* Re: [PATCH 06/10] repository: repo_submodule_init to take a submodule struct
@ 2018-10-26 22:01 4% ` Stefan Beller
0 siblings, 0 replies; 200+ results
From: Stefan Beller @ 2018-10-26 22:01 UTC (permalink / raw)
To: Jonathan Tan; +Cc: git
On Fri, Oct 26, 2018 at 12:15 PM Jonathan Tan <jonathantanmy@google.com> wrote:
[snip]
> The expected pattern.
>
> This patch looks good to me.
I'll take this as a "Reviewed-by"?
Thanks,
Stefan
^ permalink raw reply [relevance 4%]
* Re: [PATCH v7 00/10] Make submodules work if .gitmodules is not checked out
2018-10-26 1:59 4% ` Junio C Hamano
@ 2018-10-26 18:43 4% ` Stefan Beller
0 siblings, 0 replies; 200+ results
From: Stefan Beller @ 2018-10-26 18:43 UTC (permalink / raw)
To: Junio C Hamano
Cc: Antonio Ospite, git, Jonathan Nieder, Jeff King,
SZEDER Gábor
On Thu, Oct 25, 2018 at 6:59 PM Junio C Hamano <gitster@pobox.com> wrote:
>
> Stefan Beller <sbeller@google.com> writes:
>
> >> In this series I am addressing the comments by Stefan Beller about the
> >> tests in patch 9.
> >>
> >> If the new tests look OK, I'd say we try moving the series to "next" and
> >> see what happens?
> >
> > Sounds good to me.
>
> Which means (1) the plan sounds OK but I didn't look at these new
> tests or (2) the new tests look OK and I am happy to see this go to
> 'next'?
I looked at the tests and found it a pleasant read, so I think the plan of
merging it to next and seeing what will happen is a good one.
Stefan
^ permalink raw reply [relevance 4%]
* Re: [PATCH v7 00/10] Make submodules work if .gitmodules is not checked out
2018-10-25 18:49 4% ` Stefan Beller
@ 2018-10-26 1:59 4% ` Junio C Hamano
2018-10-26 18:43 4% ` Stefan Beller
0 siblings, 1 reply; 200+ results
From: Junio C Hamano @ 2018-10-26 1:59 UTC (permalink / raw)
To: Stefan Beller
Cc: Antonio Ospite, git, Jonathan Nieder, Jeff King,
SZEDER Gábor
Stefan Beller <sbeller@google.com> writes:
>> In this series I am addressing the comments by Stefan Beller about the
>> tests in patch 9.
>>
>> If the new tests look OK, I'd say we try moving the series to "next" and
>> see what happens?
>
> Sounds good to me.
Which means (1) the plan sounds OK but I didn't look at these new
tests or (2) the new tests look OK and I am happy to see this go to
'next'?
tbdiff tells me that 9/10 is the only patch different from the
previous round, and I vaguely recall that the other patches looked
OK to me (even though I admit I only skimmed them quickly).
Thanks.
^ permalink raw reply [relevance 4%]
* [PATCH 10/10] builtin/fetch: check for submodule updates in any ref update
2018-10-25 23:32 8% [PATCH 00/10] Resending sb/submodule-recursive-fetch-gets-the-tip Stefan Beller
` (7 preceding siblings ...)
2018-10-25 23:32 24% ` [PATCH 09/10] fetch: try fetching submodules if needed objects were not fetched Stefan Beller
@ 2018-10-25 23:32 17% ` Stefan Beller
8 siblings, 0 replies; 200+ results
From: Stefan Beller @ 2018-10-25 23:32 UTC (permalink / raw)
To: jonathantanmy; +Cc: git, Stefan Beller
Gerrit, the code review tool, has a different workflow than our mailing
list based approach. Usually users upload changes to a Gerrit server and
continuous integration and testing happens by bots. Sometimes however a
user wants to checkout a change locally and look at it locally. For this
use case, Gerrit offers a command line snippet to copy and paste to your
terminal, which looks like
git fetch https://<host>/gerrit refs/changes/<id> &&
git checkout FETCH_HEAD
For Gerrit changes that contain changing submodule gitlinks, it would be
easy to extend both the fetch and checkout with the '--recurse-submodules'
flag, such that this command line snippet would produce the state of a
change locally.
However the functionality added in the previous patch, which would
ensure that we fetch the objects in the submodule that the gitlink pointed
at, only works for remote tracking branches so far, not for FETCH_HEAD.
Make sure that fetching a superproject to its FETCH_HEAD, also respects
the existence checks for objects in the submodule recursion.
Signed-off-by: Stefan Beller <sbeller@google.com>
---
builtin/fetch.c | 8 ++------
t/t5526-fetch-submodules.sh | 24 ++++++++++++++++++++++++
2 files changed, 26 insertions(+), 6 deletions(-)
diff --git a/builtin/fetch.c b/builtin/fetch.c
index 95c44bf6ff..f39012d7c2 100644
--- a/builtin/fetch.c
+++ b/builtin/fetch.c
@@ -700,8 +700,6 @@ static int update_local_ref(struct ref *ref,
what = _("[new ref]");
}
- if (recurse_submodules != RECURSE_SUBMODULES_OFF)
- check_for_new_submodule_commits(&ref->new_oid);
r = s_update_ref(msg, ref, 0);
format_display(display, r ? '!' : '*', what,
r ? _("unable to update local ref") : NULL,
@@ -715,8 +713,6 @@ static int update_local_ref(struct ref *ref,
strbuf_add_unique_abbrev(&quickref, ¤t->object.oid, DEFAULT_ABBREV);
strbuf_addstr(&quickref, "..");
strbuf_add_unique_abbrev(&quickref, &ref->new_oid, DEFAULT_ABBREV);
- if (recurse_submodules != RECURSE_SUBMODULES_OFF)
- check_for_new_submodule_commits(&ref->new_oid);
r = s_update_ref("fast-forward", ref, 1);
format_display(display, r ? '!' : ' ', quickref.buf,
r ? _("unable to update local ref") : NULL,
@@ -729,8 +725,6 @@ static int update_local_ref(struct ref *ref,
strbuf_add_unique_abbrev(&quickref, ¤t->object.oid, DEFAULT_ABBREV);
strbuf_addstr(&quickref, "...");
strbuf_add_unique_abbrev(&quickref, &ref->new_oid, DEFAULT_ABBREV);
- if (recurse_submodules != RECURSE_SUBMODULES_OFF)
- check_for_new_submodule_commits(&ref->new_oid);
r = s_update_ref("forced-update", ref, 1);
format_display(display, r ? '!' : '+', quickref.buf,
r ? _("unable to update local ref") : _("forced update"),
@@ -826,6 +820,8 @@ static int store_updated_refs(const char *raw_url, const char *remote_name,
ref->force = rm->peer_ref->force;
}
+ if (recurse_submodules != RECURSE_SUBMODULES_OFF)
+ check_for_new_submodule_commits(&rm->old_oid);
if (!strcmp(rm->name, "HEAD")) {
kind = "";
diff --git a/t/t5526-fetch-submodules.sh b/t/t5526-fetch-submodules.sh
index 5a75b57852..799785783f 100755
--- a/t/t5526-fetch-submodules.sh
+++ b/t/t5526-fetch-submodules.sh
@@ -631,4 +631,28 @@ test_expect_success "fetch new submodule commits on-demand outside standard refs
)
'
+test_expect_success 'fetch new submodule commits on-demand in FETCH_HEAD' '
+ # depends on the previous test for setup
+
+ C=$(git -C submodule commit-tree -m "another change outside refs/heads" HEAD^{tree}) &&
+ git -C submodule update-ref refs/changes/1 $C &&
+ git update-index --cacheinfo 160000 $C submodule &&
+ test_tick &&
+
+ D=$(git -C sub1 commit-tree -m "another change outside refs/heads" HEAD^{tree}) &&
+ git -C sub1 update-ref refs/changes/2 $D &&
+ git update-index --cacheinfo 160000 $D sub1 &&
+
+ git commit -m "updated submodules outside of refs/heads" &&
+ E=$(git rev-parse HEAD) &&
+ git update-ref refs/changes/2 $E &&
+ (
+ cd downstream &&
+ git fetch --recurse-submodules origin refs/changes/2 &&
+ git -C submodule cat-file -t $C &&
+ git -C sub1 cat-file -t $D &&
+ git checkout --recurse-submodules FETCH_HEAD
+ )
+'
+
test_done
--
2.19.0
^ permalink raw reply related [relevance 17%]
* [PATCH 09/10] fetch: try fetching submodules if needed objects were not fetched
2018-10-25 23:32 8% [PATCH 00/10] Resending sb/submodule-recursive-fetch-gets-the-tip Stefan Beller
` (6 preceding siblings ...)
2018-10-25 23:32 17% ` [PATCH 08/10] submodule.c: fetch in submodules git directory instead of in worktree Stefan Beller
@ 2018-10-25 23:32 24% ` Stefan Beller
2018-10-25 23:32 17% ` [PATCH 10/10] builtin/fetch: check for submodule updates in any ref update Stefan Beller
8 siblings, 0 replies; 200+ results
From: Stefan Beller @ 2018-10-25 23:32 UTC (permalink / raw)
To: jonathantanmy; +Cc: git, Stefan Beller
Currently when git-fetch is asked to recurse into submodules, it dispatches
a plain "git-fetch -C <submodule-dir>" (with some submodule related options
such as prefix and recusing strategy, but) without any information of the
remote or the tip that should be fetched.
But this default fetch is not sufficient, as a newly fetched commit in
the superproject could point to a commit in the submodule that is not
in the default refspec. This is common in workflows like Gerrit's.
When fetching a Gerrit change under review (from refs/changes/??), the
commits in that change likely point to submodule commits that have not
been merged to a branch yet.
Try fetching a submodule by object id if the object id that the
superproject points to, cannot be found.
The try does not happen when the "git fetch" done at the
superproject is not storing the fetched results in remote
tracking branches (i.e. instead just recording them to
FETCH_HEAD) in this step. A later patch will fix this.
builtin/fetch used to only inspect submodules when they were fetched
"on-demand", as in either on/off case it was clear whether the submodule
needs to be fetched. However to know whether we need to try fetching the
object ids, we need to identify the object names, which is done in this
function check_for_new_submodule_commits(), so we'll also run that code
in case the submodule recursion is set to "on".
Signed-off-by: Stefan Beller <sbeller@google.com>
---
builtin/fetch.c | 9 +-
submodule.c | 192 ++++++++++++++++++++++++++++++------
t/t5526-fetch-submodules.sh | 31 ++++++
3 files changed, 198 insertions(+), 34 deletions(-)
diff --git a/builtin/fetch.c b/builtin/fetch.c
index 61bec5d213..95c44bf6ff 100644
--- a/builtin/fetch.c
+++ b/builtin/fetch.c
@@ -700,8 +700,7 @@ static int update_local_ref(struct ref *ref,
what = _("[new ref]");
}
- if ((recurse_submodules != RECURSE_SUBMODULES_OFF) &&
- (recurse_submodules != RECURSE_SUBMODULES_ON))
+ if (recurse_submodules != RECURSE_SUBMODULES_OFF)
check_for_new_submodule_commits(&ref->new_oid);
r = s_update_ref(msg, ref, 0);
format_display(display, r ? '!' : '*', what,
@@ -716,8 +715,7 @@ static int update_local_ref(struct ref *ref,
strbuf_add_unique_abbrev(&quickref, ¤t->object.oid, DEFAULT_ABBREV);
strbuf_addstr(&quickref, "..");
strbuf_add_unique_abbrev(&quickref, &ref->new_oid, DEFAULT_ABBREV);
- if ((recurse_submodules != RECURSE_SUBMODULES_OFF) &&
- (recurse_submodules != RECURSE_SUBMODULES_ON))
+ if (recurse_submodules != RECURSE_SUBMODULES_OFF)
check_for_new_submodule_commits(&ref->new_oid);
r = s_update_ref("fast-forward", ref, 1);
format_display(display, r ? '!' : ' ', quickref.buf,
@@ -731,8 +729,7 @@ static int update_local_ref(struct ref *ref,
strbuf_add_unique_abbrev(&quickref, ¤t->object.oid, DEFAULT_ABBREV);
strbuf_addstr(&quickref, "...");
strbuf_add_unique_abbrev(&quickref, &ref->new_oid, DEFAULT_ABBREV);
- if ((recurse_submodules != RECURSE_SUBMODULES_OFF) &&
- (recurse_submodules != RECURSE_SUBMODULES_ON))
+ if (recurse_submodules != RECURSE_SUBMODULES_OFF)
check_for_new_submodule_commits(&ref->new_oid);
r = s_update_ref("forced-update", ref, 1);
format_display(display, r ? '!' : '+', quickref.buf,
diff --git a/submodule.c b/submodule.c
index 67813fbe78..c978a38c81 100644
--- a/submodule.c
+++ b/submodule.c
@@ -1218,8 +1218,12 @@ struct submodule_parallel_fetch {
int result;
struct string_list changed_submodule_names;
+ struct get_next_submodule_task **fetch_specific_oids;
+ int fetch_specific_oids_nr, fetch_specific_oids_alloc;
};
-#define SPF_INIT {0, ARGV_ARRAY_INIT, NULL, NULL, 0, 0, 0, 0, STRING_LIST_INIT_DUP }
+#define SPF_INIT {0, ARGV_ARRAY_INIT, NULL, NULL, 0, 0, 0, 0, \
+ STRING_LIST_INIT_DUP, \
+ NULL, 0, 0}
static int get_fetch_recurse_config(const struct submodule *submodule,
struct submodule_parallel_fetch *spf)
@@ -1246,6 +1250,58 @@ static int get_fetch_recurse_config(const struct submodule *submodule,
return spf->default_option;
}
+struct get_next_submodule_task {
+ struct repository *repo;
+ const struct submodule *sub;
+ unsigned free_sub : 1; /* Do we need to free the submodule? */
+
+ /* fetch specific oids if set, otherwise fetch default refspec */
+ struct oid_array *commits;
+};
+
+static const struct submodule *get_default_submodule(const char *path)
+{
+ struct submodule *ret = NULL;
+ const char *name = default_name_or_path(path);
+
+ if (!name)
+ return NULL;
+
+ ret = xmalloc(sizeof(*ret));
+ memset(ret, 0, sizeof(*ret));
+ ret->path = name;
+ ret->name = name;
+
+ return (const struct submodule *) ret;
+}
+
+static struct get_next_submodule_task *get_next_submodule_task_create(
+ struct repository *r, const char *path)
+{
+ struct get_next_submodule_task *task = xmalloc(sizeof(*task));
+ memset(task, 0, sizeof(*task));
+
+ task->sub = submodule_from_path(r, &null_oid, path);
+ if (!task->sub) {
+ task->sub = get_default_submodule(path);
+ task->free_sub = 1;
+ }
+
+ return task;
+}
+
+static void get_next_submodule_task_release(struct get_next_submodule_task *p)
+{
+ if (p->free_sub)
+ free((void*)p->sub);
+ p->free_sub = 0;
+ p->sub = NULL;
+
+ if (p->repo)
+ repo_clear(p->repo);
+ FREE_AND_NULL(p->repo);
+}
+
static struct repository *get_submodule_repo_for(struct repository *r,
const struct submodule *sub)
{
@@ -1272,39 +1328,35 @@ static struct repository *get_submodule_repo_for(struct repository *r,
static int get_next_submodule(struct child_process *cp,
struct strbuf *err, void *data, void **task_cb)
{
- int ret = 0;
struct submodule_parallel_fetch *spf = data;
for (; spf->count < spf->r->index->cache_nr; spf->count++) {
- struct strbuf submodule_prefix = STRBUF_INIT;
+ int recurse_config;
const struct cache_entry *ce = spf->r->index->cache[spf->count];
const char *default_argv;
- const struct submodule *submodule;
- struct repository *repo;
- struct submodule default_submodule = SUBMODULE_INIT;
+ struct get_next_submodule_task *task;
if (!S_ISGITLINK(ce->ce_mode))
continue;
- submodule = submodule_from_path(spf->r, &null_oid, ce->name);
- if (!submodule) {
- const char *name = default_name_or_path(ce->name);
- if (name) {
- default_submodule.path = name;
- default_submodule.name = name;
- submodule = &default_submodule;
- }
+ task = get_next_submodule_task_create(spf->r, ce->name);
+
+ if (!task->sub) {
+ free(task);
+ continue;
}
- switch (get_fetch_recurse_config(submodule, spf))
+ recurse_config = get_fetch_recurse_config(task->sub, spf);
+
+ switch (recurse_config)
{
default:
case RECURSE_SUBMODULES_DEFAULT:
case RECURSE_SUBMODULES_ON_DEMAND:
- if (!submodule ||
+ if (!task->sub ||
!string_list_lookup(
&spf->changed_submodule_names,
- submodule->name))
+ task->sub->name))
continue;
default_argv = "on-demand";
break;
@@ -1315,12 +1367,12 @@ static int get_next_submodule(struct child_process *cp,
continue;
}
- strbuf_addf(&submodule_prefix, "%s%s/", spf->prefix, ce->name);
- repo = get_submodule_repo_for(spf->r, submodule);
- if (repo) {
+ task->repo = get_submodule_repo_for(spf->r, task->sub);
+ if (task->repo) {
+ struct strbuf submodule_prefix = STRBUF_INIT;
child_process_init(cp);
prepare_submodule_repo_env_in_gitdir(&cp->env_array);
- cp->dir = xstrdup(repo->gitdir);
+ cp->dir = task->repo->gitdir;
cp->git_cmd = 1;
if (!spf->quiet)
strbuf_addf(err, "Fetching submodule %s%s\n",
@@ -1329,12 +1381,22 @@ static int get_next_submodule(struct child_process *cp,
argv_array_pushv(&cp->args, spf->args.argv);
argv_array_push(&cp->args, default_argv);
argv_array_push(&cp->args, "--submodule-prefix");
+
+ strbuf_addf(&submodule_prefix, "%s%s/",
+ spf->prefix,
+ task->sub->path);
argv_array_push(&cp->args, submodule_prefix.buf);
- repo_clear(repo);
- free(repo);
- ret = 1;
+ spf->count++;
+ *task_cb = task;
+
+ strbuf_release(&submodule_prefix);
+ return 1;
} else {
+
+ get_next_submodule_task_release(task);
+ free(task);
+
/*
* An empty directory is normal,
* the submodule is not initialized
@@ -1343,12 +1405,36 @@ static int get_next_submodule(struct child_process *cp,
!is_empty_dir(ce->name))
die(_("Could not access submodule '%s'"), ce->name);
}
+ }
+
+ if (spf->fetch_specific_oids_nr) {
+ struct get_next_submodule_task *task = spf->fetch_specific_oids[spf->fetch_specific_oids_nr - 1];
+ struct strbuf submodule_prefix = STRBUF_INIT;
+ spf->fetch_specific_oids_nr--;
+
+ strbuf_addf(&submodule_prefix, "%s%s/", spf->prefix, task->sub->path);
+
+ child_process_init(cp);
+ prepare_submodule_repo_env_in_gitdir(&cp->env_array);
+ cp->git_cmd = 1;
+ cp->dir = task->repo->gitdir;
+
+ argv_array_init(&cp->args);
+ argv_array_pushv(&cp->args, spf->args.argv);
+ argv_array_push(&cp->args, "on-demand");
+ argv_array_push(&cp->args, "--submodule-prefix");
+ argv_array_push(&cp->args, submodule_prefix.buf);
+
+ /* NEEDSWORK: have get_default_remote from s--h */
+ argv_array_push(&cp->args, "origin");
+ oid_array_for_each_unique(task->commits,
+ append_oid_to_argv, &cp->args);
+
+ *task_cb = task;
strbuf_release(&submodule_prefix);
- if (ret) {
- spf->count++;
- return 1;
- }
+ return 1;
}
+
return 0;
}
@@ -1356,20 +1442,70 @@ static int fetch_start_failure(struct strbuf *err,
void *cb, void *task_cb)
{
struct submodule_parallel_fetch *spf = cb;
+ struct get_next_submodule_task *task = task_cb;
spf->result = 1;
+ get_next_submodule_task_release(task);
return 0;
}
+static int commit_exists_in_sub(const struct object_id *oid, void *data)
+{
+ struct repository *subrepo = data;
+
+ enum object_type type = oid_object_info(subrepo, oid, NULL);
+
+ return type != OBJ_COMMIT;
+}
+
static int fetch_finish(int retvalue, struct strbuf *err,
void *cb, void *task_cb)
{
struct submodule_parallel_fetch *spf = cb;
+ struct get_next_submodule_task *task = task_cb;
+ const struct submodule *sub;
+
+ struct string_list_item *it;
+ struct oid_array *commits;
if (retvalue)
spf->result = 1;
+ if (!task)
+ return 0;
+
+ sub = task->sub;
+ if (!sub)
+ goto out;
+
+ it = string_list_lookup(&spf->changed_submodule_names, sub->name);
+ if (!it)
+ goto out;
+
+ commits = it->util;
+ oid_array_filter(commits,
+ commit_exists_in_sub,
+ task->repo);
+
+ /* Are there commits that do not exist? */
+ if (commits->nr) {
+ /* We already tried fetching them, do not try again. */
+ if (task->commits)
+ return 0;
+
+ task->commits = commits;
+ ALLOC_GROW(spf->fetch_specific_oids,
+ spf->fetch_specific_oids_nr + 1,
+ spf->fetch_specific_oids_alloc);
+ spf->fetch_specific_oids[spf->fetch_specific_oids_nr] = task;
+ spf->fetch_specific_oids_nr++;
+ return 0;
+ }
+
+out:
+ get_next_submodule_task_release(task);
+
return 0;
}
diff --git a/t/t5526-fetch-submodules.sh b/t/t5526-fetch-submodules.sh
index 6c2f9b2ba2..5a75b57852 100755
--- a/t/t5526-fetch-submodules.sh
+++ b/t/t5526-fetch-submodules.sh
@@ -600,4 +600,35 @@ test_expect_success "fetch new commits when submodule got renamed" '
test_cmp expect actual
'
+test_expect_success "fetch new submodule commits on-demand outside standard refspec" '
+ # add a second submodule and ensure it is around in downstream first
+ git clone submodule sub1 &&
+ git submodule add ./sub1 &&
+ git commit -m "adding a second submodule" &&
+ git -C downstream pull &&
+ git -C downstream submodule update --init --recursive &&
+
+ git checkout --detach &&
+
+ C=$(git -C submodule commit-tree -m "new change outside refs/heads" HEAD^{tree}) &&
+ git -C submodule update-ref refs/changes/1 $C &&
+ git update-index --cacheinfo 160000 $C submodule &&
+ test_tick &&
+
+ D=$(git -C sub1 commit-tree -m "new change outside refs/heads" HEAD^{tree}) &&
+ git -C sub1 update-ref refs/changes/2 $D &&
+ git update-index --cacheinfo 160000 $D sub1 &&
+
+ git commit -m "updated submodules outside of refs/heads" &&
+ E=$(git rev-parse HEAD) &&
+ git update-ref refs/changes/2 $E &&
+ (
+ cd downstream &&
+ git fetch --recurse-submodules origin refs/changes/2:refs/heads/my_branch &&
+ git -C submodule cat-file -t $C &&
+ git -C sub1 cat-file -t $D &&
+ git checkout --recurse-submodules FETCH_HEAD
+ )
+'
+
test_done
--
2.19.0
^ permalink raw reply related [relevance 24%]
* [PATCH 08/10] submodule.c: fetch in submodules git directory instead of in worktree
2018-10-25 23:32 8% [PATCH 00/10] Resending sb/submodule-recursive-fetch-gets-the-tip Stefan Beller
` (5 preceding siblings ...)
2018-10-25 23:32 23% ` [PATCH 07/10] submodule: migrate get_next_submodule to use repository structs Stefan Beller
@ 2018-10-25 23:32 17% ` Stefan Beller
2018-10-25 23:32 24% ` [PATCH 09/10] fetch: try fetching submodules if needed objects were not fetched Stefan Beller
2018-10-25 23:32 17% ` [PATCH 10/10] builtin/fetch: check for submodule updates in any ref update Stefan Beller
8 siblings, 0 replies; 200+ results
From: Stefan Beller @ 2018-10-25 23:32 UTC (permalink / raw)
To: jonathantanmy; +Cc: git, Stefan Beller
Keep the properties introduced in 10f5c52656 (submodule: avoid
auto-discovery in prepare_submodule_repo_env(), 2016-09-01), by fixating
the git directory of the submodule.
Signed-off-by: Stefan Beller <sbeller@google.com>
---
submodule.c | 10 ++++++++--
1 file changed, 8 insertions(+), 2 deletions(-)
diff --git a/submodule.c b/submodule.c
index a1a32cab7d..67813fbe78 100644
--- a/submodule.c
+++ b/submodule.c
@@ -494,6 +494,12 @@ void prepare_submodule_repo_env(struct argv_array *out)
DEFAULT_GIT_DIR_ENVIRONMENT);
}
+static void prepare_submodule_repo_env_in_gitdir(struct argv_array *out)
+{
+ prepare_submodule_repo_env_no_git_dir(out);
+ argv_array_pushf(out, "%s=.", GIT_DIR_ENVIRONMENT);
+}
+
/* Helper function to display the submodule header line prior to the full
* summary output. If it can locate the submodule objects directory it will
* attempt to lookup both the left and right commits and put them into the
@@ -1313,8 +1319,8 @@ static int get_next_submodule(struct child_process *cp,
repo = get_submodule_repo_for(spf->r, submodule);
if (repo) {
child_process_init(cp);
- prepare_submodule_repo_env(&cp->env_array);
- cp->dir = xstrdup(repo->worktree);
+ prepare_submodule_repo_env_in_gitdir(&cp->env_array);
+ cp->dir = xstrdup(repo->gitdir);
cp->git_cmd = 1;
if (!spf->quiet)
strbuf_addf(err, "Fetching submodule %s%s\n",
--
2.19.0
^ permalink raw reply related [relevance 17%]
* [PATCH 07/10] submodule: migrate get_next_submodule to use repository structs
2018-10-25 23:32 8% [PATCH 00/10] Resending sb/submodule-recursive-fetch-gets-the-tip Stefan Beller
` (4 preceding siblings ...)
2018-10-25 23:32 25% ` [PATCH 06/10] repository: repo_submodule_init to take a submodule struct Stefan Beller
@ 2018-10-25 23:32 23% ` Stefan Beller
2018-10-25 23:32 17% ` [PATCH 08/10] submodule.c: fetch in submodules git directory instead of in worktree Stefan Beller
` (2 subsequent siblings)
8 siblings, 0 replies; 200+ results
From: Stefan Beller @ 2018-10-25 23:32 UTC (permalink / raw)
To: jonathantanmy; +Cc: git, Stefan Beller
We used to recurse into submodules, even if they were broken having
only an objects directory. The child process executed in the submodule
would fail though if the submodule was broken.
This patch tightens the check upfront, such that we do not need
to spawn a child process to find out if the submodule is broken.
Signed-off-by: Stefan Beller <sbeller@google.com>
---
submodule.c | 51 +++++++++++++++++++++++++++++++++++++++------------
1 file changed, 39 insertions(+), 12 deletions(-)
diff --git a/submodule.c b/submodule.c
index e4b494af7b..a1a32cab7d 100644
--- a/submodule.c
+++ b/submodule.c
@@ -1240,6 +1240,29 @@ static int get_fetch_recurse_config(const struct submodule *submodule,
return spf->default_option;
}
+static struct repository *get_submodule_repo_for(struct repository *r,
+ const struct submodule *sub)
+{
+ struct repository *ret = xmalloc(sizeof(*ret));
+
+ if (repo_submodule_init(ret, r, sub)) {
+ /*
+ * No entry in .gitmodules? Technically not a submodule,
+ * but historically we supported repositories that happen to be
+ * in-place where a gitlink is. Keep supporting them.
+ */
+ struct strbuf gitdir = STRBUF_INIT;
+ strbuf_repo_worktree_path(&gitdir, r, "%s/.git", sub->path);
+ if (repo_init(ret, gitdir.buf, NULL)) {
+ strbuf_release(&gitdir);
+ return NULL;
+ }
+ strbuf_release(&gitdir);
+ }
+
+ return ret;
+}
+
static int get_next_submodule(struct child_process *cp,
struct strbuf *err, void *data, void **task_cb)
{
@@ -1247,12 +1270,11 @@ static int get_next_submodule(struct child_process *cp,
struct submodule_parallel_fetch *spf = data;
for (; spf->count < spf->r->index->cache_nr; spf->count++) {
- struct strbuf submodule_path = STRBUF_INIT;
- struct strbuf submodule_git_dir = STRBUF_INIT;
struct strbuf submodule_prefix = STRBUF_INIT;
const struct cache_entry *ce = spf->r->index->cache[spf->count];
- const char *git_dir, *default_argv;
+ const char *default_argv;
const struct submodule *submodule;
+ struct repository *repo;
struct submodule default_submodule = SUBMODULE_INIT;
if (!S_ISGITLINK(ce->ce_mode))
@@ -1287,16 +1309,12 @@ static int get_next_submodule(struct child_process *cp,
continue;
}
- strbuf_repo_worktree_path(&submodule_path, spf->r, "%s", ce->name);
- strbuf_addf(&submodule_git_dir, "%s/.git", submodule_path.buf);
strbuf_addf(&submodule_prefix, "%s%s/", spf->prefix, ce->name);
- git_dir = read_gitfile(submodule_git_dir.buf);
- if (!git_dir)
- git_dir = submodule_git_dir.buf;
- if (is_directory(git_dir)) {
+ repo = get_submodule_repo_for(spf->r, submodule);
+ if (repo) {
child_process_init(cp);
- cp->dir = strbuf_detach(&submodule_path, NULL);
prepare_submodule_repo_env(&cp->env_array);
+ cp->dir = xstrdup(repo->worktree);
cp->git_cmd = 1;
if (!spf->quiet)
strbuf_addf(err, "Fetching submodule %s%s\n",
@@ -1306,10 +1324,19 @@ static int get_next_submodule(struct child_process *cp,
argv_array_push(&cp->args, default_argv);
argv_array_push(&cp->args, "--submodule-prefix");
argv_array_push(&cp->args, submodule_prefix.buf);
+
+ repo_clear(repo);
+ free(repo);
ret = 1;
+ } else {
+ /*
+ * An empty directory is normal,
+ * the submodule is not initialized
+ */
+ if (S_ISGITLINK(ce->ce_mode) &&
+ !is_empty_dir(ce->name))
+ die(_("Could not access submodule '%s'"), ce->name);
}
- strbuf_release(&submodule_path);
- strbuf_release(&submodule_git_dir);
strbuf_release(&submodule_prefix);
if (ret) {
spf->count++;
--
2.19.0
^ permalink raw reply related [relevance 23%]
* [PATCH 06/10] repository: repo_submodule_init to take a submodule struct
2018-10-25 23:32 8% [PATCH 00/10] Resending sb/submodule-recursive-fetch-gets-the-tip Stefan Beller
` (3 preceding siblings ...)
2018-10-25 23:32 16% ` [PATCH 05/10] submodule: store OIDs in changed_submodule_names Stefan Beller
@ 2018-10-25 23:32 25% ` Stefan Beller
2018-10-25 23:32 23% ` [PATCH 07/10] submodule: migrate get_next_submodule to use repository structs Stefan Beller
` (3 subsequent siblings)
8 siblings, 1 reply; 200+ results
From: Stefan Beller @ 2018-10-25 23:32 UTC (permalink / raw)
To: jonathantanmy; +Cc: git, Stefan Beller
When constructing a struct repository for a submodule for some revision
of the superproject where the submodule is not contained in the index,
it may not be present in the working tree currently either. In that
situation giving a 'path' argument is not useful. Upgrade the
repo_submodule_init function to take a struct submodule instead.
The submodule struct can be obtained via submodule_from_{path, name} or
an artificial submodule struct can be passed in.
While we are at it, rename the repository struct in the repo_submodule_init
function, which is to be initialized, to a name that is not confused with
the struct submodule as easily. Perform such renames in similar functions
as well.
Also move its documentation into the header file.
Signed-off-by: Stefan Beller <sbeller@google.com>
---
builtin/grep.c | 17 +++++++-----
builtin/ls-files.c | 12 +++++----
builtin/submodule--helper.c | 2 +-
repository.c | 27 ++++++++------------
repository.h | 12 +++++++--
t/helper/test-submodule-nested-repo-config.c | 8 +++---
6 files changed, 43 insertions(+), 35 deletions(-)
diff --git a/builtin/grep.c b/builtin/grep.c
index 7da8fef31a..ba7634258a 100644
--- a/builtin/grep.c
+++ b/builtin/grep.c
@@ -418,7 +418,10 @@ static int grep_submodule(struct grep_opt *opt, struct repository *superproject,
const struct object_id *oid,
const char *filename, const char *path)
{
- struct repository submodule;
+ struct repository subrepo;
+ const struct submodule *sub = submodule_from_path(superproject,
+ &null_oid, path);
+
int hit;
/*
@@ -434,12 +437,12 @@ static int grep_submodule(struct grep_opt *opt, struct repository *superproject,
return 0;
}
- if (repo_submodule_init(&submodule, superproject, path)) {
+ if (repo_submodule_init(&subrepo, superproject, sub)) {
grep_read_unlock();
return 0;
}
- repo_read_gitmodules(&submodule);
+ repo_read_gitmodules(&subrepo);
/*
* NEEDSWORK: This adds the submodule's object directory to the list of
@@ -451,7 +454,7 @@ static int grep_submodule(struct grep_opt *opt, struct repository *superproject,
* store is no longer global and instead is a member of the repository
* object.
*/
- add_to_alternates_memory(submodule.objects->objectdir);
+ add_to_alternates_memory(subrepo.objects->objectdir);
grep_read_unlock();
if (oid) {
@@ -476,14 +479,14 @@ static int grep_submodule(struct grep_opt *opt, struct repository *superproject,
init_tree_desc(&tree, data, size);
hit = grep_tree(opt, pathspec, &tree, &base, base.len,
- object->type == OBJ_COMMIT, &submodule);
+ object->type == OBJ_COMMIT, &subrepo);
strbuf_release(&base);
free(data);
} else {
- hit = grep_cache(opt, &submodule, pathspec, 1);
+ hit = grep_cache(opt, &subrepo, pathspec, 1);
}
- repo_clear(&submodule);
+ repo_clear(&subrepo);
return hit;
}
diff --git a/builtin/ls-files.c b/builtin/ls-files.c
index 7f9919a362..4d1649c1b3 100644
--- a/builtin/ls-files.c
+++ b/builtin/ls-files.c
@@ -206,17 +206,19 @@ static void show_files(struct repository *repo, struct dir_struct *dir);
static void show_submodule(struct repository *superproject,
struct dir_struct *dir, const char *path)
{
- struct repository submodule;
+ struct repository subrepo;
+ const struct submodule *sub = submodule_from_path(superproject,
+ &null_oid, path);
- if (repo_submodule_init(&submodule, superproject, path))
+ if (repo_submodule_init(&subrepo, superproject, sub))
return;
- if (repo_read_index(&submodule) < 0)
+ if (repo_read_index(&subrepo) < 0)
die("index file corrupt");
- show_files(&submodule, dir);
+ show_files(&subrepo, dir);
- repo_clear(&submodule);
+ repo_clear(&subrepo);
}
static void show_ce(struct repository *repo, struct dir_struct *dir,
diff --git a/builtin/submodule--helper.c b/builtin/submodule--helper.c
index 5f8a804a6e..015aa1471f 100644
--- a/builtin/submodule--helper.c
+++ b/builtin/submodule--helper.c
@@ -2037,7 +2037,7 @@ static int ensure_core_worktree(int argc, const char **argv, const char *prefix)
if (!sub)
BUG("We could get the submodule handle before?");
- if (repo_submodule_init(&subrepo, the_repository, path))
+ if (repo_submodule_init(&subrepo, the_repository, sub))
die(_("could not get a repository handle for submodule '%s'"), path);
if (!repo_config_get_string(&subrepo, "core.worktree", &cw)) {
diff --git a/repository.c b/repository.c
index 5dd1486718..aabe64ee5d 100644
--- a/repository.c
+++ b/repository.c
@@ -166,30 +166,23 @@ int repo_init(struct repository *repo,
return -1;
}
-/*
- * Initialize 'submodule' as the submodule given by 'path' in parent repository
- * 'superproject'.
- * Return 0 upon success and a non-zero value upon failure.
- */
-int repo_submodule_init(struct repository *submodule,
+int repo_submodule_init(struct repository *subrepo,
struct repository *superproject,
- const char *path)
+ const struct submodule *sub)
{
- const struct submodule *sub;
struct strbuf gitdir = STRBUF_INIT;
struct strbuf worktree = STRBUF_INIT;
int ret = 0;
- sub = submodule_from_path(superproject, &null_oid, path);
if (!sub) {
ret = -1;
goto out;
}
- strbuf_repo_worktree_path(&gitdir, superproject, "%s/.git", path);
- strbuf_repo_worktree_path(&worktree, superproject, "%s", path);
+ strbuf_repo_worktree_path(&gitdir, superproject, "%s/.git", sub->path);
+ strbuf_repo_worktree_path(&worktree, superproject, "%s", sub->path);
- if (repo_init(submodule, gitdir.buf, worktree.buf)) {
+ if (repo_init(subrepo, gitdir.buf, worktree.buf)) {
/*
* If initilization fails then it may be due to the submodule
* not being populated in the superproject's worktree. Instead
@@ -201,16 +194,16 @@ int repo_submodule_init(struct repository *submodule,
strbuf_repo_git_path(&gitdir, superproject,
"modules/%s", sub->name);
- if (repo_init(submodule, gitdir.buf, NULL)) {
+ if (repo_init(subrepo, gitdir.buf, NULL)) {
ret = -1;
goto out;
}
}
- submodule->submodule_prefix = xstrfmt("%s%s/",
- superproject->submodule_prefix ?
- superproject->submodule_prefix :
- "", path);
+ subrepo->submodule_prefix = xstrfmt("%s%s/",
+ superproject->submodule_prefix ?
+ superproject->submodule_prefix :
+ "", sub->path);
out:
strbuf_release(&gitdir);
diff --git a/repository.h b/repository.h
index 9f16c42c1e..0e482b7d49 100644
--- a/repository.h
+++ b/repository.h
@@ -116,9 +116,17 @@ void repo_set_worktree(struct repository *repo, const char *path);
void repo_set_hash_algo(struct repository *repo, int algo);
void initialize_the_repository(void);
int repo_init(struct repository *r, const char *gitdir, const char *worktree);
-int repo_submodule_init(struct repository *submodule,
+
+/*
+ * Initialize the repository 'subrepo' as the submodule given by the
+ * struct submodule 'sub' in parent repository 'superproject'.
+ * Return 0 upon success and a non-zero value upon failure, which may happen
+ * if the submodule is not found, or 'sub' is NULL.
+ */
+struct submodule;
+int repo_submodule_init(struct repository *subrepo,
struct repository *superproject,
- const char *path);
+ const struct submodule *sub);
void repo_clear(struct repository *repo);
/*
diff --git a/t/helper/test-submodule-nested-repo-config.c b/t/helper/test-submodule-nested-repo-config.c
index a31e2a9bea..bc97929bbc 100644
--- a/t/helper/test-submodule-nested-repo-config.c
+++ b/t/helper/test-submodule-nested-repo-config.c
@@ -10,19 +10,21 @@ static void die_usage(int argc, const char **argv, const char *msg)
int cmd__submodule_nested_repo_config(int argc, const char **argv)
{
- struct repository submodule;
+ struct repository subrepo;
+ const struct submodule *sub;
if (argc < 3)
die_usage(argc, argv, "Wrong number of arguments.");
setup_git_directory();
- if (repo_submodule_init(&submodule, the_repository, argv[1])) {
+ sub = submodule_from_path(the_repository, &null_oid, argv[1]);
+ if (repo_submodule_init(&subrepo, the_repository, sub)) {
die_usage(argc, argv, "Submodule not found.");
}
/* Read the config of _child_ submodules. */
- print_config_from_gitmodules(&submodule, argv[2]);
+ print_config_from_gitmodules(&subrepo, argv[2]);
submodule_free(the_repository);
--
2.19.0
^ permalink raw reply related [relevance 25%]
* [PATCH 05/10] submodule: store OIDs in changed_submodule_names
2018-10-25 23:32 8% [PATCH 00/10] Resending sb/submodule-recursive-fetch-gets-the-tip Stefan Beller
` (2 preceding siblings ...)
2018-10-25 23:32 16% ` [PATCH 04/10] submodule.c: tighten scope of changed_submodule_names struct Stefan Beller
@ 2018-10-25 23:32 16% ` Stefan Beller
2018-10-25 23:32 25% ` [PATCH 06/10] repository: repo_submodule_init to take a submodule struct Stefan Beller
` (4 subsequent siblings)
8 siblings, 0 replies; 200+ results
From: Stefan Beller @ 2018-10-25 23:32 UTC (permalink / raw)
To: jonathantanmy; +Cc: git, Stefan Beller
'calculate_changed_submodule_paths' uses a local list to compute the
changed submodules, and then produces the result by copying appropriate
items into the result list.
Instead use the result list directly and prune items afterwards
using string_list_remove_empty_items.
By doing so we'll have access to the util pointer for longer that
contains the commits that we need to fetch, which will be
useful in a later patch.
Signed-off-by: Stefan Beller <sbeller@google.com>
Reviewed-by: Jonathan Tan <jonathantanmy@google.com>
---
submodule.c | 19 ++++++++++---------
1 file changed, 10 insertions(+), 9 deletions(-)
diff --git a/submodule.c b/submodule.c
index 6fb0b9d783..e4b494af7b 100644
--- a/submodule.c
+++ b/submodule.c
@@ -1127,8 +1127,7 @@ static void calculate_changed_submodule_paths(
struct string_list *changed_submodule_names)
{
struct argv_array argv = ARGV_ARRAY_INIT;
- struct string_list changed_submodules = STRING_LIST_INIT_DUP;
- const struct string_list_item *name;
+ struct string_list_item *name;
/* No need to check if there are no submodules configured */
if (!submodule_from_path(the_repository, NULL, NULL))
@@ -1145,9 +1144,9 @@ static void calculate_changed_submodule_paths(
* Collect all submodules (whether checked out or not) for which new
* commits have been recorded upstream in "changed_submodule_names".
*/
- collect_changed_submodules(&changed_submodules, &argv);
+ collect_changed_submodules(changed_submodule_names, &argv);
- for_each_string_list_item(name, &changed_submodules) {
+ for_each_string_list_item(name, changed_submodule_names) {
struct oid_array *commits = name->util;
const struct submodule *submodule;
const char *path = NULL;
@@ -1161,12 +1160,14 @@ static void calculate_changed_submodule_paths(
if (!path)
continue;
- if (!submodule_has_commits(path, commits))
- string_list_append(changed_submodule_names,
- name->string);
+ if (submodule_has_commits(path, commits)) {
+ oid_array_clear(commits);
+ *name->string = '\0';
+ }
}
- free_submodules_oids(&changed_submodules);
+ string_list_remove_empty_items(changed_submodule_names, 1);
+
argv_array_clear(&argv);
oid_array_clear(&ref_tips_before_fetch);
oid_array_clear(&ref_tips_after_fetch);
@@ -1376,7 +1377,7 @@ int fetch_populated_submodules(struct repository *r,
argv_array_clear(&spf.args);
out:
- string_list_clear(&spf.changed_submodule_names, 1);
+ free_submodules_oids(&spf.changed_submodule_names);
return spf.result;
}
--
2.19.0
^ permalink raw reply related [relevance 16%]
* [PATCH 04/10] submodule.c: tighten scope of changed_submodule_names struct
2018-10-25 23:32 8% [PATCH 00/10] Resending sb/submodule-recursive-fetch-gets-the-tip Stefan Beller
2018-10-25 23:32 24% ` [PATCH 02/10] submodule.c: fix indentation Stefan Beller
2018-10-25 23:32 16% ` [PATCH 03/10] submodule.c: sort changed_submodule_names before searching it Stefan Beller
@ 2018-10-25 23:32 16% ` Stefan Beller
2018-10-25 23:32 16% ` [PATCH 05/10] submodule: store OIDs in changed_submodule_names Stefan Beller
` (5 subsequent siblings)
8 siblings, 0 replies; 200+ results
From: Stefan Beller @ 2018-10-25 23:32 UTC (permalink / raw)
To: jonathantanmy; +Cc: git, Stefan Beller
The `changed_submodule_names` are only used for fetching, so let's make it
part of the struct that is passed around for fetching submodules.
Signed-off-by: Stefan Beller <sbeller@google.com>
---
submodule.c | 19 +++++++++++--------
1 file changed, 11 insertions(+), 8 deletions(-)
diff --git a/submodule.c b/submodule.c
index 9fbfcfcfe1..6fb0b9d783 100644
--- a/submodule.c
+++ b/submodule.c
@@ -24,7 +24,6 @@
#include "object-store.h"
static int config_update_recurse_submodules = RECURSE_SUBMODULES_OFF;
-static struct string_list changed_submodule_names = STRING_LIST_INIT_DUP;
static int initialized_fetch_ref_tips;
static struct oid_array ref_tips_before_fetch;
static struct oid_array ref_tips_after_fetch;
@@ -1124,7 +1123,8 @@ void check_for_new_submodule_commits(struct object_id *oid)
oid_array_append(&ref_tips_after_fetch, oid);
}
-static void calculate_changed_submodule_paths(void)
+static void calculate_changed_submodule_paths(
+ struct string_list *changed_submodule_names)
{
struct argv_array argv = ARGV_ARRAY_INIT;
struct string_list changed_submodules = STRING_LIST_INIT_DUP;
@@ -1162,7 +1162,8 @@ static void calculate_changed_submodule_paths(void)
continue;
if (!submodule_has_commits(path, commits))
- string_list_append(&changed_submodule_names, name->string);
+ string_list_append(changed_submodule_names,
+ name->string);
}
free_submodules_oids(&changed_submodules);
@@ -1208,8 +1209,10 @@ struct submodule_parallel_fetch {
int default_option;
int quiet;
int result;
+
+ struct string_list changed_submodule_names;
};
-#define SPF_INIT {0, ARGV_ARRAY_INIT, NULL, NULL, 0, 0, 0, 0}
+#define SPF_INIT {0, ARGV_ARRAY_INIT, NULL, NULL, 0, 0, 0, 0, STRING_LIST_INIT_DUP }
static int get_fetch_recurse_config(const struct submodule *submodule,
struct submodule_parallel_fetch *spf)
@@ -1271,7 +1274,7 @@ static int get_next_submodule(struct child_process *cp,
case RECURSE_SUBMODULES_ON_DEMAND:
if (!submodule ||
!string_list_lookup(
- &changed_submodule_names,
+ &spf->changed_submodule_names,
submodule->name))
continue;
default_argv = "on-demand";
@@ -1363,8 +1366,8 @@ int fetch_populated_submodules(struct repository *r,
argv_array_push(&spf.args, "--recurse-submodules-default");
/* default value, "--submodule-prefix" and its value are added later */
- calculate_changed_submodule_paths();
- string_list_sort(&changed_submodule_names);
+ calculate_changed_submodule_paths(&spf.changed_submodule_names);
+ string_list_sort(&spf.changed_submodule_names);
run_processes_parallel(max_parallel_jobs,
get_next_submodule,
fetch_start_failure,
@@ -1373,7 +1376,7 @@ int fetch_populated_submodules(struct repository *r,
argv_array_clear(&spf.args);
out:
- string_list_clear(&changed_submodule_names, 1);
+ string_list_clear(&spf.changed_submodule_names, 1);
return spf.result;
}
--
2.19.0
^ permalink raw reply related [relevance 16%]
* [PATCH 02/10] submodule.c: fix indentation
2018-10-25 23:32 8% [PATCH 00/10] Resending sb/submodule-recursive-fetch-gets-the-tip Stefan Beller
@ 2018-10-25 23:32 24% ` Stefan Beller
2018-10-25 23:32 16% ` [PATCH 03/10] submodule.c: sort changed_submodule_names before searching it Stefan Beller
` (7 subsequent siblings)
8 siblings, 0 replies; 200+ results
From: Stefan Beller @ 2018-10-25 23:32 UTC (permalink / raw)
To: jonathantanmy; +Cc: git, Stefan Beller, Junio C Hamano
The submodule subsystem is really bad at staying within 80 characters.
Fix it while we are here.
Signed-off-by: Stefan Beller <sbeller@google.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
---
submodule.c | 9 ++++++---
1 file changed, 6 insertions(+), 3 deletions(-)
diff --git a/submodule.c b/submodule.c
index 2b7082b2db..e145ebbb16 100644
--- a/submodule.c
+++ b/submodule.c
@@ -1258,7 +1258,8 @@ static int get_next_submodule(struct child_process *cp,
if (!submodule) {
const char *name = default_name_or_path(ce->name);
if (name) {
- default_submodule.path = default_submodule.name = name;
+ default_submodule.path = name;
+ default_submodule.name = name;
submodule = &default_submodule;
}
}
@@ -1268,8 +1269,10 @@ static int get_next_submodule(struct child_process *cp,
default:
case RECURSE_SUBMODULES_DEFAULT:
case RECURSE_SUBMODULES_ON_DEMAND:
- if (!submodule || !unsorted_string_list_lookup(&changed_submodule_names,
- submodule->name))
+ if (!submodule ||
+ !unsorted_string_list_lookup(
+ &changed_submodule_names,
+ submodule->name))
continue;
default_argv = "on-demand";
break;
--
2.19.0
^ permalink raw reply related [relevance 24%]
* [PATCH 03/10] submodule.c: sort changed_submodule_names before searching it
2018-10-25 23:32 8% [PATCH 00/10] Resending sb/submodule-recursive-fetch-gets-the-tip Stefan Beller
2018-10-25 23:32 24% ` [PATCH 02/10] submodule.c: fix indentation Stefan Beller
@ 2018-10-25 23:32 16% ` Stefan Beller
2018-10-25 23:32 16% ` [PATCH 04/10] submodule.c: tighten scope of changed_submodule_names struct Stefan Beller
` (6 subsequent siblings)
8 siblings, 0 replies; 200+ results
From: Stefan Beller @ 2018-10-25 23:32 UTC (permalink / raw)
To: jonathantanmy; +Cc: git, Stefan Beller, Junio C Hamano
We can string_list_insert() to maintain sorted-ness of the
list as we find new items, or we can string_list_append() to
build an unsorted list and sort it at the end just once.
As we do not rely on the sortedness while building the
list, we pick the "append and sort at the end" as it
has better worst case execution times.
Signed-off-by: Stefan Beller <sbeller@google.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
---
submodule.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/submodule.c b/submodule.c
index e145ebbb16..9fbfcfcfe1 100644
--- a/submodule.c
+++ b/submodule.c
@@ -1270,7 +1270,7 @@ static int get_next_submodule(struct child_process *cp,
case RECURSE_SUBMODULES_DEFAULT:
case RECURSE_SUBMODULES_ON_DEMAND:
if (!submodule ||
- !unsorted_string_list_lookup(
+ !string_list_lookup(
&changed_submodule_names,
submodule->name))
continue;
@@ -1364,6 +1364,7 @@ int fetch_populated_submodules(struct repository *r,
/* default value, "--submodule-prefix" and its value are added later */
calculate_changed_submodule_paths();
+ string_list_sort(&changed_submodule_names);
run_processes_parallel(max_parallel_jobs,
get_next_submodule,
fetch_start_failure,
--
2.19.0
^ permalink raw reply related [relevance 16%]
* [PATCH 00/10] Resending sb/submodule-recursive-fetch-gets-the-tip
@ 2018-10-25 23:32 8% Stefan Beller
2018-10-25 23:32 24% ` [PATCH 02/10] submodule.c: fix indentation Stefan Beller
` (8 more replies)
0 siblings, 9 replies; 200+ results
From: Stefan Beller @ 2018-10-25 23:32 UTC (permalink / raw)
To: jonathantanmy; +Cc: git, Stefan Beller
Thanks Jonathan for your thoughtful comments,
here is the product of the discussion:
* I split up the patch to fetch in the worktree to be 2 patches,
each giving motivation on its own.
* the last patch is vastly simplified in code, but takes an extra test
* in [1], you remark "commits" ought not to be a pointer, but I decided against
that, as we keep the pointed-at value around for the same time span (until
we're done with that submodule) and we don't need to copy over the pointed-at
value into the new struct.
[1] https://public-inbox.org/git/20181018003954.139498-1-jonathantanmy@google.com/
This is still based on ao/submodule-wo-gitmodules-checked-out.
previous version
https://public-inbox.org/git/20181016181327.107186-1-sbeller@google.com/
Stefan Beller (10):
sha1-array: provide oid_array_filter
submodule.c: fix indentation
submodule.c: sort changed_submodule_names before searching it
submodule.c: tighten scope of changed_submodule_names struct
submodule: store OIDs in changed_submodule_names
repository: repo_submodule_init to take a submodule struct
submodule: migrate get_next_submodule to use repository structs
submodule.c: fetch in submodules git directory instead of in worktree
fetch: try fetching submodules if needed objects were not fetched
builtin/fetch: check for submodule updates in any ref update
Documentation/technical/api-oid-array.txt | 5 +
builtin/fetch.c | 11 +-
builtin/grep.c | 17 +-
builtin/ls-files.c | 12 +-
builtin/submodule--helper.c | 2 +-
repository.c | 27 +-
repository.h | 12 +-
sha1-array.c | 17 ++
sha1-array.h | 3 +
submodule.c | 265 ++++++++++++++++---
t/helper/test-submodule-nested-repo-config.c | 8 +-
t/t5526-fetch-submodules.sh | 55 ++++
12 files changed, 346 insertions(+), 88 deletions(-)
git range-diff origin/sb/submodule-recursive-fetch-gets-the-tip...
1: 3fbb06aedd ! 1: 0fdb0e2ad9 submodule.c: move global changed_submodule_names into fetch submodule struct
@@ -1,12 +1,11 @@
Author: Stefan Beller <sbeller@google.com>
- submodule.c: move global changed_submodule_names into fetch submodule struct
+ submodule.c: tighten scope of changed_submodule_names struct
The `changed_submodule_names` are only used for fetching, so let's make it
part of the struct that is passed around for fetching submodules.
Signed-off-by: Stefan Beller <sbeller@google.com>
- Signed-off-by: Junio C Hamano <gitster@pobox.com>
diff --git a/submodule.c b/submodule.c
--- a/submodule.c
@@ -16,7 +15,6 @@
static int config_update_recurse_submodules = RECURSE_SUBMODULES_OFF;
-static struct string_list changed_submodule_names = STRING_LIST_INIT_DUP;
-+
static int initialized_fetch_ref_tips;
static struct oid_array ref_tips_before_fetch;
static struct oid_array ref_tips_after_fetch;
@@ -25,22 +23,8 @@
}
-static void calculate_changed_submodule_paths(void)
-+struct submodule_parallel_fetch {
-+ int count;
-+ struct argv_array args;
-+ struct repository *r;
-+ const char *prefix;
-+ int command_line_option;
-+ int default_option;
-+ int quiet;
-+ int result;
-+
-+ struct string_list changed_submodule_names;
-+};
-+#define SPF_INIT {0, ARGV_ARRAY_INIT, NULL, NULL, 0, 0, 0, 0, STRING_LIST_INIT_DUP }
-+
+static void calculate_changed_submodule_paths(
-+ struct submodule_parallel_fetch *spf)
++ struct string_list *changed_submodule_names)
{
struct argv_array argv = ARGV_ARRAY_INIT;
struct string_list changed_submodules = STRING_LIST_INIT_DUP;
@@ -49,30 +33,23 @@
if (!submodule_has_commits(path, commits))
- string_list_append(&changed_submodule_names, name->string);
-+ string_list_append(&spf->changed_submodule_names,
++ string_list_append(changed_submodule_names,
+ name->string);
}
free_submodules_oids(&changed_submodules);
@@
- return ret;
- }
-
--struct submodule_parallel_fetch {
-- int count;
-- struct argv_array args;
-- struct repository *r;
-- const char *prefix;
-- int command_line_option;
-- int default_option;
-- int quiet;
-- int result;
--};
+ int default_option;
+ int quiet;
+ int result;
++
++ struct string_list changed_submodule_names;
+ };
-#define SPF_INIT {0, ARGV_ARRAY_INIT, NULL, NULL, 0, 0, 0, 0}
--
++#define SPF_INIT {0, ARGV_ARRAY_INIT, NULL, NULL, 0, 0, 0, 0, STRING_LIST_INIT_DUP }
+
static int get_fetch_recurse_config(const struct submodule *submodule,
struct submodule_parallel_fetch *spf)
- {
@@
case RECURSE_SUBMODULES_ON_DEMAND:
if (!submodule ||
@@ -88,7 +65,7 @@
- calculate_changed_submodule_paths();
- string_list_sort(&changed_submodule_names);
-+ calculate_changed_submodule_paths(&spf);
++ calculate_changed_submodule_paths(&spf.changed_submodule_names);
+ string_list_sort(&spf.changed_submodule_names);
run_processes_parallel(max_parallel_jobs,
get_next_submodule,
2: a64a8322a1 ! 2: a11e6e39a2 submodule.c: do not copy around submodule list
@@ -1,6 +1,6 @@
Author: Stefan Beller <sbeller@google.com>
- submodule.c: do not copy around submodule list
+ submodule: store OIDs in changed_submodule_names
'calculate_changed_submodule_paths' uses a local list to compute the
changed submodules, and then produces the result by copying appropriate
@@ -14,13 +14,13 @@
useful in a later patch.
Signed-off-by: Stefan Beller <sbeller@google.com>
- Signed-off-by: Junio C Hamano <gitster@pobox.com>
+ Reviewed-by: Jonathan Tan <jonathantanmy@google.com>
diff --git a/submodule.c b/submodule.c
--- a/submodule.c
+++ b/submodule.c
@@
- struct submodule_parallel_fetch *spf)
+ struct string_list *changed_submodule_names)
{
struct argv_array argv = ARGV_ARRAY_INIT;
- struct string_list changed_submodules = STRING_LIST_INIT_DUP;
@@ -34,10 +34,10 @@
* commits have been recorded upstream in "changed_submodule_names".
*/
- collect_changed_submodules(&changed_submodules, &argv);
-+ collect_changed_submodules(&spf->changed_submodule_names, &argv);
++ collect_changed_submodules(changed_submodule_names, &argv);
- for_each_string_list_item(name, &changed_submodules) {
-+ for_each_string_list_item(name, &spf->changed_submodule_names) {
++ for_each_string_list_item(name, changed_submodule_names) {
struct oid_array *commits = name->util;
const struct submodule *submodule;
const char *path = NULL;
@@ -46,7 +46,7 @@
continue;
- if (!submodule_has_commits(path, commits))
-- string_list_append(&spf->changed_submodule_names,
+- string_list_append(changed_submodule_names,
- name->string);
+ if (submodule_has_commits(path, commits)) {
+ oid_array_clear(commits);
@@ -55,7 +55,7 @@
}
- free_submodules_oids(&changed_submodules);
-+ string_list_remove_empty_items(&spf->changed_submodule_names, 1);
++ string_list_remove_empty_items(changed_submodule_names, 1);
+
argv_array_clear(&argv);
oid_array_clear(&ref_tips_before_fetch);
3: 9341b92c83 ! 3: 3f4e0d4b8d repository: repo_submodule_init to take a submodule struct
@@ -5,17 +5,19 @@
When constructing a struct repository for a submodule for some revision
of the superproject where the submodule is not contained in the index,
it may not be present in the working tree currently either. In that
- siutation giving a 'path' argument is not useful. Upgrade the
+ situation giving a 'path' argument is not useful. Upgrade the
repo_submodule_init function to take a struct submodule instead.
+ The submodule struct can be obtained via submodule_from_{path, name} or
+ an artificial submodule struct can be passed in.
- While we are at it, overhaul the repo_submodule_init function by renaming
- the submodule repository struct, which is to be initialized, to a name
- that is not confused with the struct submodule as easily.
+ While we are at it, rename the repository struct in the repo_submodule_init
+ function, which is to be initialized, to a name that is not confused with
+ the struct submodule as easily. Perform such renames in similar functions
+ as well.
Also move its documentation into the header file.
Signed-off-by: Stefan Beller <sbeller@google.com>
- Signed-off-by: Junio C Hamano <gitster@pobox.com>
diff --git a/builtin/grep.c b/builtin/grep.c
--- a/builtin/grep.c
@@ -104,6 +106,19 @@
static void show_ce(struct repository *repo, struct dir_struct *dir,
+diff --git a/builtin/submodule--helper.c b/builtin/submodule--helper.c
+--- a/builtin/submodule--helper.c
++++ b/builtin/submodule--helper.c
+@@
+ if (!sub)
+ BUG("We could get the submodule handle before?");
+
+- if (repo_submodule_init(&subrepo, the_repository, path))
++ if (repo_submodule_init(&subrepo, the_repository, sub))
+ die(_("could not get a repository handle for submodule '%s'"), path);
+
+ if (!repo_config_get_string(&subrepo, "core.worktree", &cw)) {
+
diff --git a/repository.c b/repository.c
--- a/repository.c
+++ b/repository.c
@@ -178,7 +193,8 @@
+/*
+ * Initialize the repository 'subrepo' as the submodule given by the
+ * struct submodule 'sub' in parent repository 'superproject'.
-+ * Return 0 upon success and a non-zero value upon failure.
++ * Return 0 upon success and a non-zero value upon failure, which may happen
++ * if the submodule is not found, or 'sub' is NULL.
+ */
+struct submodule;
+int repo_submodule_init(struct repository *subrepo,
4: cea909cbd4 ! 4: b1566069e7 submodule: fetch in submodules git directory instead of in worktree
@@ -1,44 +1,19 @@
Author: Stefan Beller <sbeller@google.com>
- submodule: fetch in submodules git directory instead of in worktree
+ submodule: migrate get_next_submodule to use repository structs
- This patch started as a refactoring to make 'get_next_submodule' more
- readable, but upon doing so, I realized that "git fetch" of the submodule
- actually doesn't need to be run in the submodules worktree. So let's run
- it in its git dir instead.
+ We used to recurse into submodules, even if they were broken having
+ only an objects directory. The child process executed in the submodule
+ would fail though if the submodule was broken.
- That should pave the way towards fetching submodules that are currently
- not checked out.
-
- This patch leaks the cp->dir in get_next_submodule, as any further
- callback in run_processes_parallel doesn't have access to the child
- process any more. In an early iteration of this patch, the function
- get_submodule_repo_for directly returned the string containing the
- git directory, which would be a better design choice for this patch.
-
- However the next patch both fixes the memory leak of cp->dir and also has
- a use case for using the full repository handle of the submodule, so
- it makes sense to introduce the get_submodule_repo_for here already.
+ This patch tightens the check upfront, such that we do not need
+ to spawn a child process to find out if the submodule is broken.
Signed-off-by: Stefan Beller <sbeller@google.com>
- Signed-off-by: Junio C Hamano <gitster@pobox.com>
diff --git a/submodule.c b/submodule.c
--- a/submodule.c
+++ b/submodule.c
-@@
- DEFAULT_GIT_DIR_ENVIRONMENT);
- }
-
-+static void prepare_submodule_repo_env_in_gitdir(struct argv_array *out)
-+{
-+ prepare_submodule_repo_env_no_git_dir(out);
-+ argv_array_pushf(out, "%s=.", GIT_DIR_ENVIRONMENT);
-+}
-+
- /* Helper function to display the submodule header line prior to the full
- * summary output. If it can locate the submodule objects directory it will
- * attempt to lookup both the left and right commits and put them into the
@@
return spf->default_option;
}
@@ -99,9 +74,8 @@
+ if (repo) {
child_process_init(cp);
- cp->dir = strbuf_detach(&submodule_path, NULL);
-- prepare_submodule_repo_env(&cp->env_array);
-+ prepare_submodule_repo_env_in_gitdir(&cp->env_array);
-+ cp->dir = xstrdup(repo->gitdir);
+ prepare_submodule_repo_env(&cp->env_array);
++ cp->dir = xstrdup(repo->worktree);
cp->git_cmd = 1;
if (!spf->quiet)
strbuf_addf(err, "Fetching submodule %s%s\n",
@@ -113,27 +87,17 @@
+ repo_clear(repo);
+ free(repo);
ret = 1;
++ } else {
++ /*
++ * An empty directory is normal,
++ * the submodule is not initialized
++ */
++ if (S_ISGITLINK(ce->ce_mode) &&
++ !is_empty_dir(ce->name))
++ die(_("Could not access submodule '%s'"), ce->name);
}
- strbuf_release(&submodule_path);
- strbuf_release(&submodule_git_dir);
strbuf_release(&submodule_prefix);
if (ret) {
spf->count++;
-
-diff --git a/t/t5526-fetch-submodules.sh b/t/t5526-fetch-submodules.sh
---- a/t/t5526-fetch-submodules.sh
-+++ b/t/t5526-fetch-submodules.sh
-@@
-
- test_must_fail git -C dst status &&
- test_must_fail git -C dst diff &&
-- test_must_fail git -C dst fetch --recurse-submodules
-+
-+ # git-fetch cannot find the git directory of the submodule,
-+ # so it will do nothing, successfully, as it cannot distinguish between
-+ # this broken submodule and a submodule that was just set active but
-+ # not cloned yet
-+ git -C dst fetch --recurse-submodules
- '
-
- test_expect_success "fetch new commits when submodule got renamed" '
-: ---------- > 5: 2d98ff1201 submodule.c: fetch in submodules git directory instead of in worktree
5: 9ad0125310 ! 6: 092b9cbcba fetch: retry fetching submodules if needed objects were not fetched
@@ -1,25 +1,23 @@
Author: Stefan Beller <sbeller@google.com>
- fetch: retry fetching submodules if needed objects were not fetched
+ fetch: try fetching submodules if needed objects were not fetched
Currently when git-fetch is asked to recurse into submodules, it dispatches
a plain "git-fetch -C <submodule-dir>" (with some submodule related options
such as prefix and recusing strategy, but) without any information of the
remote or the tip that should be fetched.
- This works surprisingly well in some workflows (such as using submodules
- as a third party library), while not so well in other scenarios, such
- as in a Gerrit topic-based workflow, that can tie together changes
- (potentially across repositories) on the server side. One of the parts
- of such a Gerrit workflow is to download a change when wanting to examine
- it, and you'd want to have its submodule changes that are in the same
- topic downloaded as well. However these submodule changes reside in their
- own repository in their own ref (refs/changes/<int>).
+ But this default fetch is not sufficient, as a newly fetched commit in
+ the superproject could point to a commit in the submodule that is not
+ in the default refspec. This is common in workflows like Gerrit's.
+ When fetching a Gerrit change under review (from refs/changes/??), the
+ commits in that change likely point to submodule commits that have not
+ been merged to a branch yet.
- Retry fetching a submodule by object name if the object id that the
+ Try fetching a submodule by object id if the object id that the
superproject points to, cannot be found.
- This retrying does not happen when the "git fetch" done at the
+ The try does not happen when the "git fetch" done at the
superproject is not storing the fetched results in remote
tracking branches (i.e. instead just recording them to
FETCH_HEAD) in this step. A later patch will fix this.
@@ -32,7 +30,6 @@
in case the submodule recursion is set to "on".
Signed-off-by: Stefan Beller <sbeller@google.com>
- Signed-off-by: Junio C Hamano <gitster@pobox.com>
diff --git a/builtin/fetch.c b/builtin/fetch.c
--- a/builtin/fetch.c
@@ -75,16 +72,16 @@
int result;
struct string_list changed_submodule_names;
-+ struct get_next_submodule_task **retry;
-+ int retry_nr, retry_alloc;
++ struct get_next_submodule_task **fetch_specific_oids;
++ int fetch_specific_oids_nr, fetch_specific_oids_alloc;
};
-#define SPF_INIT {0, ARGV_ARRAY_INIT, NULL, NULL, 0, 0, 0, 0, STRING_LIST_INIT_DUP }
+#define SPF_INIT {0, ARGV_ARRAY_INIT, NULL, NULL, 0, 0, 0, 0, \
+ STRING_LIST_INIT_DUP, \
+ NULL, 0, 0}
- static void calculate_changed_submodule_paths(
- struct submodule_parallel_fetch *spf)
+ static int get_fetch_recurse_config(const struct submodule *submodule,
+ struct submodule_parallel_fetch *spf)
@@
return spf->default_option;
}
@@ -93,6 +90,8 @@
+ struct repository *repo;
+ const struct submodule *sub;
+ unsigned free_sub : 1; /* Do we need to free the submodule? */
++
++ /* fetch specific oids if set, otherwise fetch default refspec */
+ struct oid_array *commits;
+};
+
@@ -224,24 +223,29 @@
- repo_clear(repo);
- free(repo);
- ret = 1;
-- }
-- strbuf_release(&submodule_prefix);
-- if (ret) {
- spf->count++;
++ spf->count++;
+ *task_cb = task;
+
+ strbuf_release(&submodule_prefix);
- return 1;
-+ } else {
++ return 1;
+ } else {
++
+ get_next_submodule_task_release(task);
+ free(task);
++
+ /*
+ * An empty directory is normal,
+ * the submodule is not initialized
+@@
+ !is_empty_dir(ce->name))
+ die(_("Could not access submodule '%s'"), ce->name);
}
- }
++ }
+
-+ if (spf->retry_nr) {
-+ struct get_next_submodule_task *task = spf->retry[spf->retry_nr - 1];
++ if (spf->fetch_specific_oids_nr) {
++ struct get_next_submodule_task *task = spf->fetch_specific_oids[spf->fetch_specific_oids_nr - 1];
+ struct strbuf submodule_prefix = STRBUF_INIT;
-+ spf->retry_nr--;
++ spf->fetch_specific_oids_nr--;
+
+ strbuf_addf(&submodule_prefix, "%s%s/", spf->prefix, task->sub->path);
+
@@ -262,9 +266,13 @@
+ append_oid_to_argv, &cp->args);
+
+ *task_cb = task;
-+ strbuf_release(&submodule_prefix);
+ strbuf_release(&submodule_prefix);
+- if (ret) {
+- spf->count++;
+- return 1;
+- }
+ return 1;
-+ }
+ }
+
return 0;
}
@@ -326,9 +334,11 @@
+ return 0;
+
+ task->commits = commits;
-+ ALLOC_GROW(spf->retry, spf->retry_nr + 1, spf->retry_alloc);
-+ spf->retry[spf->retry_nr] = task;
-+ spf->retry_nr++;
++ ALLOC_GROW(spf->fetch_specific_oids,
++ spf->fetch_specific_oids_nr + 1,
++ spf->fetch_specific_oids_alloc);
++ spf->fetch_specific_oids[spf->fetch_specific_oids_nr] = task;
++ spf->fetch_specific_oids_nr++;
+ return 0;
+ }
+
@@ -346,18 +356,33 @@
test_cmp expect actual
'
-+test_expect_success "fetch new commits on-demand when they are not reachable" '
++test_expect_success "fetch new submodule commits on-demand outside standard refspec" '
++ # add a second submodule and ensure it is around in downstream first
++ git clone submodule sub1 &&
++ git submodule add ./sub1 &&
++ git commit -m "adding a second submodule" &&
++ git -C downstream pull &&
++ git -C downstream submodule update --init --recursive &&
++
+ git checkout --detach &&
++
+ C=$(git -C submodule commit-tree -m "new change outside refs/heads" HEAD^{tree}) &&
+ git -C submodule update-ref refs/changes/1 $C &&
+ git update-index --cacheinfo 160000 $C submodule &&
-+ git commit -m "updated submodule outside of refs/heads" &&
-+ D=$(git rev-parse HEAD) &&
-+ git update-ref refs/changes/2 $D &&
++ test_tick &&
++
++ D=$(git -C sub1 commit-tree -m "new change outside refs/heads" HEAD^{tree}) &&
++ git -C sub1 update-ref refs/changes/2 $D &&
++ git update-index --cacheinfo 160000 $D sub1 &&
++
++ git commit -m "updated submodules outside of refs/heads" &&
++ E=$(git rev-parse HEAD) &&
++ git update-ref refs/changes/2 $E &&
+ (
+ cd downstream &&
-+ git fetch --recurse-submodules --recurse-submodules-default on-demand origin refs/changes/2:refs/heads/my_branch &&
++ git fetch --recurse-submodules origin refs/changes/2:refs/heads/my_branch &&
+ git -C submodule cat-file -t $C &&
++ git -C sub1 cat-file -t $D &&
+ git checkout --recurse-submodules FETCH_HEAD
+ )
+'
6: b8db3b45bf < -: ---------- builtin/fetch: check for submodule updates for non branch fetches
7: 905a4f0d4f < -: ---------- fixup! repository: repo_submodule_init to take a submodule struct
-: ---------- > 7: 11bf819782 builtin/fetch: check for submodule updates in any ref update
^ permalink raw reply [relevance 8%]
* Re: [PATCH 8/9] fetch: retry fetching submodules if needed objects were not fetched
@ 2018-10-25 21:42 7% ` Stefan Beller
0 siblings, 0 replies; 200+ results
From: Stefan Beller @ 2018-10-25 21:42 UTC (permalink / raw)
To: Jonathan Tan; +Cc: Junio C Hamano, git
On Tue, Oct 23, 2018 at 4:38 PM Jonathan Tan <jonathantanmy@google.com> wrote:
>
> > > Another thing you need to clarify is what happens if the fetch-by-commit
> > > fails. Right now, it seems that it will make the whole thing fail, which
> > > might be a surprising change in behavior.
> >
> > But a positive surprise, I would assume?
>
> Whether positive or negative, I think that this needs to be mentioned in
> the commit message.
>
> As for positive or negative, I tend to agree that it's positive - sure,
> some previously successful fetches would now fail, but the results of
> those fetches could not be recursively checked out anyway.
>
> > > The test stores the result in a normal branch, not a remote tracking
> > > branch. Is storing in a normal branch required?
> >
> > In the test we fetch from another repository, such that in the
> > repository-under-test this will show up in a remote tracking branch?
I messed up there. Yes, we need to fetch into a normal branch
such that the logic of check_for_new_submodule_commits triggers
no matter where it is on the remote.
Your experiment below shows that we cannot fetch into FETCH_HEAD:
> If that were true, I would expect that when this line:
>
> > git fetch --recurse-submodules --recurse-submodules-default on-demand origin refs/changes/2:refs/heads/my_branch &&
>
> is replaced by this line:
>
> > git fetch --recurse-submodules --recurse-submodules-default on-demand origin refs/changes/2 &&
>
> then things would still work. The tests pass with the first line (after
> I fixed a type mismatch) but not with the second. (Also I don't think a
> remote-tracking branch is generated here - the output printed doesn't
> indicate so, and refs/changes/2 is not a branch anyway.)
> > > Also, do you know why this is required? A naive reading of the patch
> > > leads me to believe that this should work even if merely fetching to
> > > FETCH_HEAD.
> >
> > See the next patch, check_for_new_submodule_commits() is missing
> > for FETCH_HEAD.
>
> I see in the next patch that there is an "if" branch in
> store_updated_refs() without update_local_ref() in which
> "check_for_new_submodule_commits(&rm->old_oid)" needs to be inserted. I
> think this is a symptom that maybe check_for_new_submodule_commits()
> needs to be extracted from update_local_ref() and put into
> store_updated_refs() instead? In update_local_ref(), it is called on
> ref->new_oid, which is actually the same as rm->old_oid anyway (there is
> an oidcpy earlier).
I'll look into that.
> > > What is a "default" submodule and why would you need one?
> >
> > s/default/artificial/. Such a submodule is a submodule that has no
> > config in the .gitmodules file and its name == path.
> > We need to keep those around for historic reasons AFAICT, c.f.
> > c68f837576 (implement fetching of moved submodules, 2017-10-16)
>
> Ah, OK. I would call it a fake submodule then, and copy over the "No
> entry in .gitmodules?" comment.
"fake submodule" sounds like
http://debuggable.com/posts/git-fake-submodules:4b563ee4-f3cc-4061-967e-0e48cbdd56cb
which is what I think of when hearing fake submodules.
^ permalink raw reply [relevance 7%]
* Re: New semantic patches vs. in-flight topics [was: Re: [PATCH 00/19] Bring more repository handles into our code base]
@ 2018-10-25 19:25 3% ` Stefan Beller
0 siblings, 0 replies; 200+ results
From: Stefan Beller @ 2018-10-25 19:25 UTC (permalink / raw)
To: SZEDER Gábor; +Cc: Jeff King, git, Jonathan Tan
On Wed, Oct 24, 2018 at 6:59 PM SZEDER Gábor <szeder.dev@gmail.com> wrote:
>
> On Mon, Oct 22, 2018 at 11:54:06AM -0700, Stefan Beller wrote:
>
> > For the sake of a good history, I would think running 'make coccicheck'
> > and applying the resulting patches would be best as part of the (dirty)
> > merge of any topic that proposes new semantic patches, but that would
> > add load to Junio as it would be an extra step during the merge.
> >
> > One could argue that the step of applying such transformations into
> > the dirty merge is cheaper than resolving merge conflicts that are
> > had when the topic includes the transformation.
>
> Please consider that merge commits' have uglier diffs than regular
> commits, and that merge commits cause additional complications when
> 'git bisect' points the finger at them, both of which are exacerbated
> by additional changes squeezed into evil merges.
>
> > > Consequently, 'make coccicheck' won't run clean and the
> > > static analysis build job will fail until all those topics reach
> > > 'master', and the remaining transformations are applied on top.
> > >
> > > This was (and still is!) an issue with the hasheq()/oideq() series
> > > as well: that series was added on 2018-08-28, and the static
> > > analysis build job is red on 'pu' ever since. See the follow-up
> > > patch e43d2dcce1 (more oideq/hasheq conversions, 2018-10-02), and
> > > one more follow-up will be necessary after the builtin stash topic
> > > is merged to 'master'.
> >
> > In my understanding this follow up is a feature, as it helps to avoid
> > merge conflicts with other topics in flight.
>
> I don't see how such a follow up patch helps to avoid merge conflicts.
Well, you merge first (the new topic and the cocci patches), and then
do the transformation. But that is putting a lot more work on Junio
as the way to integrate is not just merge a new topic into the pile of
topics (whether it is pu/next/master), but to first perform a merge
of the topic and the coccinelle patches, apply the transformation
and then merge to the pile, assuming the pile is already transformed
(as it happened in "treewide: apply cocci patch" in next/pu).
> > as 'make coccicheck' is an integral part of your review?
>
> Erm, right, "review" was not the right word here. Anyway, as it is,
> 'make coccicheck' is an integral part of our automated tests, not only
> on Travis CI but on the upcoming Azure thing as well. I just try to
> pay attention to its results and the results of a bunch of my
> additional builds, and complain or even send a fix when something goes
> reproducibly wrong. This has certainly became more cumbersome with
> the permanently failing static analysis build job in the last couple
> of weeks.
I seem to not pay as much attention as I should to these,
mostly because they are unreliable on the aggregate level (a failure
there most likely means another topic than the one I am interested
broke; except in this case, where we discuss the fallout there via
this topic.)
> > I like the approach of having separate classes of semantic patches:
> > (a) the regular "we need to keep checking these" as they address
> > undesirable code patterns, which is what we currently have,
> > and what 'make coccicheck' would complain about.
> > (b) The pending patches as you propose. However I would
> > argue that we'd not want to include the transformation into
> > the same patch as then the patch will have merge conflicts.
>
> Since we have a lot of parallel running topics, merge conflicts are
> basically unavoidable anyway. If the conflicts from the
> transformation are really that severe, then perhaps the whole series
> should be postponed to a calmer, more suitable time.
Merge conflicts of this kind could be avoided, by running the
transformation on both sides before merging (or not running them
on both sides, but only after merging).
So maybe for these larger 'the_repository.pending.cocci' patches
we could get them into the tree and wait until all (most) of the topics
are including that semantic patch in their base, such that application
of the patch is easy whether before or after writing a series
(as the semantic patch is in its base).
Another short term plan would be renaming the_repository.cocci
such that it would break the 'make coccicheck'
> In the case of 'the_repository.cocci', merging its transformations
> into 'pu' resulted in only four conflicts, and I found all four on the
> easy side to resolve. I don't think it's worth waiting with the
> transformations in this particular case.
I am not worried about the current conflicts, but about those to come
in new series.
>
> > Ideally we'd have an automated process/bot that would apply
> > all pending semantic patches onto master and then checks for
> > conflicts in HEAD..pu, and only sends off the non-conflicting
> > diffs as a topic.
>
> New semantic patches didn't pop up all that frequently in the past, so
> I'm not sure it's worth investing in such an automation. Of course
> they can become more frequent in the future, and in that case we might
> want to reconsider it. Unfortunately, however, Coccinelle's results
> can't be completely trusted, either because our semantic patches or
> because Coccinelle itself are buggy...
For the first, we could just be become better at reviewing the Cocci
patch. ;-) (Well, they are harder to review than usual patches, so
this doesn't surprise me. Also not fully understanding the whole
tool may have impact on reviewability)
>
> > Then after a couple integration cycles we'd have all pending
> > changes in, with no conflicts on Junios side.
> >
> > So I think we should add a patch like you post, but we would
> > need to discuss the exact approach how to deal with pending
> > patches. Is it the original dev who should push forward on their
> > own pending patches, or does it become a pooled effort?
>
> Well, it makes sense to me that whoever proposes a change with an
> accompanying new semantic patch should also deal with the necessary
> followups. However, it doesn't really matter who deals with them, as
> long as somebody deals with them. I don't think it's much different
> from e.g. sending a followup bugfix to someone else's patch series.
Ok, I plan on resending these patches after I get
origin/sb/submodule-recursive-fetch-gets-the-tip going
resent.
Stefan
^ permalink raw reply [relevance 3%]
* Re: [PATCH v7 00/10] Make submodules work if .gitmodules is not checked out
2018-10-25 16:18 6% [PATCH v7 00/10] Make submodules work if .gitmodules is not checked out Antonio Ospite
@ 2018-10-25 18:49 4% ` Stefan Beller
2018-10-26 1:59 4% ` Junio C Hamano
0 siblings, 1 reply; 200+ results
From: Stefan Beller @ 2018-10-25 18:49 UTC (permalink / raw)
To: Antonio Ospite
Cc: Junio C Hamano, git, Jonathan Nieder, Jeff King,
SZEDER Gábor
> In this series I am addressing the comments by Stefan Beller about the
> tests in patch 9.
>
> If the new tests look OK, I'd say we try moving the series to "next" and
> see what happens?
Sounds good to me.
Thanks,
Stefan
^ permalink raw reply [relevance 4%]
* Re: Confusing behavior with ignored submodules and `git commit -a`
@ 2018-10-25 18:26 7% ` Stefan Beller
0 siblings, 0 replies; 200+ results
From: Stefan Beller @ 2018-10-25 18:26 UTC (permalink / raw)
To: mforney; +Cc: git
On Thu, Oct 25, 2018 at 11:03 AM Michael Forney <mforney@mforney.org> wrote:
>
> On 2018-03-16, Michael Forney <mforney@mforney.org> wrote:
> > Hi,
> >
> > In the past few months have noticed some confusing behavior with
> > ignored submodules. I finally got around to bisecting this to commit
> > 5556808690ea245708fb80383be5c1afee2fb3eb (add, reset: ensure
> > submodules can be added or reset).
Uh. :(
See the discussion starting at
https://public-inbox.org/git/20170725213928.125998-4-bmwill@google.com/
specifically
https://public-inbox.org/git/xmqqinieq49v.fsf@gitster.mtv.corp.google.com/
> >
> > Here is a demonstration of the problem:
> >
[...]
> > Up to here is all expected.
Makes sense.
> > However, if I go to update `foo.txt` and
> > commit with `git commit -a`, changes to inner get recorded
> > unexpectedly. What's worse is the shortstat output of `git commit -a`,
> > and the diff output of `git show` give no indication that the
> > submodule was changed.
This is really bad. git-status and git-commit share some code,
and we'll populate the commit message with a status output.
So it seems reasonable to expect the status and the commit to match,
i.e. if status tells me there is no change, then commit should not record
the submodule update.
> > $ git commit -a -m 'update foo.txt'
> > [master 6ec564c] update foo.txt
> > 1 file changed, 1 insertion(+), 1 deletion(-)
> > $ git show
> > commit 6ec564c15ddae099c71f01750b4c434557525653 (HEAD -> master)
> > Author: Michael Forney <mforney@mforney.org>
> > Date: Fri Mar 16 20:18:37 2018 -0700
> >
> > update foo.txt
> >
> > diff --git a/foo.txt b/foo.txt
> > index d00491f..0cfbf08 100644
> > --- a/foo.txt
> > +++ b/foo.txt
> > @@ -1 +1 @@
> > -1
> > +2
> > $
> >
> > There have been a couple occasions where I accidentally pushed local
> > changes to ignored submodules because of this. Since they don't show
> > up in the log output, it is difficult to figure out what actually has
> > gone wrong.
How was it prevented before? Just by git commit -a not picking up the
submodule change?
> >
> > Anyway, since the bisected commit (555680869) only mentions add and
> > reset, I'm assuming that this is a regression and not a deliberate
> > behavior change. The documentation for submodule.<name>.ignore states
> > that the setting should only affect `git status` and the diff family.
> > In terms of my expectations, I would go further and say it should only
> > affect `git status` and diffs against the working tree.
> >
> > I took a brief look through the relevant sources, and it wasn't clear
> > to me how to fix this without accidentally changing the behavior of
> > other subcommands.
> >
> > Any help with this issue is appreciated!
I guess reverting that commit is not a good idea now, as
I would expect something to break.
Maybe looking through the series 614ea03a71
(Merge branch 'bw/submodule-config-cleanup', 2017-08-26)
to understand why it happened in the context would be a good start.
> I accidentally pushed local changes to ignored submodules again due to this.
>
> Can anyone confirm whether this is the intended behavior of ignore? If
> it is, then at least the documentation needs an update saying that
> `commit -a` will commit all submodule changes, even if they are
> ignored.
The docs say "(but it will nonetheless show up in the output of
status and commit when it has been staged)" as well, so that commit
sounds like a regression?
Stefan
^ permalink raw reply [relevance 7%]
* [PATCH v7 00/10] Make submodules work if .gitmodules is not checked out
@ 2018-10-25 16:18 6% Antonio Ospite
2018-10-25 18:49 4% ` Stefan Beller
0 siblings, 1 reply; 200+ results
From: Antonio Ospite @ 2018-10-25 16:18 UTC (permalink / raw)
To: gitster
Cc: git, Jonathan Nieder, Stefan Beller, Jeff King, SZEDER Gábor,
Antonio Ospite
Hi,
this series teaches git to try and read the .gitmodules file from the
index (:.gitmodules) or from the current branch (HEAD:.gitmodules) when
the file is not readily available in the working tree.
This can be used, together with sparse checkouts, to enable submodule
usage with programs like vcsh[1] which manage multiple repositories with
their working trees sharing the same path.
[1] https://github.com/RichiH/vcsh
In the previous series there were some comments about not using the enum
in patch 8, but I decided to leave the code as it was as I still think
that it make sense to use an enum there, and have the default value
unnamed; during the discussion I pointed out that other code in git do
the same.
In this series I am addressing the comments by Stefan Beller about the
tests in patch 9.
If the new tests look OK, I'd say we try moving the series to "next" and
see what happens?
I am available to address any further concerns in follow-up patches.
v6 of the series is here:
https://public-inbox.org/git/20181005130601.15879-1-ao2@ao2.it/
v5 of the series is here:
https://public-inbox.org/git/20180917140940.3839-1-ao2@ao2.it/
v4 of the series is here:
https://public-inbox.org/git/20180824132951.8000-1-ao2@ao2.it/
v3 of the series is here:
https://public-inbox.org/git/20180814110525.17801-1-ao2@ao2.it/
v2 of the series is here:
https://public-inbox.org/git/20180802134634.10300-1-ao2@ao2.it/
v1 of the series, with some background info, is here:
https://public-inbox.org/git/20180514105823.8378-1-ao2@ao2.it/
Changes since v6:
* Renamed t7416-submodule-sparse-gitmodules.sh to
t7418-submodule-sparse-gitmodules.sh because t7416 was already
taken. This has been already taken care of by Junio in "pu".
* Improved tests in t7418: now, instead of just testing the return
value of "git submodule ..." commands when .gitmodules is not in the
working tree, the actual use case is checked in each test, with pre-
and post-conditions.
Thank you,
Antonio
Antonio Ospite (10):
submodule: add a print_config_from_gitmodules() helper
submodule: factor out a config_set_in_gitmodules_file_gently function
t7411: merge tests 5 and 6
t7411: be nicer to future tests and really clean things up
submodule--helper: add a new 'config' subcommand
submodule: use the 'submodule--helper config' command
t7506: clean up .gitmodules properly before setting up new scenario
submodule: add a helper to check if it is safe to write to .gitmodules
submodule: support reading .gitmodules when it's not in the working
tree
t/helper: add test-submodule-nested-repo-config
Makefile | 1 +
builtin/grep.c | 17 ++-
builtin/submodule--helper.c | 40 ++++++
cache.h | 2 +
git-submodule.sh | 13 +-
submodule-config.c | 68 ++++++++-
submodule-config.h | 2 +
submodule.c | 28 +++-
submodule.h | 1 +
t/helper/test-submodule-nested-repo-config.c | 30 ++++
t/helper/test-tool.c | 1 +
t/helper/test-tool.h | 1 +
t/t7411-submodule-config.sh | 141 +++++++++++++++++--
t/t7418-submodule-sparse-gitmodules.sh | 122 ++++++++++++++++
t/t7506-status-submodule.sh | 3 +-
t/t7814-grep-recurse-submodules.sh | 16 +++
16 files changed, 454 insertions(+), 32 deletions(-)
create mode 100644 t/helper/test-submodule-nested-repo-config.c
create mode 100755 t/t7418-submodule-sparse-gitmodules.sh
--
Antonio Ospite
https://ao2.it
https://twitter.com/ao2it
A: Because it messes up the order in which people normally read text.
See http://en.wikipedia.org/wiki/Posting_style
Q: Why is top-posting such a bad thing?
^ permalink raw reply [relevance 6%]
* Re: [PATCH 18/19] submodule: use submodule repos for object lookup
2018-10-16 23:35 20% ` [PATCH 18/19] submodule: use submodule repos for object lookup Stefan Beller
@ 2018-10-25 9:14 8% ` SZEDER Gábor
0 siblings, 0 replies; 200+ results
From: SZEDER Gábor @ 2018-10-25 9:14 UTC (permalink / raw)
To: Stefan Beller; +Cc: Derrick Stolee, git, jonathantanmy
On Tue, Oct 16, 2018 at 04:35:49PM -0700, Stefan Beller wrote:
> This converts the 'show_submodule_header' function to use
> the repository API properly, such that the submodule objects
> are not added to the main object store.
This patch breaks the test suite with 'GIT_TEST_COMMIT_GRAPH=1', in
particular 't4041-diff-submodule-option.sh' fails with:
expecting success:
git diff-index -p --submodule=log HEAD >actual &&
cat >expected <<-EOF &&
Submodule sm1 $head2..$head3 (rewind):
< Add foo3 ($added foo3)
< Add foo2 ($added foo2)
EOF
test_cmp expected actual
+ git diff-index -p --submodule=log HEAD
+ cat
+ test_cmp expected actual
+ diff -u expected actual
--- expected 2018-10-25 09:10:00.541610016 +0000
+++ actual 2018-10-25 09:10:00.537609885 +0000
@@ -1,3 +1,5 @@
-Submodule sm1 30b9670..dafb207 (rewind):
+Submodule sm1 30b9670...dafb207:
< Add foo3 (hinzugefügt foo3)
< Add foo2 (hinzugefügt foo2)
+ > Add foo1 (hinzugefügt foo1)
+ < Add foo1 (hinzugefügt foo1)
error: last command exited with $?=1
not ok 9 - modified submodule(backward)
and 't4060-diff-submodule-option-diff-format.sh' fails with:
expecting success:
git diff-index -p --submodule=diff HEAD >actual &&
cat >expected <<-EOF &&
Submodule sm1 $head2..$head3 (rewind):
diff --git a/sm1/foo2 b/sm1/foo2
deleted file mode 100644
index 54b060e..0000000
--- a/sm1/foo2
+++ /dev/null
@@ -1 +0,0 @@
-foo2
diff --git a/sm1/foo3 b/sm1/foo3
deleted file mode 100644
index c1ec6c6..0000000
--- a/sm1/foo3
+++ /dev/null
@@ -1 +0,0 @@
-foo3
EOF
test_cmp expected actual
+ git diff-index -p --submodule=diff HEAD
+ cat
+ test_cmp expected actual
+ diff -u expected actual
--- expected 2018-10-25 09:10:38.854868800 +0000
+++ actual 2018-10-25 09:10:38.854868800 +0000
@@ -1,4 +1,4 @@
-Submodule sm1 30b9670..dafb207 (rewind):
+Submodule sm1 30b9670...dafb207:
diff --git a/sm1/foo2 b/sm1/foo2
deleted file mode 100644
index 54b060e..0000000
error: last command exited with $?=1
not ok 10 - modified submodule(backward)
^ permalink raw reply [relevance 8%]
* Re: Bug with "git mv" and submodules, and with "git submodule add something --force"
2018-10-22 21:52 12% ` Stefan Beller
@ 2018-10-24 7:18 7% ` Junio C Hamano
0 siblings, 0 replies; 200+ results
From: Junio C Hamano @ 2018-10-24 7:18 UTC (permalink / raw)
To: Stefan Beller; +Cc: git, jrnieder, juergen.vogl
Stefan Beller <sbeller@google.com> writes:
>> You would want to be able to remove a submodule and replace it with
>> a directory, but you can probably do it in two steps, i.e.
>>
>> git reset --hard
>> git rm --cached sha1collisiondetection
>> echo Now a regular dir >sha1collisiondetection/READ.ME
>> find sha1collisiondetection ! -type d -print0 |
>> git update-index --add --stdin -z
>
> "Ignoring path sha1collisiondetection/.git"
>
> Nice!
There actually is another possible outcome that anybody following
along must be aware of and be careful about: not even .git directory
exist there, i.e. it is possible that the submodule has never been
init'ed.
So, it is not that nice X-<.
^ permalink raw reply [relevance 7%]
* Re: git pull defaults for recursesubmodules
2018-10-23 21:57 5% ` Stefan Beller
@ 2018-10-24 5:32 2% ` Tommi Vainikainen
0 siblings, 0 replies; 200+ results
From: Tommi Vainikainen @ 2018-10-24 5:32 UTC (permalink / raw)
To: sbeller; +Cc: git
ke 24. lokak. 2018 klo 0.57 Stefan Beller (sbeller@google.com) kirjoitti:
> On Tue, Oct 23, 2018 at 2:04 PM Tommi Vainikainen <tvainika@gmail.com> wrote:
> > I would expect that if git-config has fetch.recurseSubmodules set,
> > also git pull should use this setting, or at least similar option such
>
> This makes sense to me and the patch looks good to me.
> It is unclear to me if this is a regression or an oversight of
> of a6d7eb2c7a (pull: optionally rebase submodules (remote
> submodule changes only), 2017-06-23)
With my testing, no it was not regression at least from that commit.
It did not work as I expected before that commit.
What was unclear to me is why this config needs to be read as
pull calls fetch, why fetch does not use this configuration as is?
If fetch.prune=1, should git pull command also prune or not during
fetch? There does not seem to be test case for that behavior.
--
Tommi Vainikainen
^ permalink raw reply [relevance 2%]
* Re: [PATCH 7/9] submodule: fetch in submodules git directory instead of in worktree
@ 2018-10-23 23:01 5% ` Stefan Beller
0 siblings, 0 replies; 200+ results
From: Stefan Beller @ 2018-10-23 23:01 UTC (permalink / raw)
To: Jonathan Tan; +Cc: Junio C Hamano, git
On Tue, Oct 23, 2018 at 3:55 PM Jonathan Tan <jonathantanmy@google.com> wrote:
> > When adding the comment here, we'd also want to have
> > the comment in prepare_submodule_repo_env, which
> > could be its own preparation commit.
>
> I agree with the protection. As for the preparation commit, I don't
> think it's always the code author's responsibility to tidy up the
> surrounding code, but since you're adding an identical comment here,
> it's probably worth it to add the comment there too.
Am I the only one who dislikes inconsistent files? ;-)
(ie. clean in one place, not cleaned up in another)
I can see your point. Will add a comment
> > Thinking of that, maybe we need to announce that in get_next_submodule
>
> The consequence of getting caught changes, though. Currently,
> spf->result is set to 1 whenever a child process fails. But in this
> patch, some of these repositories would be entirely skipped, meaning
> that no child process is run, and spf->result is never modified.
Right.
> > If the working tree directory is empty for that submodule, it means
> > it is likely not initialized. But why would we use that as a signal to
> > skip the submodule?
>
> What I meant was: if empty, skip it completely. Otherwise, do the
> repo_submodule_init() and repo_init() thing, and if they both fail, set
> spf->result to 1, preserving existing behavior.
I did it the other way round:
If repo_[submodule_]init fails, see if we have a gitlink in tree and
an empty dir in the FS, to decide if we need to signal failure.
I can switch it around again, but it seemed easier to write as
that puts corner cases away into one else {} case.
^ permalink raw reply [relevance 5%]
* Re: [PATCH 8/9] fetch: retry fetching submodules if needed objects were not fetched
@ 2018-10-23 22:37 7% ` Stefan Beller
0 siblings, 1 reply; 200+ results
From: Stefan Beller @ 2018-10-23 22:37 UTC (permalink / raw)
To: Jonathan Tan; +Cc: Junio C Hamano, git
On Wed, Oct 17, 2018 at 5:40 PM Jonathan Tan <jonathantanmy@google.com> wrote:
>
> > Currently when git-fetch is asked to recurse into submodules, it dispatches
> > a plain "git-fetch -C <submodule-dir>" (with some submodule related options
> > such as prefix and recusing strategy, but) without any information of the
> > remote or the tip that should be fetched.
> >
> > This works surprisingly well in some workflows (such as using submodules
> > as a third party library), while not so well in other scenarios, such
> > as in a Gerrit topic-based workflow, that can tie together changes
> > (potentially across repositories) on the server side. One of the parts
> > of such a Gerrit workflow is to download a change when wanting to examine
> > it, and you'd want to have its submodule changes that are in the same
> > topic downloaded as well. However these submodule changes reside in their
> > own repository in their own ref (refs/changes/<int>).
>
> It seems that the pertinent situation is when, in the superproject, you
> fetch a commit (which may be the target of a ref, or an ancestor of the
> target of a ref) that points to a submodule commit that was not fetched
> by the default refspec-less fetch that you describe in the first
> paragraph. I would just describe it as follows:
>
> But this default fetch is not sufficient, as a newly fetched commit in
> the superproject could point to a commit in the submodule that is not
> in the default refspec. This is common in workflows like Gerrit's.
> When fetching a Gerrit change under review (from refs/changes/??), the
> commits in that change likely point to submodule commits that have not
> been merged to a branch yet.
Makes sense.
> Another thing you need to clarify is what happens if the fetch-by-commit
> fails. Right now, it seems that it will make the whole thing fail, which
> might be a surprising change in behavior.
But a positive surprise, I would assume?
> The test stores the result in a normal branch, not a remote tracking
> branch. Is storing in a normal branch required?
In the test we fetch from another repository, such that in the
repository-under-test this will show up in a remote tracking branch?
> Also, do you know why this is required? A naive reading of the patch
> leads me to believe that this should work even if merely fetching to
> FETCH_HEAD.
See the next patch, check_for_new_submodule_commits() is missing
for FETCH_HEAD.
>
> > +struct get_next_submodule_task {
> > + struct repository *repo;
> > + const struct submodule *sub;
> > + unsigned free_sub : 1; /* Do we need to free the submodule? */
> > + struct oid_array *commits;
> > +};
>
> Firstly, I don't think "commits" needs to be a pointer.
>
> Document at least "commits". As far as I can tell, if NULL (or empty if
> you take my suggestion), this means that this task is the first pass for
> this particular submodule and the fetch needs to be done using the
> default refspec. Otherwise, this task is the second pass for this
> particular submodule and the fetch needs to be done passing the
> contained OIDs.
Makes sense. I think I'll make it a non-pointer, but will introduce another
flag or counter for the phase.
>
> > +static const struct submodule *get_default_submodule(const char *path)
> > +{
> > + struct submodule *ret = NULL;
> > + const char *name = default_name_or_path(path);
> > +
> > + if (!name)
> > + return NULL;
> > +
> > + ret = xmalloc(sizeof(*ret));
> > + memset(ret, 0, sizeof(*ret));
> > + ret->path = name;
> > + ret->name = name;
> > +
> > + return (const struct submodule *) ret;
> > +}
>
> What is a "default" submodule and why would you need one?
s/default/artificial/. Such a submodule is a submodule that has no
config in the .gitmodules file and its name == path.
We need to keep those around for historic reasons AFAICT, c.f.
c68f837576 (implement fetching of moved submodules, 2017-10-16)
> > + task = get_next_submodule_task_create(spf->r, ce->name);
> > +
> > + if (!task->sub) {
> > + free(task);
> > + continue;
> > }
>
> Will task->sub ever be NULL?
Yes, if we fall back to these "default" submodule and just try if it
can be handled
as a submodule, but it cannot be handled as such,
get_next_submodule_task_create has
task->sub = submodule_from_path(r, &null_oid, path);
if (!task->sub) {
task->sub = get_default_submodule(path);
and get_default_submodule can return NULL.
>
> > + if (spf->retry_nr) {
> > + struct get_next_submodule_task *task = spf->retry[spf->retry_nr - 1];
> > + struct strbuf submodule_prefix = STRBUF_INIT;
> > + spf->retry_nr--;
> > +
> > + strbuf_addf(&submodule_prefix, "%s%s/", spf->prefix, task->sub->path);
> > +
> > + child_process_init(cp);
> > + prepare_submodule_repo_env_in_gitdir(&cp->env_array);
> > + cp->git_cmd = 1;
> > + cp->dir = task->repo->gitdir;
> > +
> > + argv_array_init(&cp->args);
> > + argv_array_pushv(&cp->args, spf->args.argv);
> > + argv_array_push(&cp->args, "on-demand");
>
> This means that we need to trust that the last entry in spf->args can
> take an "on-demand" parameter. Could we supply that argument here
> explicitly instead?
>
> > + argv_array_push(&cp->args, "--submodule-prefix");
> > + argv_array_push(&cp->args, submodule_prefix.buf);
> > +
> > + /* NEEDSWORK: have get_default_remote from s--h */
>
> What is s--h?
builtin/submodule--helper, will clarify
>
> > +static int commit_exists_in_sub(const struct object_id *oid, void *data)
> > +{
> > + struct repository *subrepo = data;
> > +
> > + enum object_type type = oid_object_info(subrepo, oid, NULL);
> > +
> > + return type != OBJ_COMMIT;
> > +}
>
> Shouldn't the function name be commit_missing_in_sub?
yes.
>
> > static int fetch_finish(int retvalue, struct strbuf *err,
> > void *cb, void *task_cb)
> > {
> > struct submodule_parallel_fetch *spf = cb;
> > + struct get_next_submodule_task *task = task_cb;
> > + const struct submodule *sub;
> > +
> > + struct string_list_item *it;
> > + struct oid_array *commits;
> >
> > if (retvalue)
> > spf->result = 1;
> >
> > + if (!task)
> > + return 0;
>
> When will task be NULL?
>
> > +
> > + sub = task->sub;
> > + if (!sub)
> > + goto out;
>
> Same as above - when will sub be NULL?
>
> > + it = string_list_lookup(&spf->changed_submodule_names, sub->name);
> > + if (!it)
> > + goto out;
>
> And "it" as well.
I'll add comments.
>
> > + commits = it->util;
> > + oid_array_filter(commits,
> > + commit_exists_in_sub,
> > + task->repo);
> > +
> > + /* Are there commits that do not exist? */
> > + if (commits->nr) {
> > + /* We already tried fetching them, do not try again. */
> > + if (task->commits)
> > + return 0;
>
> Clearer and more efficient if the check for task->commits was first
> before all the filtering.
>
> > +test_expect_success "fetch new commits on-demand when they are not reachable" '
>
> "not reachable" confused me - they are indeed reachable, just not from
> the default refspec.
Makes sense
>
> > + git checkout --detach &&
> > + C=$(git -C submodule commit-tree -m "new change outside refs/heads" HEAD^{tree}) &&
> > + git -C submodule update-ref refs/changes/1 $C &&
> > + git update-index --cacheinfo 160000 $C submodule &&
> > + git commit -m "updated submodule outside of refs/heads" &&
> > + D=$(git rev-parse HEAD) &&
> > + git update-ref refs/changes/2 $D &&
> > + (
> > + cd downstream &&
> > + git fetch --recurse-submodules --recurse-submodules-default on-demand origin refs/changes/2:refs/heads/my_branch &&
> > + git -C submodule cat-file -t $C &&
> > + git checkout --recurse-submodules FETCH_HEAD
> > + )
> > +'
>
> For maximum test coverage, I think this test should involve 2
> submodules:
>
> B C E F
> \ / \ /
> A D
>
> and the upstream superproject having:
>
> G -> H -> I
>
> The downstream superproject already has G, and is fetching I. In H, the
> submodule gitlinks point to B and E respectively, and in I, the
> submodule gitlinks point to C and F respectively. This ensures that both
> multiple submodules work, and that submodule commits are also fetched
> for interior superproject commits.
ok.
^ permalink raw reply [relevance 7%]
* Re: git pull defaults for recursesubmodules
@ 2018-10-23 21:57 5% ` Stefan Beller
2018-10-24 5:32 2% ` Tommi Vainikainen
0 siblings, 1 reply; 200+ results
From: Stefan Beller @ 2018-10-23 21:57 UTC (permalink / raw)
To: tvainika; +Cc: git
On Tue, Oct 23, 2018 at 2:04 PM Tommi Vainikainen <tvainika@gmail.com> wrote:
>
> I configured my local git to fetch with recurseSubmodules = on-demand,
> which I found the most convenient setting. However then I noticed that
> I mostly use git pull actually to fetch from remotes, but git pull
> does not utilize any recurseSubmoddules setting now, or at least I
> could not find such.
>
> I would expect that if git-config has fetch.recurseSubmodules set,
> also git pull should use this setting, or at least similar option such
> as pull.recurseSubmodules should be available. I'd prefer sharing
> fetch.recurseSubmodules setting here.
>
> I've attached a minimal patch, which I believe implements this
> configuration usage, and a test case to show my expected behavior for
> git pull.
This makes sense to me and the patch looks good to me.
It is unclear to me if this is a regression or an oversight of
of a6d7eb2c7a (pull: optionally rebase submodules (remote
submodule changes only), 2017-06-23)
Thanks,
Stefan
^ permalink raw reply [relevance 5%]
* Re: [PATCH 7/9] submodule: fetch in submodules git directory instead of in worktree
@ 2018-10-23 18:26 7% ` Stefan Beller
0 siblings, 1 reply; 200+ results
From: Stefan Beller @ 2018-10-23 18:26 UTC (permalink / raw)
To: Jonathan Tan; +Cc: Junio C Hamano, git
On Wed, Oct 17, 2018 at 3:58 PM Jonathan Tan <jonathantanmy@google.com> wrote:
>
> > This patch started as a refactoring to make 'get_next_submodule' more
> > readable, but upon doing so, I realized that "git fetch" of the submodule
> > actually doesn't need to be run in the submodules worktree. So let's run
> > it in its git dir instead.
>
> The commit message needs to be updated, I think - this patch does
> significantly more than fetching in the gitdir.
From my point of view, it is not significant, but refactoring.
I'll think how to write a better commit message.
> > This patch leaks the cp->dir in get_next_submodule, as any further
> > callback in run_processes_parallel doesn't have access to the child
> > process any more.
>
> The cp->dir is already leaked - probably better to write "cp->dir in
> get_next_submodule() is still leaked, but this will be fixed in a
> subsequent patch".
... which fails to mention the reason why (as it is hard to do given
the current design) but is more concise.
> > +static void prepare_submodule_repo_env_in_gitdir(struct argv_array *out)
> > +{
> > + prepare_submodule_repo_env_no_git_dir(out);
> > + argv_array_pushf(out, "%s=.", GIT_DIR_ENVIRONMENT);
>
> Why does GIT_DIR need to be set? Is it to avoid subcommands recursively
> checking the parent directories in case the CWD is a malformed Git
> repository? If yes, maybe it's worth adding a comment.
It is copying the structure from prepare_submodule_repo_env,
specifically 10f5c52656 (submodule: avoid auto-discovery in
prepare_submodule_repo_env(), 2016-09-01), which sounds
appealing (and brings real benefits for the working directory),
but I have not thought about this protection for the git dir.
Maybe another approach is to not set the cwd for the child process
and instead point GIT_DIR_ENVIRONMENT only to the right
directory.
Then the use of GIT_DIR_ENVIRONMENT is obvious and
is not just for protection of corner cases.
However I think this protection is really valuable for the
.git dir as well as the submodule may be broken and we do not
want to end up in an infinite loop (as the discovery would find
the superproject which then tries to recurse, again, into the
submodule with the broken git dir)
When adding the comment here, we'd also want to have
the comment in prepare_submodule_repo_env, which
could be its own preparation commit.
> > +static struct repository *get_submodule_repo_for(struct repository *r,
> > + const struct submodule *sub)
> > +{
> > + struct repository *ret = xmalloc(sizeof(*ret));
> > +
> > + if (repo_submodule_init(ret, r, sub)) {
> > + /*
> > + * No entry in .gitmodules? Technically not a submodule,
> > + * but historically we supported repositories that happen to be
> > + * in-place where a gitlink is. Keep supporting them.
> > + */
> > + struct strbuf gitdir = STRBUF_INIT;
> > + strbuf_repo_worktree_path(&gitdir, r, "%s/.git", sub->path);
> > + if (repo_init(ret, gitdir.buf, NULL)) {
> > + strbuf_release(&gitdir);
> > + return NULL;
> > + }
> > + strbuf_release(&gitdir);
> > + }
> > +
> > + return ret;
> > +}
>
> This is the significant thing that this patch does more - an unskipped
> submodule is now something that either passes the checks in
> repo_submodule_init() or the checks in repo_init(), which seems to be
> stricter than the current check that ".git" points to a directory or is
> one. This means that we skip certain broken repositories, and this
> necessitates a change in the test.
I see. However there is no change in function, the check in repo_init
(or repo_submodule_init) is less strict than the check in the child process.
So if there are broken submodule repositories, the difference of this
patch is the layer at which it is caught, i.e. we would not spawn a child
that fails, but skip the submodule.
Thinking of that, maybe we need to announce that in get_next_submodule
>
> I think we should be more particular about what we're allowed to skip -
> in particular, maybe if we're planning to skip this submodule, its
> corresponding directory in the worktree (if one exists) needs to be
> empty.
If the working tree directory is empty for that submodule, it means
it is likely not initialized. But why would we use that as a signal to
skip the submodule?
> > - cp->dir = strbuf_detach(&submodule_path, NULL);
> > - prepare_submodule_repo_env(&cp->env_array);
> > + prepare_submodule_repo_env_in_gitdir(&cp->env_array);
> > + cp->dir = xstrdup(repo->gitdir);
>
> Here is where the functionality change (fetch in ".git") described in
> the commit message occurs.
True.
Thanks for the review, I'll try to split up this commit a bit more and
explain each part on its own.
^ permalink raw reply [relevance 7%]
* Bug with "git mv" and submodules, and with "git submodule add something --force"
2018-10-22 2:52 5% ` Junio C Hamano
@ 2018-10-22 21:52 12% ` Stefan Beller
2018-10-24 7:18 7% ` Junio C Hamano
0 siblings, 1 reply; 200+ results
From: Stefan Beller @ 2018-10-22 21:52 UTC (permalink / raw)
To: gitster; +Cc: git, jrnieder, juergen.vogl, sbeller
On Sun, Oct 21, 2018 at 7:52 PM Junio C Hamano <gitster@pobox.com> wrote:
>
> Jonathan Nieder <jrnieder@gmail.com> writes:
>
> > Stefan Beller wrote:
> >
> >> Maybe for now we can do with just an update of the documentation/bugs
> >> section and say we cannot move files in and out of submodules?
> >
> > I think we have some existing logic to prevent "git add"-ing a file
> > within a submodule to the superproject, for example.
>
> There is die_path_inside_submodule() that sanity-checks the pathspec
> and rejects. But I think that is done primarily to give an error
> message and not strictly necesary for correctness.
c08397e3aa (pathspec: remove PATHSPEC_STRIP_SUBMODULE_SLASH_EXPENSIVE
flag, 2017-05-11) seems to be relevant here, as that factors out the
warning.
> The real safety of "git add" is its call to dir.c::fill_directory();
> it collects untracked paths that match the pathspec so that they can
> be added as new paths, but because it won't cross the module
> boundary, you won't get such a path in the index to begin with.
It would not cross the boundary and fail silently as it would
treat a path into the submodule as a no-op.
> > So "git mv" should learn the same trick. And perhaps the trick needs
> > to be moved down a layer (e.g. into the index API). Hints?
Yeah, I think we'd want to teach git-mv about that trick.
Unfortunately git-mv is one of the last remainders of not
properly using pathspecs, and die_path_inside_submodule
expects a pathspec. :-/
> You would want to be able to remove a submodule and replace it with
> a directory, but you can probably do it in two steps, i.e.
>
> git reset --hard
> git rm --cached sha1collisiondetection
> echo Now a regular dir >sha1collisiondetection/READ.ME
> find sha1collisiondetection ! -type d -print0 |
> git update-index --add --stdin -z
"Ignoring path sha1collisiondetection/.git"
Nice!
>
> So from that point of view, forbidding (starting from the same state
> of our project) this sequence:
>
> git reset --hard
> echo Now a regular dir >sha1collisiondetection/READ.ME
> find sha1collisiondetection ! -type d -print0 |
> git update-index --add --remove --stdin -z
>
> that would nuke the submodule and replace it with a directory within
> which there are files would be OK. Making the latter's default
> rejection overridable with ADD_CACHE_OK_TO_REPLACE would also be
> fine.
I am having trouble of relating these commands to the original git-mv
across submodule boundaries.
Moving files from the submodule out to the superproject, seems
to fail properly, though having a less-than-optimal error message:
$ git mv sha1collisiondetection/LICENSE.txt .
fatal: not under version control, source=sha1collisiondetection/LICENSE.txt, destination=LICENSE.txt
and moving things inside was the original report, below is a proof of concept
that would yield
./git mv -v cache.h sha1collisiondetection/
fatal: moving across submodule boundaries not supported, source=cache.h, destination=sha1collisiondetection/cache.h
--8<--
Subject: [WIP/PATCH] builtin/mv.c: disallow moving across submodule boundaries
Signed-off-by: Stefan Beller <sbeller@google.com>
---
builtin/mv.c | 9 ++++++++-
1 file changed, 8 insertions(+), 1 deletion(-)
diff --git a/builtin/mv.c b/builtin/mv.c
index 80bb967a63..9ec4b2f0a3 100644
--- a/builtin/mv.c
+++ b/builtin/mv.c
@@ -172,7 +172,7 @@ int cmd_mv(int argc, const char **argv, const char *prefix)
/* Checking */
for (i = 0; i < argc; i++) {
const char *src = source[i], *dst = destination[i];
- int length, src_is_dir;
+ int length, src_is_dir, pos;
const char *bad = NULL;
if (show_only)
@@ -243,6 +243,13 @@ int cmd_mv(int argc, const char **argv, const char *prefix)
else
string_list_insert(&src_for_dst, dst);
+ pos = cache_name_pos(dst, strlen(dst));
+ if (pos < 0) {
+ pos = -(pos + 1);
+ if (!S_ISGITLINK(active_cache[pos]->ce_mode))
+ bad = _("moving across submodule boundaries not supported");
+ }
+
if (!bad)
continue;
if (!ignore_errors)
--
2.19.0
^ permalink raw reply related [relevance 12%]
* Re: Bug with "git mv" and submodules, and with "git submodule add something --force"
2018-10-19 20:58 7% ` Jonathan Nieder
@ 2018-10-22 2:52 5% ` Junio C Hamano
2018-10-22 21:52 12% ` Stefan Beller
0 siblings, 1 reply; 200+ results
From: Junio C Hamano @ 2018-10-22 2:52 UTC (permalink / raw)
To: Jonathan Nieder; +Cc: Stefan Beller, juergen.vogl, git
Jonathan Nieder <jrnieder@gmail.com> writes:
> Stefan Beller wrote:
>
>> Maybe for now we can do with just an update of the documentation/bugs
>> section and say we cannot move files in and out of submodules?
>
> I think we have some existing logic to prevent "git add"-ing a file
> within a submodule to the superproject, for example.
There is die_path_inside_submodule() that sanity-checks the pathspec
and rejects. But I think that is done primarily to give an error
message and not strictly necesary for correctness.
The real safety of "git add" is its call to dir.c::fill_directory();
it collects untracked paths that match the pathspec so that they can
be added as new paths, but because it won't cross the module
boundary, you won't get such a path in the index to begin with.
> So "git mv" should learn the same trick. And perhaps the trick needs
> to be moved down a layer (e.g. into the index API). Hints?
You would want to be able to remove a submodule and replace it with
a directory, but you can probably do it in two steps, i.e.
git reset --hard
git rm --cached sha1collisiondetection
echo Now a regular dir >sha1collisiondetection/READ.ME
find sha1collisiondetection ! -type d -print0 |
git update-index --add --stdin -z
So from that point of view, forbidding (starting from the same state
of our project) this sequence:
git reset --hard
echo Now a regular dir >sha1collisiondetection/READ.ME
find sha1collisiondetection ! -type d -print0 |
git update-index --add --remove --stdin -z
that would nuke the submodule and replace it with a directory within
which there are files would be OK. Making the latter's default
rejection overridable with ADD_CACHE_OK_TO_REPLACE would also be
fine.
^ permalink raw reply [relevance 5%]
* Re: Bug with "git mv" and submodules, and with "git submodule add something --force"
2018-10-19 20:40 7% ` Stefan Beller
@ 2018-10-19 20:58 7% ` Jonathan Nieder
2018-10-22 2:52 5% ` Junio C Hamano
0 siblings, 1 reply; 200+ results
From: Jonathan Nieder @ 2018-10-19 20:58 UTC (permalink / raw)
To: Stefan Beller; +Cc: juergen.vogl, git
Stefan Beller wrote:
> Maybe for now we can do with just an update of the documentation/bugs
> section and say we cannot move files in and out of submodules?
I think we have some existing logic to prevent "git add"-ing a file
within a submodule to the superproject, for example.
So "git mv" should learn the same trick. And perhaps the trick needs
to be moved down a layer (e.g. into the index API). Hints?
Thanks,
Jonathan
^ permalink raw reply [relevance 7%]
* Re: Bug with "git mv" and submodules, and with "git submodule add something --force"
@ 2018-10-19 20:40 7% ` Stefan Beller
2018-10-19 20:58 7% ` Jonathan Nieder
0 siblings, 1 reply; 200+ results
From: Stefan Beller @ 2018-10-19 20:40 UTC (permalink / raw)
To: juergen.vogl; +Cc: git
On Fri, Oct 19, 2018 at 5:43 AM Juergen Vogl <juergen.vogl@jku.at> wrote:
>
> Hi there,
>
> tested on both, git 2.18 and git 2.19.1:
>
> moving a file with `git mv` from a project to a submodule results in an
> **undefined state** of the local repository.
Luckily we do have a submodule in git.git itself, so we can
try it out here as well (I'll use a separate worktree to not
hose my main repo):
git worktree add ../testgit && cd ../testgit
git checkout v2.19.1 && make install
git --version
git version 2.19.1
git submodule update --init
Cloning into '/home/sbeller/testgit/sha1collisiondetection'...
Submodule path 'sha1collisiondetection': checked out
'232357eb2ea0397388254a4b188333a227bf5b10'
git mv cache.h sha1collisiondetection/
git status
HEAD detached at v2.19.1
Changes to be committed:
(use "git reset HEAD <file>..." to unstage)
deleted: sha1collisiondetection
renamed: cache.h -> sha1collisiondetection/cache.h
Untracked files:
(use "git add <file>..." to include in what will be committed)
sha1collisiondetection/.gitignore
sha1collisiondetection/.travis.yml
sha1collisiondetection/LICENSE.txt
sha1collisiondetection/Makefile
sha1collisiondetection/README.md
sha1collisiondetection/lib/
sha1collisiondetection/src/
sha1collisiondetection/test/
sha1collisiondetection/vs2015/
> It breaks up the submodule (it's still in .gitmodules, but not
> accessable via `git submodule`), and is not reversible on local repository.
This seems like what I just did. However reversing can be done via:
git checkout -f
git status
HEAD detached at v2.19.1
nothing to commit, working tree clean
git submodule status
232357eb2ea0397388254a4b188333a227bf5b10 sha1collisiondetection
(stable-v1.0.3-31-g232357e)
So I think it's just "git-mv" that doesn't respect submodule
boundaries, which we would want to address.
The man page of git mv
(https://git-scm.com/docs/git-mv)
actually has a short note about submodules, but that is about
moving *a* submodule not about moving things in and out.
Maybe for now we can do with just an update of the documentation/bugs
section and say we cannot move files in and out of submodules?
>
> Either `git mv submodule/file .`
which is just running the command in reverse, (i.e. swapping
destination and target),
git mv cache.h sha1collisiondetection/
git mv sha1collisiondetection/cache.h cache.h
git status
HEAD detached at v2.19.1
Changes to be committed:
(use "git reset HEAD <file>..." to unstage)
deleted: sha1collisiondetection
Untracked files:
(use "git add <file>..." to include in what will be committed)
sha1collisiondetection/
which seems like the submodule is gone,
> nor deleting the folder works. For the
> locale repo the submodule is gone.
Yup, that seems to be the case.
> But: trying to add it with `git
> submodule add` also do not work and results in an error message (with
> and without `--force` flag):
Would checkout out a state where the submodule still exists
(such as HEAD) and then running "git submodule update --init"
fix it instead?
>
> $ git submodule add git@github.com:-----------/wiki-public.git public
> --force
> A git directory for 'public' is found locally with remote(s):
> origin git@github.com:-----------/wiki-public.git
> If you want to reuse this local git directory instead of cloning again from
> git@github.com:-----------/wiki-public.git
> use the '--force' option. If the local git directory is not the correct repo
> or you are unsure what this means choose another name with the '--name'
> option.
>
> Therefore, it's in a undefined, broken state.
>
>
> Another bug I've got by testing upper line:
> * --force will be used as folder name * when used in `git submodule add
> git@github.com:someone/some.git --force`:
Yes, the order of arguments is important, the options
(such as --force) goes before the URL and path.
> /usr/libexec/git-core/git-submodule: line 273: cd: --: invalid option
> cd: usage: cd [-L|-P] [dir]
> Unable to checkout submodule '--force'
Gah! We'd need to escape the path after the options,
i.e. cd -- --force
would cd into that directory, but I am unsure if that
is accepted by all shells.
Thanks,
Stefan
^ permalink raw reply [relevance 7%]
* Re: [PATCH 19/19] submodule: don't add submodule as odb for push
2018-10-16 23:35 14% ` [PATCH 19/19] submodule: don't add submodule as odb for push Stefan Beller
@ 2018-10-19 20:39 2% ` Jonathan Tan
0 siblings, 0 replies; 200+ results
From: Jonathan Tan @ 2018-10-19 20:39 UTC (permalink / raw)
To: sbeller; +Cc: git, jonathantanmy
> In push_submodule(), because we do not actually need access to objects
> in the submodule, do not invoke add_submodule_odb().
> (for_each_remote_ref_submodule() does not require access to those
> objects, and the actual push is done by spawning another process,
> which handles object access by itself.)
The code looks good - my analysis is the same as that in my review of
the previous version [1].
Can you mention, in the commit message, the tests that exercise the
functionality here (and say that they still pass)?
[1] https://public-inbox.org/git/20181011230028.200488-1-jonathantanmy@google.com/
>
> Signed-off-by: Stefan Beller <sbeller@google.com>
> ---
> submodule.c | 3 ---
> 1 file changed, 3 deletions(-)
>
> diff --git a/submodule.c b/submodule.c
> index 7305ae2e10..e623e6bf7f 100644
> --- a/submodule.c
> +++ b/submodule.c
> @@ -1024,9 +1024,6 @@ static int push_submodule(const char *path,
> const struct string_list *push_options,
> int dry_run)
> {
> - if (add_submodule_odb(path))
> - return 1;
> -
> if (for_each_remote_ref_submodule(path, has_remote, NULL) > 0) {
> struct child_process cp = CHILD_PROCESS_INIT;
> argv_array_push(&cp.args, "push");
> --
> 2.19.0
>
^ permalink raw reply [relevance 2%]
* Re: [PATCH] submodule.c: remove some of the_repository references
@ 2018-10-19 17:46 7% ` Stefan Beller
0 siblings, 0 replies; 200+ results
From: Stefan Beller @ 2018-10-19 17:46 UTC (permalink / raw)
To: Duy Nguyen; +Cc: git, Jeff King, Junio C Hamano
On Fri, Oct 19, 2018 at 10:34 AM Nguyễn Thái Ngọc Duy <pclouds@gmail.com> wrote:
>
>
> struct collect_changed_submodules_cb_data {
> + struct repository *repo;
This slightly overlaps with sb/submodule-recursive-fetch-gets-the-tip,
but we can have this patch on its own as I have to rebuild that
series, will build on top of this.
The patch looks good,
Stefan
^ permalink raw reply [relevance 7%]
* Re: [PATCH/WIP 00/19] Kill the_index, final part
@ 2018-10-19 17:38 4% ` Stefan Beller
0 siblings, 0 replies; 200+ results
From: Stefan Beller @ 2018-10-19 17:38 UTC (permalink / raw)
To: Duy Nguyen; +Cc: git
On Fri, Oct 19, 2018 at 7:52 AM Nguyễn Thái Ngọc Duy <pclouds@gmail.com> wrote:
>
> Stefan I see you start removing more references of the_repository in
> another series (haven't really looked at it yet) so this is just heads
> up so we could coordinate if needed. Still WIP, not really ready for
> review yet.
I'll take a brief look anyway, otherwise you wouldn't have put in on
a mailing list :-P
Yes, coordination would be good; stolee brought up a good point
regarding testing these changes. Killing of the index in itself can be
just tested by checking if behavior stays the same, but killing of
the_repository is a slightly different beast, as it relates to submodules,
which means any global state that is still around, need to go into
the repository struct (we might have missed some there, whereas
the index struct has been around for a long time).
> This series removes use of the_index outside builtin/ and t/helper/.
> The only the_index reference left is in repository.c. The macro
> NO_THE_REPOSITORY_COMPATIBILITY_MACROS is now flipped to
> USE_THE_INDEX_COMPATIBILITY_MACROS. "the_index" is forbidden by
> default.
Wow, that is really cool. Looking forward for the patches. :-)
> After this I think we can start pushing the_repository outside library
> code. Then perhaps clean them up in builtin code too and you can
> finally get rid of it. But I don't think that'll happen in a year's
> time.
That sounds realistic.
Stefan
^ permalink raw reply [relevance 4%]
* Re: [PATCH v2 19/24] submodule.c: remove implicit dependency on the_index
@ 2018-10-19 16:57 4% ` Stefan Beller
0 siblings, 0 replies; 200+ results
From: Stefan Beller @ 2018-10-19 16:57 UTC (permalink / raw)
To: Jeff King; +Cc: Duy Nguyen, git
> >
> > Yeah. Killing the_index is just the first small step (didn't look that
> > small when I started). Now it's all about the_repository ;-) and we
> > have like 400 references of it just in library code.
>
> I suspect it is much worse than that, even. There are many spots that
> likely are relying on global data that _should_ be in the repository
> struct but aren't yet. I don't think there's even an easy way to count
> those. ;)
This is a very interesting part of the discussion,
please note the series "RFC Bring the_repository into cmd_foo"[1]
as that proposes one way how to deal with this, exposing the
repository/index changes into the existing test suite.
[1] https://public-inbox.org/git/20181018183758.81186-1-sbeller@google.com/
I'll read on.
^ permalink raw reply [relevance 4%]
* Re: [PATCH 4/9] submodule.c: move global changed_submodule_names into fetch submodule struct
@ 2018-10-18 19:09 5% ` Stefan Beller
0 siblings, 0 replies; 200+ results
From: Stefan Beller @ 2018-10-18 19:09 UTC (permalink / raw)
To: Jonathan Tan; +Cc: Junio C Hamano, git
On Wed, Oct 17, 2018 at 2:26 PM Jonathan Tan <jonathantanmy@google.com> wrote:
>
> > The `changed_submodule_names` are only used for fetching, so let's make it
> > part of the struct that is passed around for fetching submodules.
>
> Keep the titles of commit messages to 50 characters or under.
renamed
>
> > +static void calculate_changed_submodule_paths(
> > + struct submodule_parallel_fetch *spf)
>
> Instead of taking the entire struct, could this just take the list of
> changed submodule names instead?
I think so, done.
^ permalink raw reply [relevance 5%]
* [RFC PATCH 1/2] repository: have get_the_repository() to remove the_repository dependency
@ 2018-10-18 18:37 7% ` Stefan Beller
0 siblings, 0 replies; 200+ results
From: Stefan Beller @ 2018-10-18 18:37 UTC (permalink / raw)
To: sbeller; +Cc: git, jonathantanmy, stolee
The struct 'the_repo' contains all data that of the main repository.
As we move more and more globals into this struct, the usual way of
accessing these is using 'the_repository', which can be used as a drop in
replacement for accessing the migrated globals.
However during the migration of globals into the repository object, it is
not clear if some code path rely on the_repository or can work on an
arbitrary repository (as we'd eventually want for submodules) due to the
excessive use of the_repository throughout the code base.
To address this, introduce a function 'get_the_repository(void)' which
will return the main repository and set the_repository to NULL when the
environment variable GIT_NO_THE_REPOSITORY is set.
This function is to be strictly used only at the beginning of builtin
command to assign it to a local repository pointer that we'll use to
pass through the code base.
By having the possibility to set the_repository to NULL, we'll get
a segfault when we try to access the_repository instead of using the
handle that we pass around.
This approach let's us have the_repository in the setup code, which
in its current form is not yet able to transition into a world where
the repository handle is passed around and only test the passing around
of the repository handle for later stage code.
Eventually in the future the setup code will produce the repository
handle and each 'cmd_foo(int argc, char **argv)' builtin would get the
repository via an additional parameter.
Signed-off-by: Stefan Beller <sbeller@google.com>
---
repository.c | 10 ++++++++++
repository.h | 13 ++++++++++++-
2 files changed, 22 insertions(+), 1 deletion(-)
diff --git a/repository.c b/repository.c
index 5dd1486718..aaaababb88 100644
--- a/repository.c
+++ b/repository.c
@@ -20,6 +20,16 @@ void initialize_the_repository(void)
repo_set_hash_algo(&the_repo, GIT_HASH_SHA1);
}
+struct repository *get_the_repository(void)
+{
+ struct repository *r = the_repository;
+
+ if (getenv("GIT_NO_THE_REPOSITORY"))
+ the_repository = NULL;
+
+ return r;
+}
+
static void expand_base_dir(char **out, const char *in,
const char *base_dir, const char *def_in)
{
diff --git a/repository.h b/repository.h
index 9f16c42c1e..26f5d64f68 100644
--- a/repository.h
+++ b/repository.h
@@ -114,13 +114,24 @@ void repo_set_gitdir(struct repository *repo, const char *root,
const struct set_gitdir_args *extra_args);
void repo_set_worktree(struct repository *repo, const char *path);
void repo_set_hash_algo(struct repository *repo, int algo);
-void initialize_the_repository(void);
int repo_init(struct repository *r, const char *gitdir, const char *worktree);
int repo_submodule_init(struct repository *submodule,
struct repository *superproject,
const char *path);
void repo_clear(struct repository *repo);
+/*
+ * Initializes the repository 'the_repository', which is used in the transition
+ * phase of moving globals into the repository struct.
+ */
+void initialize_the_repository(void);
+
+/*
+ * To be called once; after the call use only returned repository, and do not
+ * use the_repository any more
+ */
+struct repository *get_the_repository(void);
+
/*
* Populates the repository's index from its index_file, an index struct will
* be allocated if needed.
--
2.19.0
^ permalink raw reply related [relevance 7%]
* Re: [PATCH 0/9] Resending sb/submodule-recursive-fetch-gets-the-tip
2018-10-18 7:30 14% ` Junio C Hamano
@ 2018-10-18 18:00 7% ` Stefan Beller
0 siblings, 0 replies; 200+ results
From: Stefan Beller @ 2018-10-18 18:00 UTC (permalink / raw)
To: Junio C Hamano; +Cc: git
On Thu, Oct 18, 2018 at 12:30 AM Junio C Hamano <gitster@pobox.com> wrote:
>
> Stefan Beller <sbeller@google.com> writes:
>
> > This is based on ao/submodule-wo-gitmodules-checked-out.
> >
> > This resends origin/sb/submodule-recursive-fetch-gets-the-tip, resolving
> > the issues pointed out via origin/xxx/sb-submodule-recursive-fetch-gets-the-tip-in-pu
> > by basing this series on origin/ao/submodule-wo-gitmodules-checked-out
>
> Applying this round to the result of merging ao/submodule-* to
> 'master' requires this to work, it seems, as you've introduced a
> call to repo-init thing in the meantime with another topic.
Unfortunately, yes, sb/submodule-update-in-c had one such call.
> > This is based on ao/submodule-wo-gitmodules-checked-out.
> Thanks. I had an impression that we were not entirely happy with
> the topic and are expecting a reroll, but let's hope that the part
> we expect to be updated won't have much overlaps.
I slowly came to the realization that this topic might be ripped up
into 2 or more topics, as some of the cleanups seem to be orthogonal.
^ permalink raw reply [relevance 7%]
* Re: [PATCH 0/9] Resending sb/submodule-recursive-fetch-gets-the-tip
2018-10-16 18:13 9% [PATCH 0/9] Resending sb/submodule-recursive-fetch-gets-the-tip Stefan Beller
` (8 preceding siblings ...)
2018-10-18 2:30 4% ` [PATCH 0/9] Resending sb/submodule-recursive-fetch-gets-the-tip Junio C Hamano
@ 2018-10-18 7:30 14% ` Junio C Hamano
2018-10-18 18:00 7% ` Stefan Beller
9 siblings, 1 reply; 200+ results
From: Junio C Hamano @ 2018-10-18 7:30 UTC (permalink / raw)
To: Stefan Beller; +Cc: git
Stefan Beller <sbeller@google.com> writes:
> This is based on ao/submodule-wo-gitmodules-checked-out.
>
> This resends origin/sb/submodule-recursive-fetch-gets-the-tip, resolving
> the issues pointed out via origin/xxx/sb-submodule-recursive-fetch-gets-the-tip-in-pu
> by basing this series on origin/ao/submodule-wo-gitmodules-checked-out
Applying this round to the result of merging ao/submodule-* to
'master' requires this to work, it seems, as you've introduced a
call to repo-init thing in the meantime with another topic.
Subject: [PATCH] fixup! repository: repo_submodule_init to take a submodule struct
---
builtin/submodule--helper.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/builtin/submodule--helper.c b/builtin/submodule--helper.c
index 5f8a804a6e..015aa1471f 100644
--- a/builtin/submodule--helper.c
+++ b/builtin/submodule--helper.c
@@ -2037,7 +2037,7 @@ static int ensure_core_worktree(int argc, const char **argv, const char *prefix)
if (!sub)
BUG("We could get the submodule handle before?");
- if (repo_submodule_init(&subrepo, the_repository, path))
+ if (repo_submodule_init(&subrepo, the_repository, sub))
die(_("could not get a repository handle for submodule '%s'"), path);
if (!repo_config_get_string(&subrepo, "core.worktree", &cw)) {
--
2.19.1-450-ga4b8ab5363
^ permalink raw reply related [relevance 14%]
* Re: [PATCH 0/9] Resending sb/submodule-recursive-fetch-gets-the-tip
2018-10-16 18:13 9% [PATCH 0/9] Resending sb/submodule-recursive-fetch-gets-the-tip Stefan Beller
` (7 preceding siblings ...)
2018-10-16 18:13 21% ` [PATCH 9/9] builtin/fetch: check for submodule updates for non branch fetches Stefan Beller
@ 2018-10-18 2:30 4% ` Junio C Hamano
2018-10-18 7:30 14% ` Junio C Hamano
9 siblings, 0 replies; 200+ results
From: Junio C Hamano @ 2018-10-18 2:30 UTC (permalink / raw)
To: Stefan Beller; +Cc: git
Stefan Beller <sbeller@google.com> writes:
> This is based on ao/submodule-wo-gitmodules-checked-out.
Thanks. I had an impression that we were not entirely happy with
the topic and are expecting a reroll, but let's hope that the part
we expect to be updated won't have much overlaps.
> A range-diff below shows how picking a different base changed the patches,
> apart from that no further adjustments have been made.
Thanks. Let's see how well this plays together with other topics in 'pu'.q
^ permalink raw reply [relevance 4%]
* Re: [PATCH 00/19] Bring more repository handles into our code base
2018-10-17 12:41 4% ` [PATCH 00/19] Bring more repository handles into our code base Derrick Stolee
@ 2018-10-17 17:53 5% ` Stefan Beller
0 siblings, 1 reply; 200+ results
From: Stefan Beller @ 2018-10-17 17:53 UTC (permalink / raw)
To: Derrick Stolee; +Cc: git, Jonathan Tan
On Wed, Oct 17, 2018 at 5:41 AM Derrick Stolee <stolee@gmail.com> wrote:
>
> On 10/16/2018 7:35 PM, Stefan Beller wrote:
> >
> > This series takes another approach as it doesn't change the signature of
> > functions, but introduces new functions that can deal with arbitrary
> > repositories, keeping the old function signature around using a shallow wrapper.
> I think this is a good direction, and the changes look good to me.
>
> > Additionally each patch adds a semantic patch, that would port from the old to
> > the new function. These semantic patches are all applied in the very last patch,
> > but we could omit applying the last patch if it causes too many merge conflicts
> > and trickl in the semantic patches over time when there are no merge conflicts.
>
> The semantic patches are a good idea. At some point in the future, we
> can submit a series that applies the patches to any leftover calls and
> removes the old callers. We can hopefully rely on review (and the
> semantic patches warning that there is work to do) to prevent new
> callers from being introduced in future topics. That doesn't count
> topics that come around while this one isn't merged down.
For those topics still in flight, I added re-defines, e.g.
#ifndef NO_THE_REPOSITORY_COMPATIBILITY_MACROS
#define get_merge_bases(r1, r2)
repo_get_merge_bases(the_repository, r1, r2)
#endif
so the base function still keeps working, and we can cleanup
multiple times, until eventually, we can get rid of the base function.
> I had one high-level question: How are we testing that these "arbitrary
> repository" changes are safe?
I did the bare minimum in conversions in this series, such that the
submodule code tests successfully. So if we'd revert some parts,
the submodule tests would break.
> I just remember the issue we had with the
> commit-graph code relying on arbitrary repositories, but then adding a
> reference to the replace objects broke the code (because replace-objects
> wasn't using arbitrary repos correctly, but the commit-graph was tested
> with arbitrary repositories using "test-tool repository"). It would be
> nice to introduce more method calls in t/helper/test-repository.c that
> help us know these are safe conversions.
Or instead we could accelerate the long term plan of removing a
hard coded the_repository and have each cmd builtin take an additional
repository pointer from the init code, such that we'd bring all of Git to
work on arbitrary repositories. Then the standard test suite should be
okay, as there is no special case for the_repository any more.
> Otherwise, we are essentially
> waiting until we try to take submodule things in-process and find out
> what breaks. As we discovered with the refstore, we can't just ensure
> that all references to the_repository are removed.
Yes, that is correct. We had a similar case with partial clone,
as laid out in the cover letter for the RFC.
I'll explore both the test tool approach as well as
repository-fication of the code base.
Thanks,
Stefan
^ permalink raw reply [relevance 5%]
* Re: [PATCH 00/19] Bring more repository handles into our code base
2018-10-16 23:35 7% [PATCH 00/19] Bring more repository handles into our code base Stefan Beller
2018-10-16 23:35 20% ` [PATCH 18/19] submodule: use submodule repos for object lookup Stefan Beller
2018-10-16 23:35 14% ` [PATCH 19/19] submodule: don't add submodule as odb for push Stefan Beller
@ 2018-10-17 12:41 4% ` Derrick Stolee
2018-10-17 17:53 5% ` Stefan Beller
3 siblings, 1 reply; 200+ results
From: Derrick Stolee @ 2018-10-17 12:41 UTC (permalink / raw)
To: Stefan Beller, git; +Cc: jonathantanmy
On 10/16/2018 7:35 PM, Stefan Beller wrote:
>
> This series takes another approach as it doesn't change the signature of
> functions, but introduces new functions that can deal with arbitrary
> repositories, keeping the old function signature around using a shallow wrapper.
I think this is a good direction, and the changes look good to me.
> Additionally each patch adds a semantic patch, that would port from the old to
> the new function. These semantic patches are all applied in the very last patch,
> but we could omit applying the last patch if it causes too many merge conflicts
> and trickl in the semantic patches over time when there are no merge conflicts.
The semantic patches are a good idea. At some point in the future, we
can submit a series that applies the patches to any leftover calls and
removes the old callers. We can hopefully rely on review (and the
semantic patches warning that there is work to do) to prevent new
callers from being introduced in future topics. That doesn't count
topics that come around while this one isn't merged down.
I had one high-level question: How are we testing that these "arbitrary
repository" changes are safe? I just remember the issue we had with the
commit-graph code relying on arbitrary repositories, but then adding a
reference to the replace objects broke the code (because replace-objects
wasn't using arbitrary repos correctly, but the commit-graph was tested
with arbitrary repositories using "test-tool repository"). It would be
nice to introduce more method calls in t/helper/test-repository.c that
help us know these are safe conversions. Otherwise, we are essentially
waiting until we try to take submodule things in-process and find out
what breaks. As we discovered with the refstore, we can't just ensure
that all references to the_repository are removed.
Thanks,
-Stolee
^ permalink raw reply [relevance 4%]
* Re: [PATCH] builtin/submodule--helper: remove debugging leftover tracing
2018-10-16 23:45 17% [PATCH] builtin/submodule--helper: remove debugging leftover tracing Stefan Beller
@ 2018-10-17 2:52 4% ` Jonathan Nieder
0 siblings, 0 replies; 200+ results
From: Jonathan Nieder @ 2018-10-17 2:52 UTC (permalink / raw)
To: Stefan Beller; +Cc: gitster, git
Stefan Beller wrote:
> I noticed 74d4731da1 (submodule--helper: replace connect-gitdir-workingtree
> by ensure-core-worktree, 2018-08-13) had two leftover debugging statements
> when reading The coverage report [1]. Remove them.
>
> https://public-inbox.org/git/e30a9c05-87d8-1f2b-182c-6d6a5fefe43c@gmail.com/
>
> Signed-off-by: Stefan Beller <sbeller@google.com>
> ---
> builtin/submodule--helper.c | 2 --
> 1 file changed, 2 deletions(-)
Doh. Glad you caught it!
Is there some reference for The Coverage Report other than the mailing
list? E.g. I suspect a reference to
make coverage-test
make coverage-report
would be useful to readers finding this commit later.
> To be applied on (or squashed into the tip of)
> sb/submodule-update-in-c
Looks like that's already in "master", so not a candidate for
squashing.
Reviewed-by: Jonathan Nieder <jrnieder@gmail.com>
^ permalink raw reply [relevance 4%]
* [PATCH] builtin/submodule--helper: remove debugging leftover tracing
@ 2018-10-16 23:45 17% Stefan Beller
2018-10-17 2:52 4% ` Jonathan Nieder
0 siblings, 1 reply; 200+ results
From: Stefan Beller @ 2018-10-16 23:45 UTC (permalink / raw)
To: gitster; +Cc: git, Stefan Beller
I noticed 74d4731da1 (submodule--helper: replace connect-gitdir-workingtree
by ensure-core-worktree, 2018-08-13) had two leftover debugging statements
when reading The coverage report [1]. Remove them.
https://public-inbox.org/git/e30a9c05-87d8-1f2b-182c-6d6a5fefe43c@gmail.com/
Signed-off-by: Stefan Beller <sbeller@google.com>
---
To be applied on (or squashed into the tip of)
sb/submodule-update-in-c
builtin/submodule--helper.c | 2 --
1 file changed, 2 deletions(-)
diff --git a/builtin/submodule--helper.c b/builtin/submodule--helper.c
index 5c9d1fb496..c7d3841ffc 100644
--- a/builtin/submodule--helper.c
+++ b/builtin/submodule--helper.c
@@ -1459,7 +1459,6 @@ static void determine_submodule_update_strategy(struct repository *r,
key = xstrfmt("submodule.%s.update", sub->name);
if (update) {
- trace_printf("parsing update");
if (parse_submodule_update_strategy(update, out) < 0)
die(_("Invalid update mode '%s' for submodule path '%s'"),
update, path);
@@ -1468,7 +1467,6 @@ static void determine_submodule_update_strategy(struct repository *r,
die(_("Invalid update mode '%s' configured for submodule path '%s'"),
val, path);
} else if (sub->update_strategy.type != SM_UPDATE_UNSPECIFIED) {
- trace_printf("loaded thing");
out->type = sub->update_strategy.type;
out->command = sub->update_strategy.command;
} else
--
2.19.0
^ permalink raw reply related [relevance 17%]
* [PATCH 19/19] submodule: don't add submodule as odb for push
2018-10-16 23:35 7% [PATCH 00/19] Bring more repository handles into our code base Stefan Beller
2018-10-16 23:35 20% ` [PATCH 18/19] submodule: use submodule repos for object lookup Stefan Beller
@ 2018-10-16 23:35 14% ` Stefan Beller
2018-10-19 20:39 2% ` Jonathan Tan
2018-10-17 12:41 4% ` [PATCH 00/19] Bring more repository handles into our code base Derrick Stolee
3 siblings, 1 reply; 200+ results
From: Stefan Beller @ 2018-10-16 23:35 UTC (permalink / raw)
To: git; +Cc: jonathantanmy, Stefan Beller
In push_submodule(), because we do not actually need access to objects
in the submodule, do not invoke add_submodule_odb().
(for_each_remote_ref_submodule() does not require access to those
objects, and the actual push is done by spawning another process,
which handles object access by itself.)
Signed-off-by: Stefan Beller <sbeller@google.com>
---
submodule.c | 3 ---
1 file changed, 3 deletions(-)
diff --git a/submodule.c b/submodule.c
index 7305ae2e10..e623e6bf7f 100644
--- a/submodule.c
+++ b/submodule.c
@@ -1024,9 +1024,6 @@ static int push_submodule(const char *path,
const struct string_list *push_options,
int dry_run)
{
- if (add_submodule_odb(path))
- return 1;
-
if (for_each_remote_ref_submodule(path, has_remote, NULL) > 0) {
struct child_process cp = CHILD_PROCESS_INIT;
argv_array_push(&cp.args, "push");
--
2.19.0
^ permalink raw reply related [relevance 14%]
* [PATCH 18/19] submodule: use submodule repos for object lookup
2018-10-16 23:35 7% [PATCH 00/19] Bring more repository handles into our code base Stefan Beller
@ 2018-10-16 23:35 20% ` Stefan Beller
2018-10-25 9:14 8% ` SZEDER Gábor
2018-10-16 23:35 14% ` [PATCH 19/19] submodule: don't add submodule as odb for push Stefan Beller
` (2 subsequent siblings)
3 siblings, 1 reply; 200+ results
From: Stefan Beller @ 2018-10-16 23:35 UTC (permalink / raw)
To: git; +Cc: jonathantanmy, Stefan Beller
This converts the 'show_submodule_header' function to use
the repository API properly, such that the submodule objects
are not added to the main object store.
Signed-off-by: Stefan Beller <sbeller@google.com>
---
submodule.c | 75 ++++++++++++++++++++++++++++++++++++++++++-----------
1 file changed, 60 insertions(+), 15 deletions(-)
diff --git a/submodule.c b/submodule.c
index 4ee69cc014..7305ae2e10 100644
--- a/submodule.c
+++ b/submodule.c
@@ -444,7 +444,7 @@ static int prepare_submodule_summary(struct rev_info *rev, const char *path,
return prepare_revision_walk(rev);
}
-static void print_submodule_summary(struct rev_info *rev, struct diff_options *o)
+static void print_submodule_summary(struct repository *r, struct rev_info *rev, struct diff_options *o)
{
static const char format[] = " %m %s";
struct strbuf sb = STRBUF_INIT;
@@ -455,7 +455,8 @@ static void print_submodule_summary(struct rev_info *rev, struct diff_options *o
ctx.date_mode = rev->date_mode;
ctx.output_encoding = get_log_output_encoding();
strbuf_setlen(&sb, 0);
- format_commit_message(commit, format, &sb, &ctx);
+ repo_format_commit_message(r, commit, format, &sb,
+ &ctx);
strbuf_addch(&sb, '\n');
if (commit->object.flags & SYMMETRIC_LEFT)
diff_emit_submodule_del(o, sb.buf);
@@ -482,14 +483,46 @@ void prepare_submodule_repo_env(struct argv_array *out)
DEFAULT_GIT_DIR_ENVIRONMENT);
}
-/* Helper function to display the submodule header line prior to the full
- * summary output. If it can locate the submodule objects directory it will
- * attempt to lookup both the left and right commits and put them into the
- * left and right pointers.
+/*
+ * Initialize 'out' based on the provided submodule path.
+ *
+ * Unlike repo_submodule_init, this tolerates submodules not present
+ * in .gitmodules. This function exists only to preserve historical behavior,
+ *
+ * Returns 0 on success, -1 when the submodule is not present.
+ */
+static int open_submodule(struct repository *out, const char *path)
+{
+ struct strbuf sb = STRBUF_INIT;
+
+ if (submodule_to_gitdir(&sb, path) || repo_init(out, sb.buf, NULL)) {
+ strbuf_release(&sb);
+ return -1;
+ }
+
+ out->submodule_prefix = xstrdup(path);
+ out->submodule_prefix = xstrfmt("%s%s/",
+ the_repository->submodule_prefix ?
+ the_repository->submodule_prefix :
+ "", path);
+
+ strbuf_release(&sb);
+ return 0;
+}
+
+/*
+ * Helper function to display the submodule header line prior to the full
+ * summary output.
+ *
+ * If it can locate the submodule git directory it will create a repository
+ * handle for the submodule and lookup both the left and right commits and
+ * put them into the left and right pointers.
*/
-static void show_submodule_header(struct diff_options *o, const char *path,
+static void show_submodule_header(struct diff_options *o,
+ const char *path,
struct object_id *one, struct object_id *two,
unsigned dirty_submodule,
+ struct repository *sub,
struct commit **left, struct commit **right,
struct commit_list **merge_bases)
{
@@ -508,7 +541,7 @@ static void show_submodule_header(struct diff_options *o, const char *path,
else if (is_null_oid(two))
message = "(submodule deleted)";
- if (add_submodule_odb(path)) {
+ if (!sub) {
if (!message)
message = "(commits not present)";
goto output_header;
@@ -518,8 +551,8 @@ static void show_submodule_header(struct diff_options *o, const char *path,
* Attempt to lookup the commit references, and determine if this is
* a fast forward or fast backwards update.
*/
- *left = lookup_commit_reference(the_repository, one);
- *right = lookup_commit_reference(the_repository, two);
+ *left = lookup_commit_reference(sub, one);
+ *right = lookup_commit_reference(sub, two);
/*
* Warn about missing commits in the submodule project, but only if
@@ -529,7 +562,7 @@ static void show_submodule_header(struct diff_options *o, const char *path,
(!is_null_oid(two) && !*right))
message = "(commits not present)";
- *merge_bases = get_merge_bases(*left, *right);
+ *merge_bases = repo_get_merge_bases(sub, *left, *right);
if (*merge_bases) {
if ((*merge_bases)->item == *left)
fast_forward = 1;
@@ -563,16 +596,20 @@ void show_submodule_summary(struct diff_options *o, const char *path,
struct rev_info rev;
struct commit *left = NULL, *right = NULL;
struct commit_list *merge_bases = NULL;
+ struct repository subrepo, *sub = &subrepo;
+
+ if (open_submodule(&subrepo, path) < 0)
+ sub = NULL;
show_submodule_header(o, path, one, two, dirty_submodule,
- &left, &right, &merge_bases);
+ sub, &left, &right, &merge_bases);
/*
* If we don't have both a left and a right pointer, there is no
* reason to try and display a summary. The header line should contain
* all the information the user needs.
*/
- if (!left || !right)
+ if (!left || !right || !sub)
goto out;
/* Treat revision walker failure the same as missing commits */
@@ -581,13 +618,15 @@ void show_submodule_summary(struct diff_options *o, const char *path,
goto out;
}
- print_submodule_summary(&rev, o);
+ print_submodule_summary(sub, &rev, o);
out:
if (merge_bases)
free_commit_list(merge_bases);
clear_commit_marks(left, ~0);
clear_commit_marks(right, ~0);
+ if (sub)
+ repo_clear(sub);
}
void show_submodule_inline_diff(struct diff_options *o, const char *path,
@@ -599,9 +638,13 @@ void show_submodule_inline_diff(struct diff_options *o, const char *path,
struct commit_list *merge_bases = NULL;
struct child_process cp = CHILD_PROCESS_INIT;
struct strbuf sb = STRBUF_INIT;
+ struct repository subrepo, *sub = &subrepo;
+
+ if (open_submodule(&subrepo, path) < 0)
+ sub = NULL;
show_submodule_header(o, path, one, two, dirty_submodule,
- &left, &right, &merge_bases);
+ sub, &left, &right, &merge_bases);
/* We need a valid left and right commit to display a difference */
if (!(left || is_null_oid(one)) ||
@@ -662,6 +705,8 @@ void show_submodule_inline_diff(struct diff_options *o, const char *path,
clear_commit_marks(left, ~0);
if (right)
clear_commit_marks(right, ~0);
+ if (sub)
+ repo_clear(sub);
}
int should_update_submodules(void)
--
2.19.0
^ permalink raw reply related [relevance 20%]
* [PATCH 00/19] Bring more repository handles into our code base
@ 2018-10-16 23:35 7% Stefan Beller
2018-10-16 23:35 20% ` [PATCH 18/19] submodule: use submodule repos for object lookup Stefan Beller
` (3 more replies)
0 siblings, 4 replies; 200+ results
From: Stefan Beller @ 2018-10-16 23:35 UTC (permalink / raw)
To: git; +Cc: jonathantanmy, Stefan Beller
This rerolls sb/more-repo-in-api.
It applies on nd/the-index merged with ds/reachable and is available via
git fetch https://github.com/stefanbeller/git object-store-final-3
I addressed all of Jonathans comments, see range-diff below;
the last patch (applying the semantic patches) has been omitted as that
would produce a lot of merge conflicts. Without that patch, this merges
cleanly to next.
As for when to apply the semantic patches, I wondered if we would prefer a dirty merge
(created via "make coccicheck && git apply contrib/coccinelle/the_repository.cocci.patch")
of the semantic patches, or if we'd actually trickle in the changes over some time?
Thanks,
Stefan
An earlier RFC was sent out at
https://public-inbox.org/git/20181011211754.31369-1-sbeller@google.com/ and said:
This applies on nd/the-index (b3c7eef9b05) and is the logical continuation of
the object store series, which I sent over the last year.
The previous series did take a very slow and pedantic approach,
using a #define trick, see cfc62fc98c for details, but it turns out,
that it doesn't work:
When changing the signature of widely used functions, it burdens the
maintainer in resolving the semantic conflicts.
In the orginal approach this was called a feature, as then we can ensure
that not bugs creep into the code base during the merge window (while such
a refactoring series wanders from pu to master). It turns out this
was not well received and was just burdensome.
The #define trick doesn't buy us much to begin with when dealing with
non-merge-conflicts. For example, see deref_tag at tag.c:68, which got
the repository argument in 286d258d4f (tag.c: allow deref_tag to handle
arbitrary repositories, 2018-06-28) but lost its property of working on any
repository while 8c4cc32689 (tag: don't warn if target is missing but
promised, 2018-07-12) was in flight simultaneously.
Another example of failure of this approach is seen in patch 5, which
shows that the pedantry was missed.
This series takes another approach as it doesn't change the signature of
functions, but introduces new functions that can deal with arbitrary
repositories, keeping the old function signature around using a shallow wrapper.
Additionally each patch adds a semantic patch, that would port from the old to
the new function. These semantic patches are all applied in the very last patch,
but we could omit applying the last patch if it causes too many merge conflicts
and trickl in the semantic patches over time when there are no merge conflicts.
The original goal of all these refactoring series was to remove add_submodule_odb
in submodule.c, which was partially reached with this series. I'll investigate the
remaining calls in another series, but it shows we're close to be done with these
large refactorings as far as I am concerned.
Thanks,
Stefan
Stefan Beller (19):
sha1_file: allow read_object to read objects in arbitrary repositories
packfile: allow has_packed_and_bad to handle arbitrary repositories
object-store: allow read_object_file_extended to read from arbitrary
repositories
object-store: prepare read_object_file to deal with arbitrary
repositories
object-store: prepare has_{sha1, object}_file[_with_flags] to handle
arbitrary repositories
object: parse_object to honor its repository argument
commit: allow parse_commit* to handle arbitrary repositories
commit-reach.c: allow paint_down_to_common to handle arbitrary
repositories
commit-reach.c: allow merge_bases_many to handle arbitrary
repositories
commit-reach.c: allow remove_redundant to handle arbitrary
repositories
commit-reach.c: allow get_merge_bases_many_0 to handle arbitrary
repositories
commit-reach: prepare get_merge_bases to handle arbitrary repositories
commit-reach: prepare in_merge_bases[_many] to handle arbitrary
repositories
commit: prepare get_commit_buffer to handle arbitrary repositories
commit: prepare repo_unuse_commit_buffer to handle arbitrary
repositories
commit: prepare logmsg_reencode to handle arbitrary repositories
pretty: prepare format_commit_message to handle arbitrary repositories
submodule: use submodule repos for object lookup
submodule: don't add submodule as odb for push
commit-reach.c | 73 +++++++-----
commit-reach.h | 38 +++++--
commit.c | 32 ++++--
commit.h | 39 ++++++-
contrib/coccinelle/the_repository.cocci | 144 ++++++++++++++++++++++++
object-store.h | 35 ++++--
object.c | 6 +-
packfile.c | 5 +-
packfile.h | 2 +-
pretty.c | 28 ++---
pretty.h | 7 +-
sha1-file.c | 34 +++---
streaming.c | 2 +-
submodule.c | 78 ++++++++++---
14 files changed, 402 insertions(+), 121 deletions(-)
create mode 100644 contrib/coccinelle/the_repository.cocci
git range-diff origin/sb/more-repo-in-api...
-: ---------- > 1: 5227c38566 commit-reach: move walk methods from commit.c
-: ---------- > 2: 6404355657 commit.h: remove method declarations
-: ---------- > 3: 1d614d41e5 commit-reach: move ref_newer from remote.c
-: ---------- > 4: 920f93ca1c commit-reach: move commit_contains from ref-filter
-: ---------- > 5: f044bb49ad upload-pack: make reachable() more generic
-: ---------- > 6: 921bf7734f upload-pack: refactor ok_to_give_up()
-: ---------- > 7: 118be5785e upload-pack: generalize commit date cutoff
-: ---------- > 8: ba3ca1edce commit-reach: move can_all_from_reach_with_flags
-: ---------- > 9: ab176ac4ae test-reach: create new test tool for ref_newer
-: ---------- > 10: 5cd52de326 test-reach: test in_merge_bases
-: ---------- > 11: 6255232ec1 test-reach: test is_descendant_of
-: ---------- > 12: 324dec0191 test-reach: test get_merge_bases_many
-: ---------- > 13: 0c89f715d0 test-reach: test reduce_heads
-: ---------- > 14: 1792bc1250 test-reach: test can_all_from_reach_with_flags
-: ---------- > 15: 1fee124257 test-reach: test commit_contains
-: ---------- > 16: 1e3497a24c commit-reach: replace ref_newer logic
-: ---------- > 17: 4fbcca4eff commit-reach: make can_all_from_reach... linear
-: ---------- > 18: 6cc017431c commit-reach: use can_all_from_reach
-: ---------- > 19: 6621c83874 commit-reach: correct accidental #include of C file
-: ---------- > 20: b67f6b26e3 commit-reach: properly peel tags
-: ---------- > 21: 4067a64672 commit-reach: fix memory and flag leaks
1: 2c13c6101f = 22: 14c12a5350 sha1_file: allow read_object to read objects in arbitrary repositories
2: 82848c2571 = 23: f978950604 packfile: allow has_packed_and_bad to handle arbitrary repositories
3: 717023d801 = 24: 852f6696d1 object-store: allow read_object_file_extended to read from arbitrary repositories
4: fab6bdfad0 ! 25: b185b88ad8 object-store: prepare read_object_file to deal with arbitrary repositories
@@ -9,7 +9,7 @@
Introduce repo_read_object_file which takes the repository argument, and
hide the original read_object_file as a macro behind
- NO_THE_REPOSITORY_COMPATIBILITY_MACROS, which we planned for in
+ NO_THE_REPOSITORY_COMPATIBILITY_MACROS, similar to
e675765235 (diff.c: remove implicit dependency on the_index, 2018-09-21)
Add a coccinelle patch to convert existing callers, but do not apply
-: ---------- > 26: 009d6f350b object-store: prepare has_{sha1, object}_file[_with_flags] to handle arbitrary repositories
5: 9e34d76c53 ! 27: df2feb2baf object: parse_object to honor its repository argument
@@ -12,6 +12,17 @@
diff --git a/object.c b/object.c
--- a/object.c
+++ b/object.c
+@@
+ if (obj && obj->parsed)
+ return obj;
+
+- if ((obj && obj->type == OBJ_BLOB && has_object_file(oid)) ||
+- (!obj && has_object_file(oid) &&
++ if ((obj && obj->type == OBJ_BLOB && repo_has_object_file(r, oid)) ||
++ (!obj && repo_has_object_file(r, oid) &&
+ oid_object_info(r, oid, NULL) == OBJ_BLOB)) {
+ if (check_object_signature(repl, NULL, 0, NULL) < 0) {
+ error(_("sha1 mismatch %s"), oid_to_hex(oid));
@@
return lookup_object(r, oid->hash);
}
6: 9cb7efc7b2 ! 28: 18b7073c25 commit: allow parse_commit* to handle arbitrary repositories
@@ -97,9 +97,10 @@
--- a/contrib/coccinelle/the_repository.cocci
+++ b/contrib/coccinelle/the_repository.cocci
@@
- + repo_read_object_file(the_repository,
- E, F, G)
-
+ - has_object_file_with_flags(
+ + repo_has_object_file_with_flags(the_repository,
+ E)
++
+@@
+expression E;
+expression F;
@@ -123,4 +124,3 @@
+- parse_commit(
++ repo_parse_commit(the_repository,
+ E)
-+
7: fcab4a4000 ! 29: 48f8999e95 commit.c: allow paint_down_to_common to handle arbitrary repositories
@@ -1,15 +1,14 @@
Author: Stefan Beller <sbeller@google.com>
- commit.c: allow paint_down_to_common to handle arbitrary repositories
+ commit-reach.c: allow paint_down_to_common to handle arbitrary repositories
As the function is file local and not widely used, migrate it all at once.
Signed-off-by: Stefan Beller <sbeller@google.com>
- Signed-off-by: Junio C Hamano <gitster@pobox.com>
-diff --git a/commit.c b/commit.c
---- a/commit.c
-+++ b/commit.c
+diff --git a/commit-reach.c b/commit-reach.c
+--- a/commit-reach.c
++++ b/commit-reach.c
@@
}
8: 324422f0d3 ! 30: b84d32c972 commit.c: allow merge_bases_many to handle arbitrary repositories
@@ -1,13 +1,12 @@
Author: Stefan Beller <sbeller@google.com>
- commit.c: allow merge_bases_many to handle arbitrary repositories
+ commit-reach.c: allow merge_bases_many to handle arbitrary repositories
Signed-off-by: Stefan Beller <sbeller@google.com>
- Signed-off-by: Junio C Hamano <gitster@pobox.com>
-diff --git a/commit.c b/commit.c
---- a/commit.c
-+++ b/commit.c
+diff --git a/commit-reach.c b/commit-reach.c
+--- a/commit-reach.c
++++ b/commit-reach.c
@@
return result;
}
9: 9c43e74d6c ! 31: cd0e595b8d commit.c: allow remove_redundant to handle arbitrary repositories
@@ -1,13 +1,12 @@
Author: Stefan Beller <sbeller@google.com>
- commit.c: allow remove_redundant to handle arbitrary repositories
+ commit-reach.c: allow remove_redundant to handle arbitrary repositories
Signed-off-by: Stefan Beller <sbeller@google.com>
- Signed-off-by: Junio C Hamano <gitster@pobox.com>
-diff --git a/commit.c b/commit.c
---- a/commit.c
-+++ b/commit.c
+diff --git a/commit-reach.c b/commit-reach.c
+--- a/commit-reach.c
++++ b/commit-reach.c
@@
return ret;
}
10: e5c2e2fa8e ! 32: 40ebd46acd commit: allow get_merge_bases_many_0 to handle arbitrary repositories
@@ -1,13 +1,12 @@
Author: Stefan Beller <sbeller@google.com>
- commit: allow get_merge_bases_many_0 to handle arbitrary repositories
+ commit-reach.c: allow get_merge_bases_many_0 to handle arbitrary repositories
Signed-off-by: Stefan Beller <sbeller@google.com>
- Signed-off-by: Junio C Hamano <gitster@pobox.com>
-diff --git a/commit.c b/commit.c
---- a/commit.c
-+++ b/commit.c
+diff --git a/commit-reach.c b/commit-reach.c
+--- a/commit-reach.c
++++ b/commit-reach.c
@@
return filled;
}
11: 27d4669ccb ! 33: 151cc8cd9b commit: prepare get_merge_bases to handle arbitrary repositories
@@ -1,6 +1,6 @@
Author: Stefan Beller <sbeller@google.com>
- commit: prepare get_merge_bases to handle arbitrary repositories
+ commit-reach: prepare get_merge_bases to handle arbitrary repositories
Similarly to previous patches, the get_merge_base functions are used
often in the code base, which makes migrating them hard.
@@ -9,11 +9,10 @@
functions behind a wrapper macro.
Signed-off-by: Stefan Beller <sbeller@google.com>
- Signed-off-by: Junio C Hamano <gitster@pobox.com>
-diff --git a/commit.c b/commit.c
---- a/commit.c
-+++ b/commit.c
+diff --git a/commit-reach.c b/commit-reach.c
+--- a/commit-reach.c
++++ b/commit-reach.c
@@
return result;
}
@@ -53,16 +52,21 @@
/*
-diff --git a/commit.h b/commit.h
---- a/commit.h
-+++ b/commit.h
+diff --git a/commit-reach.h b/commit-reach.h
+--- a/commit-reach.h
++++ b/commit-reach.h
@@
- int register_commit_graft(struct repository *r, struct commit_graft *, int);
- struct commit_graft *lookup_commit_graft(struct repository *r, const struct object_id *oid);
+ struct contains_cache;
+ struct ref_filter;
--extern struct commit_list *get_merge_bases(struct commit *rev1, struct commit *rev2);
--extern struct commit_list *get_merge_bases_many(struct commit *one, int n, struct commit **twos);
--extern struct commit_list *get_octopus_merge_bases(struct commit_list *in);
+-struct commit_list *get_merge_bases_many(struct commit *one,
+- int n,
+- struct commit **twos);
+-struct commit_list *get_merge_bases_many_dirty(struct commit *one,
+- int n,
+- struct commit **twos);
+-struct commit_list *get_merge_bases(struct commit *one, struct commit *two);
+-struct commit_list *get_octopus_merge_bases(struct commit_list *in);
-
+struct commit_list *repo_get_merge_bases(struct repository *r,
+ struct commit *rev1,
@@ -71,7 +75,7 @@
+ struct commit *one, int n,
+ struct commit **twos);
/* To be used only when object flags after this call no longer matter */
--extern struct commit_list *get_merge_bases_many_dirty(struct commit *one, int n, struct commit **twos);
+-struct commit_list *get_merge_bases_many_dirty(struct commit *one, int n, struct commit **twos);
+struct commit_list *repo_get_merge_bases_many_dirty(struct repository *r,
+ struct commit *one, int n,
+ struct commit **twos);
@@ -80,18 +84,20 @@
+#define get_merge_bases_many(one, n, two) repo_get_merge_bases_many(the_repository, one, n, two)
+#define get_merge_bases_many_dirty(one, n, twos) repo_get_merge_bases_many_dirty(the_repository, one, n, twos)
+#endif
-+extern struct commit_list *get_octopus_merge_bases(struct commit_list *in);
++
++struct commit_list *get_octopus_merge_bases(struct commit_list *in);
- /* largest positive number a signed 32-bit integer can contain */
- #define INFINITE_DEPTH 0x7fffffff
+ int is_descendant_of(struct commit *commit, struct commit_list *with_commit);
+ int in_merge_bases_many(struct commit *commit, int nr_reference, struct commit **reference);
diff --git a/contrib/coccinelle/the_repository.cocci b/contrib/coccinelle/the_repository.cocci
--- a/contrib/coccinelle/the_repository.cocci
+++ b/contrib/coccinelle/the_repository.cocci
@@
+ - parse_commit(
+ repo_parse_commit(the_repository,
E)
-
++
+@@
+expression E;
+expression F;
13: 977359ec86 ! 34: b508acf4fd commit: prepare in_merge_bases[_many] to handle arbitrary repositories
@@ -1,13 +1,12 @@
Author: Stefan Beller <sbeller@google.com>
- commit: prepare in_merge_bases[_many] to handle arbitrary repositories
+ commit-reach: prepare in_merge_bases[_many] to handle arbitrary repositories
Signed-off-by: Stefan Beller <sbeller@google.com>
- Signed-off-by: Junio C Hamano <gitster@pobox.com>
-diff --git a/commit.c b/commit.c
---- a/commit.c
-+++ b/commit.c
+diff --git a/commit-reach.c b/commit-reach.c
+--- a/commit-reach.c
++++ b/commit-reach.c
@@
/*
* Is "commit" an ancestor of one of the "references"?
@@ -53,33 +52,36 @@
struct commit_list *reduce_heads(struct commit_list *heads)
-diff --git a/commit.h b/commit.h
---- a/commit.h
-+++ b/commit.h
+diff --git a/commit-reach.h b/commit-reach.h
+--- a/commit-reach.h
++++ b/commit-reach.h
@@
- extern struct trace_key trace_shallow;
+ struct commit_list *get_octopus_merge_bases(struct commit_list *in);
- int is_descendant_of(struct commit *, struct commit_list *);
--int in_merge_bases(struct commit *, struct commit *);
--int in_merge_bases_many(struct commit *, int, struct commit **);
-+int repo_in_merge_bases(struct repository *r, struct commit *, struct commit *);
-+int repo_in_merge_bases_many(struct repository *r, struct commit *, int, struct commit **);
+ int is_descendant_of(struct commit *commit, struct commit_list *with_commit);
+-int in_merge_bases_many(struct commit *commit, int nr_reference, struct commit **reference);
+-int in_merge_bases(struct commit *commit, struct commit *reference);
++int repo_in_merge_bases(struct repository *r,
++ struct commit *commit,
++ struct commit *reference);
++int repo_in_merge_bases_many(struct repository *r,
++ struct commit *commit,
++ int nr_reference, struct commit **reference);
+#ifndef NO_THE_REPOSITORY_COMPATIBILITY_MACROS
+#define in_merge_bases(c1, c2) repo_in_merge_bases(the_repository, c1, c2)
+#define in_merge_bases_many(c1, n, cs) repo_in_merge_bases_many(the_repository, c1, n, cs)
+#endif
- extern int interactive_add(int argc, const char **argv, const char *prefix, int patch);
- extern int run_add_interactive(const char *revision, const char *patch_mode,
+ /*
+ * Takes a list of commits and returns a new list where those
diff --git a/contrib/coccinelle/the_repository.cocci b/contrib/coccinelle/the_repository.cocci
--- a/contrib/coccinelle/the_repository.cocci
+++ b/contrib/coccinelle/the_repository.cocci
@@
- - get_commit_buffer(
- + repo_get_commit_buffer(the_repository,
- E, F);
-+
+ + repo_get_merge_bases_many_dirty(the_repository,
+ E, F, G);
+
+@@
+expression E;
+expression F;
12: 044bf2917c ! 35: c825f2703c commit: prepare get_commit_buffer to handle arbitrary repositories
@@ -3,7 +3,6 @@
commit: prepare get_commit_buffer to handle arbitrary repositories
Signed-off-by: Stefan Beller <sbeller@google.com>
- Signed-off-by: Junio C Hamano <gitster@pobox.com>
diff --git a/commit.c b/commit.c
--- a/commit.c
@@ -50,9 +49,10 @@
--- a/contrib/coccinelle/the_repository.cocci
+++ b/contrib/coccinelle/the_repository.cocci
@@
- + repo_get_merge_bases_many_dirty(the_repository,
+ - in_merge_bases_many(
+ + repo_in_merge_bases_many(the_repository,
E, F, G);
-
++
+@@
+expression E;
+expression F;
14: 65771970b9 ! 36: bea6ff2e6d commit: prepare repo_unuse_commit_buffer to handle arbitrary repositories
@@ -3,7 +3,6 @@
commit: prepare repo_unuse_commit_buffer to handle arbitrary repositories
Signed-off-by: Stefan Beller <sbeller@google.com>
- Signed-off-by: Junio C Hamano <gitster@pobox.com>
diff --git a/commit.c b/commit.c
--- a/commit.c
@@ -46,9 +45,10 @@
--- a/contrib/coccinelle/the_repository.cocci
+++ b/contrib/coccinelle/the_repository.cocci
@@
+ - get_commit_buffer(
+ repo_get_commit_buffer(the_repository,
E, F);
-
++
+@@
+expression E;
+expression F;
@@ -56,7 +56,3 @@
+- unuse_commit_buffer(
++ repo_unuse_commit_buffer(the_repository,
+ E, F);
-+
- @@
- expression E;
- expression F;
15: ddef24685d ! 37: 11ea084cd2 commit: prepare logmsg_reencode to handle arbitrary repositories
@@ -28,9 +28,9 @@
--- a/contrib/coccinelle/the_repository.cocci
+++ b/contrib/coccinelle/the_repository.cocci
@@
- - in_merge_bases_many(
- + repo_in_merge_bases_many(the_repository,
- E, F, G);
+ - unuse_commit_buffer(
+ + repo_unuse_commit_buffer(the_repository,
+ E, F);
+
+@@
+expression E;
16: 8612153e23 = 38: 9f50edd155 pretty: prepare format_commit_message to handle arbitrary repositories
17: eba296fc18 ! 39: 1a4a392233 submodule: use submodule repos for object lookup
@@ -7,7 +7,6 @@
are not added to the main object store.
Signed-off-by: Stefan Beller <sbeller@google.com>
- Signed-off-by: Junio C Hamano <gitster@pobox.com>
diff --git a/submodule.c b/submodule.c
--- a/submodule.c
@@ -35,12 +34,15 @@
DEFAULT_GIT_DIR_ENVIRONMENT);
}
+-/* Helper function to display the submodule header line prior to the full
+- * summary output. If it can locate the submodule objects directory it will
+- * attempt to lookup both the left and right commits and put them into the
+- * left and right pointers.
+/*
+ * Initialize 'out' based on the provided submodule path.
+ *
+ * Unlike repo_submodule_init, this tolerates submodules not present
-+ * in .gitmodules. NEEDSWORK: The repo_submodule_init behavior is
-+ * preferrable. This function exists only to preserve historical behavior.
++ * in .gitmodules. This function exists only to preserve historical behavior,
+ *
+ * Returns 0 on success, -1 when the submodule is not present.
+ */
@@ -54,28 +56,38 @@
+ }
+
+ out->submodule_prefix = xstrdup(path);
++ out->submodule_prefix = xstrfmt("%s%s/",
++ the_repository->submodule_prefix ?
++ the_repository->submodule_prefix :
++ "", path);
+
+ strbuf_release(&sb);
+ return 0;
+}
+
- /* Helper function to display the submodule header line prior to the full
- * summary output. If it can locate the submodule objects directory it will
- * attempt to lookup both the left and right commits and put them into the
- * left and right pointers.
++/*
++ * Helper function to display the submodule header line prior to the full
++ * summary output.
++ *
++ * If it can locate the submodule git directory it will create a repository
++ * handle for the submodule and lookup both the left and right commits and
++ * put them into the left and right pointers.
*/
-static void show_submodule_header(struct diff_options *o, const char *path,
-+static void show_submodule_header(struct diff_options *o, struct repository *sub,
++static void show_submodule_header(struct diff_options *o,
+ const char *path,
struct object_id *one, struct object_id *two,
unsigned dirty_submodule,
++ struct repository *sub,
struct commit **left, struct commit **right,
+ struct commit_list **merge_bases)
+ {
@@
else if (is_null_oid(two))
message = "(submodule deleted)";
- if (add_submodule_odb(path)) {
-+ if (open_submodule(sub, path) < 0) {
++ if (!sub) {
if (!message)
message = "(commits not present)";
goto output_header;
@@ -103,30 +115,63 @@
struct rev_info rev;
struct commit *left = NULL, *right = NULL;
struct commit_list *merge_bases = NULL;
-+ struct repository sub;
++ struct repository subrepo, *sub = &subrepo;
++
++ if (open_submodule(&subrepo, path) < 0)
++ sub = NULL;
-- show_submodule_header(o, path, one, two, dirty_submodule,
-+ show_submodule_header(o, &sub, path, one, two, dirty_submodule,
- &left, &right, &merge_bases);
+ show_submodule_header(o, path, one, two, dirty_submodule,
+- &left, &right, &merge_bases);
++ sub, &left, &right, &merge_bases);
/*
+ * If we don't have both a left and a right pointer, there is no
+ * reason to try and display a summary. The header line should contain
+ * all the information the user needs.
+ */
+- if (!left || !right)
++ if (!left || !right || !sub)
+ goto out;
+
+ /* Treat revision walker failure the same as missing commits */
@@
goto out;
}
- print_submodule_summary(&rev, o);
-+ print_submodule_summary(&sub, &rev, o);
++ print_submodule_summary(sub, &rev, o);
out:
if (merge_bases)
+ free_commit_list(merge_bases);
+ clear_commit_marks(left, ~0);
+ clear_commit_marks(right, ~0);
++ if (sub)
++ repo_clear(sub);
+ }
+
+ void show_submodule_inline_diff(struct diff_options *o, const char *path,
@@
struct commit_list *merge_bases = NULL;
struct child_process cp = CHILD_PROCESS_INIT;
struct strbuf sb = STRBUF_INIT;
-+ struct repository sub;
++ struct repository subrepo, *sub = &subrepo;
++
++ if (open_submodule(&subrepo, path) < 0)
++ sub = NULL;
-- show_submodule_header(o, path, one, two, dirty_submodule,
-+ show_submodule_header(o, &sub, path, one, two, dirty_submodule,
- &left, &right, &merge_bases);
+ show_submodule_header(o, path, one, two, dirty_submodule,
+- &left, &right, &merge_bases);
++ sub, &left, &right, &merge_bases);
/* We need a valid left and right commit to display a difference */
+ if (!(left || is_null_oid(one)) ||
+@@
+ clear_commit_marks(left, ~0);
+ if (right)
+ clear_commit_marks(right, ~0);
++ if (sub)
++ repo_clear(sub);
+ }
+
+ int should_update_submodules(void)
18: 4b2033017a ! 40: 1ce2d0a867 submodule: don't add submodule as odb for push
@@ -2,20 +2,13 @@
submodule: don't add submodule as odb for push
- The submodule was added as an alternative in eb21c732d6 (push: teach
- --recurse-submodules the on-demand option, 2012-03-29), but was
- not explained, why.
-
- In similar code, submodule_has_commits, the submodule is added as an
- alternative to perform a quick check if we need to dive into the submodule.
-
- However in push_submodule
- (a) for_each_remote_ref_submodule will also provide the quick check and
- (b) after that we don't need to have submodule objects around, as all
- further code is to spawn a separate process.
+ In push_submodule(), because we do not actually need access to objects
+ in the submodule, do not invoke add_submodule_odb().
+ (for_each_remote_ref_submodule() does not require access to those
+ objects, and the actual push is done by spawning another process,
+ which handles object access by itself.)
Signed-off-by: Stefan Beller <sbeller@google.com>
- Signed-off-by: Junio C Hamano <gitster@pobox.com>
diff --git a/submodule.c b/submodule.c
--- a/submodule.c
19: c74af1496b < -: ---------- Apply semantic patches from previous patches
^ permalink raw reply [relevance 7%]
* Re: [PATCH 17/19] submodule: use submodule repos for object lookup
@ 2018-10-16 23:16 7% ` Stefan Beller
0 siblings, 0 replies; 200+ results
From: Stefan Beller @ 2018-10-16 23:16 UTC (permalink / raw)
To: Jonathan Tan; +Cc: git
On Tue, Oct 16, 2018 at 4:13 PM Jonathan Tan <jonathantanmy@google.com> wrote:
>
> > Thanks for the review of the whole series!
> >
> > I have redone this series, addressing all your comments. I addressed
> > this comment differently than suggested, and put the submodule
> > repository argument at the end of the parameter list, such that it
> > goes with all the other arguments to be filled in.
>
> Sounds good.
Actually I changed my mind on that after figuring out how to free
the submodule repository appropriately and went with your original suggestion.
>
> > I was about to resend the series, but test-merged with the other
> > submodule series in flight (origin/sb/submodule-recursive-fetch-gets-the-tip)
> > which had some conflicts that I can easily resolve by rebasing on top.
>
> I presume you are talking about [1]? Maybe consider rebasing that one on
> top of this instead, since this is just a refactoring whereas
> submodule-recursive-fetch-gets-the-tip changes functionality, from what
> I understand of patches 8 and 9.
From my understanding, that series is further along than this one,
so I would not want to mix up their order.
Currently I am rebasing this on top of select topics from next,
(ds/reachable) as that are the other conflicts that I'd have to deal with.
> [1] https://public-inbox.org/git/20181016181327.107186-1-sbeller@google.com/
>
> > It conflicts a lot when merging to next, due to the latest patch
> > ("Apply semantic patches from previous patches"), so I am not sure
> > how to proceed properly. Maybe we'd omit that patch and
> > run 'make coccicheck' on next to apply the semantic patches
> > there instead.
>
> Omitting the patch sounds good to me. For me, just stating that you have
> excluded any coccinelle-generated patches in order to ease merging into
> the various branches is sufficient, and people can test the coccinelle
> patches included by running "make coccicheck" then "patch -p1
> <contrib/coccinelle/the_repository.cocci.patch".
ok.
Thanks,
Stefan
^ permalink raw reply [relevance 7%]
* Re: [PATCH] submodule helper: convert relative URL to absolute URL if needed
2018-10-16 17:27 21% ` Stefan Beller
@ 2018-10-16 21:05 4% ` Jonathan Nieder
0 siblings, 0 replies; 200+ results
From: Jonathan Nieder @ 2018-10-16 21:05 UTC (permalink / raw)
To: Stefan Beller; +Cc: gitster, git, jungjw
Stefan Beller wrote:
> Reported-by: Jaewoong Jung <jungjw@google.com>
> Signed-off-by: Stefan Beller <sbeller@google.com>
> ---
> builtin/submodule--helper.c | 51 ++++++++++++++++++++++++-------------
> t/t7400-submodule-basic.sh | 24 +++++++++++++++++
> 2 files changed, 58 insertions(+), 17 deletions(-)
Reviewed-by: Jonathan Nieder <jrnieder@gmail.com>
Thanks for your patient work.
^ permalink raw reply [relevance 4%]
* Re: [PATCH 17/19] submodule: use submodule repos for object lookup
2018-10-13 0:20 8% ` Stefan Beller
@ 2018-10-16 19:30 8% ` Stefan Beller
1 sibling, 1 reply; 200+ results
From: Stefan Beller @ 2018-10-16 19:30 UTC (permalink / raw)
To: Jonathan Tan; +Cc: git
On Thu, Oct 11, 2018 at 3:41 PM Jonathan Tan <jonathantanmy@google.com> wrote:
>
> > +/*
> > + * Initialize 'out' based on the provided submodule path.
> > + *
> > + * Unlike repo_submodule_init, this tolerates submodules not present
> > + * in .gitmodules. NEEDSWORK: The repo_submodule_init behavior is
> > + * preferrable. This function exists only to preserve historical behavior.
>
> What do you mean by "The repo_submodule_init behavior is preferable"? If
> we need to preserve historical behavior, then it seems that the most
> preferable one is something that meets our needs (like open_submodule()
> in this patch).
>
> > +static int open_submodule(struct repository *out, const char *path)
> > +{
> > + struct strbuf sb = STRBUF_INIT;
> > +
> > + if (submodule_to_gitdir(&sb, path) || repo_init(out, sb.buf, NULL)) {
> > + strbuf_release(&sb);
> > + return -1;
> > + }
> > +
> > + out->submodule_prefix = xstrdup(path);
>
> Do we need to set submodule_prefix?
>
> > @@ -507,7 +533,7 @@ static void show_submodule_header(struct diff_options *o, const char *path,
> > else if (is_null_oid(two))
> > message = "(submodule deleted)";
> >
> > - if (add_submodule_odb(path)) {
> > + if (open_submodule(sub, path) < 0) {
>
> This function, as a side effect, writes the open repository to "sub" for
> its caller to use. I think it's better if its callers open "sub" and
> then pass it to show_submodule_header() if successful. If opening the
> submodule is expected to fail sometimes (like it seems here), then we
> can allow callers to also pass NULL, and document what happens when NULL
> is passed.
Thanks for the review of the whole series!
I have redone this series, addressing all your comments. I addressed
this comment differently than suggested, and put the submodule
repository argument at the end of the parameter list, such that it
goes with all the other arguments to be filled in.
I was about to resend the series, but test-merged with the other
submodule series in flight (origin/sb/submodule-recursive-fetch-gets-the-tip)
which had some conflicts that I can easily resolve by rebasing on top.
It conflicts a lot when merging to next, due to the latest patch
("Apply semantic patches from previous patches"), so I am not sure
how to proceed properly. Maybe we'd omit that patch and
run 'make coccicheck' on next to apply the semantic patches
there instead.
^ permalink raw reply [relevance 8%]
* [PATCH 9/9] builtin/fetch: check for submodule updates for non branch fetches
2018-10-16 18:13 9% [PATCH 0/9] Resending sb/submodule-recursive-fetch-gets-the-tip Stefan Beller
` (6 preceding siblings ...)
2018-10-16 18:13 23% ` [PATCH 8/9] fetch: retry fetching submodules if needed objects were not fetched Stefan Beller
@ 2018-10-16 18:13 21% ` Stefan Beller
2018-10-18 2:30 4% ` [PATCH 0/9] Resending sb/submodule-recursive-fetch-gets-the-tip Junio C Hamano
2018-10-18 7:30 14% ` Junio C Hamano
9 siblings, 0 replies; 200+ results
From: Stefan Beller @ 2018-10-16 18:13 UTC (permalink / raw)
To: gitster; +Cc: git, Stefan Beller
Gerrit, the code review tool, has a different workflow than our mailing
list based approach. Usually users upload changes to a Gerrit server and
continuous integration and testing happens by bots. Sometimes however a
user wants to checkout a change locally and look at it locally. For this
use case, Gerrit offers a command line snippet to copy and paste to your
terminal, which looks like
git fetch https://<host>/gerrit refs/changes/<id> &&
git checkout FETCH_HEAD
For Gerrit changes that contain changing submodule gitlinks, it would be
easy to extend both the fetch and checkout with the '--recurse-submodules'
flag, such that this command line snippet would produce the state of a
change locally.
However the functionality added in the previous patch, which would
ensure that we fetch the objects in the submodule that the gitlink pointed
at, only works for remote tracking branches so far, not for FETCH_HEAD.
Make sure that fetching a superproject to its FETCH_HEAD, also respects
the existence checks for objects in the submodule recursion.
Signed-off-by: Stefan Beller <sbeller@google.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
---
builtin/fetch.c | 5 ++++-
t/t5526-fetch-submodules.sh | 2 +-
2 files changed, 5 insertions(+), 2 deletions(-)
diff --git a/builtin/fetch.c b/builtin/fetch.c
index 95c44bf6ff..ea6ecd123e 100644
--- a/builtin/fetch.c
+++ b/builtin/fetch.c
@@ -887,11 +887,14 @@ static int store_updated_refs(const char *raw_url, const char *remote_name,
rc |= update_local_ref(ref, what, rm, ¬e,
summary_width);
free(ref);
- } else
+ } else {
+ check_for_new_submodule_commits(&rm->old_oid);
format_display(¬e, '*',
*kind ? kind : "branch", NULL,
*what ? what : "HEAD",
"FETCH_HEAD", summary_width);
+ }
+
if (note.len) {
if (verbosity >= 0 && !shown_url) {
fprintf(stderr, _("From %.*s\n"),
diff --git a/t/t5526-fetch-submodules.sh b/t/t5526-fetch-submodules.sh
index af12c50e7d..a509eabb04 100755
--- a/t/t5526-fetch-submodules.sh
+++ b/t/t5526-fetch-submodules.sh
@@ -615,7 +615,7 @@ test_expect_success "fetch new commits on-demand when they are not reachable" '
git update-ref refs/changes/2 $D &&
(
cd downstream &&
- git fetch --recurse-submodules --recurse-submodules-default on-demand origin refs/changes/2:refs/heads/my_branch &&
+ git fetch --recurse-submodules origin refs/changes/2 &&
git -C submodule cat-file -t $C &&
git checkout --recurse-submodules FETCH_HEAD
)
--
2.19.0
^ permalink raw reply related [relevance 21%]
* [PATCH 8/9] fetch: retry fetching submodules if needed objects were not fetched
2018-10-16 18:13 9% [PATCH 0/9] Resending sb/submodule-recursive-fetch-gets-the-tip Stefan Beller
` (5 preceding siblings ...)
2018-10-16 18:13 26% ` [PATCH 7/9] submodule: fetch in submodules git directory instead of in worktree Stefan Beller
@ 2018-10-16 18:13 23% ` Stefan Beller
2018-10-16 18:13 21% ` [PATCH 9/9] builtin/fetch: check for submodule updates for non branch fetches Stefan Beller
` (2 subsequent siblings)
9 siblings, 1 reply; 200+ results
From: Stefan Beller @ 2018-10-16 18:13 UTC (permalink / raw)
To: gitster; +Cc: git, Stefan Beller
Currently when git-fetch is asked to recurse into submodules, it dispatches
a plain "git-fetch -C <submodule-dir>" (with some submodule related options
such as prefix and recusing strategy, but) without any information of the
remote or the tip that should be fetched.
This works surprisingly well in some workflows (such as using submodules
as a third party library), while not so well in other scenarios, such
as in a Gerrit topic-based workflow, that can tie together changes
(potentially across repositories) on the server side. One of the parts
of such a Gerrit workflow is to download a change when wanting to examine
it, and you'd want to have its submodule changes that are in the same
topic downloaded as well. However these submodule changes reside in their
own repository in their own ref (refs/changes/<int>).
Retry fetching a submodule by object name if the object id that the
superproject points to, cannot be found.
This retrying does not happen when the "git fetch" done at the
superproject is not storing the fetched results in remote
tracking branches (i.e. instead just recording them to
FETCH_HEAD) in this step. A later patch will fix this.
builtin/fetch used to only inspect submodules when they were fetched
"on-demand", as in either on/off case it was clear whether the submodule
needs to be fetched. However to know whether we need to try fetching the
object ids, we need to identify the object names, which is done in this
function check_for_new_submodule_commits(), so we'll also run that code
in case the submodule recursion is set to "on".
Signed-off-by: Stefan Beller <sbeller@google.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
---
builtin/fetch.c | 9 +-
submodule.c | 185 ++++++++++++++++++++++++++++++------
t/t5526-fetch-submodules.sh | 16 ++++
3 files changed, 177 insertions(+), 33 deletions(-)
diff --git a/builtin/fetch.c b/builtin/fetch.c
index 61bec5d213..95c44bf6ff 100644
--- a/builtin/fetch.c
+++ b/builtin/fetch.c
@@ -700,8 +700,7 @@ static int update_local_ref(struct ref *ref,
what = _("[new ref]");
}
- if ((recurse_submodules != RECURSE_SUBMODULES_OFF) &&
- (recurse_submodules != RECURSE_SUBMODULES_ON))
+ if (recurse_submodules != RECURSE_SUBMODULES_OFF)
check_for_new_submodule_commits(&ref->new_oid);
r = s_update_ref(msg, ref, 0);
format_display(display, r ? '!' : '*', what,
@@ -716,8 +715,7 @@ static int update_local_ref(struct ref *ref,
strbuf_add_unique_abbrev(&quickref, ¤t->object.oid, DEFAULT_ABBREV);
strbuf_addstr(&quickref, "..");
strbuf_add_unique_abbrev(&quickref, &ref->new_oid, DEFAULT_ABBREV);
- if ((recurse_submodules != RECURSE_SUBMODULES_OFF) &&
- (recurse_submodules != RECURSE_SUBMODULES_ON))
+ if (recurse_submodules != RECURSE_SUBMODULES_OFF)
check_for_new_submodule_commits(&ref->new_oid);
r = s_update_ref("fast-forward", ref, 1);
format_display(display, r ? '!' : ' ', quickref.buf,
@@ -731,8 +729,7 @@ static int update_local_ref(struct ref *ref,
strbuf_add_unique_abbrev(&quickref, ¤t->object.oid, DEFAULT_ABBREV);
strbuf_addstr(&quickref, "...");
strbuf_add_unique_abbrev(&quickref, &ref->new_oid, DEFAULT_ABBREV);
- if ((recurse_submodules != RECURSE_SUBMODULES_OFF) &&
- (recurse_submodules != RECURSE_SUBMODULES_ON))
+ if (recurse_submodules != RECURSE_SUBMODULES_OFF)
check_for_new_submodule_commits(&ref->new_oid);
r = s_update_ref("forced-update", ref, 1);
format_display(display, r ? '!' : '+', quickref.buf,
diff --git a/submodule.c b/submodule.c
index 30c06507e3..7246b776f3 100644
--- a/submodule.c
+++ b/submodule.c
@@ -1141,8 +1141,12 @@ struct submodule_parallel_fetch {
int result;
struct string_list changed_submodule_names;
+ struct get_next_submodule_task **retry;
+ int retry_nr, retry_alloc;
};
-#define SPF_INIT {0, ARGV_ARRAY_INIT, NULL, NULL, 0, 0, 0, 0, STRING_LIST_INIT_DUP }
+#define SPF_INIT {0, ARGV_ARRAY_INIT, NULL, NULL, 0, 0, 0, 0, \
+ STRING_LIST_INIT_DUP, \
+ NULL, 0, 0}
static void calculate_changed_submodule_paths(
struct submodule_parallel_fetch *spf)
@@ -1247,6 +1251,56 @@ static int get_fetch_recurse_config(const struct submodule *submodule,
return spf->default_option;
}
+struct get_next_submodule_task {
+ struct repository *repo;
+ const struct submodule *sub;
+ unsigned free_sub : 1; /* Do we need to free the submodule? */
+ struct oid_array *commits;
+};
+
+static const struct submodule *get_default_submodule(const char *path)
+{
+ struct submodule *ret = NULL;
+ const char *name = default_name_or_path(path);
+
+ if (!name)
+ return NULL;
+
+ ret = xmalloc(sizeof(*ret));
+ memset(ret, 0, sizeof(*ret));
+ ret->path = name;
+ ret->name = name;
+
+ return (const struct submodule *) ret;
+}
+
+static struct get_next_submodule_task *get_next_submodule_task_create(
+ struct repository *r, const char *path)
+{
+ struct get_next_submodule_task *task = xmalloc(sizeof(*task));
+ memset(task, 0, sizeof(*task));
+
+ task->sub = submodule_from_path(r, &null_oid, path);
+ if (!task->sub) {
+ task->sub = get_default_submodule(path);
+ task->free_sub = 1;
+ }
+
+ return task;
+}
+
+static void get_next_submodule_task_release(struct get_next_submodule_task *p)
+{
+ if (p->free_sub)
+ free((void*)p->sub);
+ p->free_sub = 0;
+ p->sub = NULL;
+
+ if (p->repo)
+ repo_clear(p->repo);
+ FREE_AND_NULL(p->repo);
+}
+
static struct repository *get_submodule_repo_for(struct repository *r,
const struct submodule *sub)
{
@@ -1273,39 +1327,35 @@ static struct repository *get_submodule_repo_for(struct repository *r,
static int get_next_submodule(struct child_process *cp,
struct strbuf *err, void *data, void **task_cb)
{
- int ret = 0;
struct submodule_parallel_fetch *spf = data;
for (; spf->count < spf->r->index->cache_nr; spf->count++) {
- struct strbuf submodule_prefix = STRBUF_INIT;
+ int recurse_config;
const struct cache_entry *ce = spf->r->index->cache[spf->count];
const char *default_argv;
- const struct submodule *submodule;
- struct repository *repo;
- struct submodule default_submodule = SUBMODULE_INIT;
+ struct get_next_submodule_task *task;
if (!S_ISGITLINK(ce->ce_mode))
continue;
- submodule = submodule_from_path(spf->r, &null_oid, ce->name);
- if (!submodule) {
- const char *name = default_name_or_path(ce->name);
- if (name) {
- default_submodule.path = name;
- default_submodule.name = name;
- submodule = &default_submodule;
- }
+ task = get_next_submodule_task_create(spf->r, ce->name);
+
+ if (!task->sub) {
+ free(task);
+ continue;
}
- switch (get_fetch_recurse_config(submodule, spf))
+ recurse_config = get_fetch_recurse_config(task->sub, spf);
+
+ switch (recurse_config)
{
default:
case RECURSE_SUBMODULES_DEFAULT:
case RECURSE_SUBMODULES_ON_DEMAND:
- if (!submodule ||
+ if (!task->sub ||
!string_list_lookup(
&spf->changed_submodule_names,
- submodule->name))
+ task->sub->name))
continue;
default_argv = "on-demand";
break;
@@ -1316,12 +1366,12 @@ static int get_next_submodule(struct child_process *cp,
continue;
}
- strbuf_addf(&submodule_prefix, "%s%s/", spf->prefix, ce->name);
- repo = get_submodule_repo_for(spf->r, submodule);
- if (repo) {
+ task->repo = get_submodule_repo_for(spf->r, task->sub);
+ if (task->repo) {
+ struct strbuf submodule_prefix = STRBUF_INIT;
child_process_init(cp);
prepare_submodule_repo_env_in_gitdir(&cp->env_array);
- cp->dir = xstrdup(repo->gitdir);
+ cp->dir = task->repo->gitdir;
cp->git_cmd = 1;
if (!spf->quiet)
strbuf_addf(err, "Fetching submodule %s%s\n",
@@ -1330,18 +1380,51 @@ static int get_next_submodule(struct child_process *cp,
argv_array_pushv(&cp->args, spf->args.argv);
argv_array_push(&cp->args, default_argv);
argv_array_push(&cp->args, "--submodule-prefix");
+
+ strbuf_addf(&submodule_prefix, "%s%s/",
+ spf->prefix,
+ task->sub->path);
argv_array_push(&cp->args, submodule_prefix.buf);
- repo_clear(repo);
- free(repo);
- ret = 1;
- }
- strbuf_release(&submodule_prefix);
- if (ret) {
spf->count++;
+ *task_cb = task;
+
+ strbuf_release(&submodule_prefix);
return 1;
+ } else {
+ get_next_submodule_task_release(task);
+ free(task);
}
}
+
+ if (spf->retry_nr) {
+ struct get_next_submodule_task *task = spf->retry[spf->retry_nr - 1];
+ struct strbuf submodule_prefix = STRBUF_INIT;
+ spf->retry_nr--;
+
+ strbuf_addf(&submodule_prefix, "%s%s/", spf->prefix, task->sub->path);
+
+ child_process_init(cp);
+ prepare_submodule_repo_env_in_gitdir(&cp->env_array);
+ cp->git_cmd = 1;
+ cp->dir = task->repo->gitdir;
+
+ argv_array_init(&cp->args);
+ argv_array_pushv(&cp->args, spf->args.argv);
+ argv_array_push(&cp->args, "on-demand");
+ argv_array_push(&cp->args, "--submodule-prefix");
+ argv_array_push(&cp->args, submodule_prefix.buf);
+
+ /* NEEDSWORK: have get_default_remote from s--h */
+ argv_array_push(&cp->args, "origin");
+ oid_array_for_each_unique(task->commits,
+ append_oid_to_argv, &cp->args);
+
+ *task_cb = task;
+ strbuf_release(&submodule_prefix);
+ return 1;
+ }
+
return 0;
}
@@ -1349,20 +1432,68 @@ static int fetch_start_failure(struct strbuf *err,
void *cb, void *task_cb)
{
struct submodule_parallel_fetch *spf = cb;
+ struct get_next_submodule_task *task = task_cb;
spf->result = 1;
+ get_next_submodule_task_release(task);
return 0;
}
+static int commit_exists_in_sub(const struct object_id *oid, void *data)
+{
+ struct repository *subrepo = data;
+
+ enum object_type type = oid_object_info(subrepo, oid, NULL);
+
+ return type != OBJ_COMMIT;
+}
+
static int fetch_finish(int retvalue, struct strbuf *err,
void *cb, void *task_cb)
{
struct submodule_parallel_fetch *spf = cb;
+ struct get_next_submodule_task *task = task_cb;
+ const struct submodule *sub;
+
+ struct string_list_item *it;
+ struct oid_array *commits;
if (retvalue)
spf->result = 1;
+ if (!task)
+ return 0;
+
+ sub = task->sub;
+ if (!sub)
+ goto out;
+
+ it = string_list_lookup(&spf->changed_submodule_names, sub->name);
+ if (!it)
+ goto out;
+
+ commits = it->util;
+ oid_array_filter(commits,
+ commit_exists_in_sub,
+ task->repo);
+
+ /* Are there commits that do not exist? */
+ if (commits->nr) {
+ /* We already tried fetching them, do not try again. */
+ if (task->commits)
+ return 0;
+
+ task->commits = commits;
+ ALLOC_GROW(spf->retry, spf->retry_nr + 1, spf->retry_alloc);
+ spf->retry[spf->retry_nr] = task;
+ spf->retry_nr++;
+ return 0;
+ }
+
+out:
+ get_next_submodule_task_release(task);
+
return 0;
}
diff --git a/t/t5526-fetch-submodules.sh b/t/t5526-fetch-submodules.sh
index 42692219a1..af12c50e7d 100755
--- a/t/t5526-fetch-submodules.sh
+++ b/t/t5526-fetch-submodules.sh
@@ -605,4 +605,20 @@ test_expect_success "fetch new commits when submodule got renamed" '
test_cmp expect actual
'
+test_expect_success "fetch new commits on-demand when they are not reachable" '
+ git checkout --detach &&
+ C=$(git -C submodule commit-tree -m "new change outside refs/heads" HEAD^{tree}) &&
+ git -C submodule update-ref refs/changes/1 $C &&
+ git update-index --cacheinfo 160000 $C submodule &&
+ git commit -m "updated submodule outside of refs/heads" &&
+ D=$(git rev-parse HEAD) &&
+ git update-ref refs/changes/2 $D &&
+ (
+ cd downstream &&
+ git fetch --recurse-submodules --recurse-submodules-default on-demand origin refs/changes/2:refs/heads/my_branch &&
+ git -C submodule cat-file -t $C &&
+ git checkout --recurse-submodules FETCH_HEAD
+ )
+'
+
test_done
--
2.19.0
^ permalink raw reply related [relevance 23%]
* [PATCH 7/9] submodule: fetch in submodules git directory instead of in worktree
2018-10-16 18:13 9% [PATCH 0/9] Resending sb/submodule-recursive-fetch-gets-the-tip Stefan Beller
` (4 preceding siblings ...)
2018-10-16 18:13 24% ` [PATCH 6/9] repository: repo_submodule_init to take a submodule struct Stefan Beller
@ 2018-10-16 18:13 26% ` Stefan Beller
2018-10-16 18:13 23% ` [PATCH 8/9] fetch: retry fetching submodules if needed objects were not fetched Stefan Beller
` (3 subsequent siblings)
9 siblings, 1 reply; 200+ results
From: Stefan Beller @ 2018-10-16 18:13 UTC (permalink / raw)
To: gitster; +Cc: git, Stefan Beller
This patch started as a refactoring to make 'get_next_submodule' more
readable, but upon doing so, I realized that "git fetch" of the submodule
actually doesn't need to be run in the submodules worktree. So let's run
it in its git dir instead.
That should pave the way towards fetching submodules that are currently
not checked out.
This patch leaks the cp->dir in get_next_submodule, as any further
callback in run_processes_parallel doesn't have access to the child
process any more. In an early iteration of this patch, the function
get_submodule_repo_for directly returned the string containing the
git directory, which would be a better design choice for this patch.
However the next patch both fixes the memory leak of cp->dir and also has
a use case for using the full repository handle of the submodule, so
it makes sense to introduce the get_submodule_repo_for here already.
Signed-off-by: Stefan Beller <sbeller@google.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
---
submodule.c | 51 +++++++++++++++++++++++++++----------
t/t5526-fetch-submodules.sh | 7 ++++-
2 files changed, 44 insertions(+), 14 deletions(-)
diff --git a/submodule.c b/submodule.c
index cbefe5f54d..30c06507e3 100644
--- a/submodule.c
+++ b/submodule.c
@@ -495,6 +495,12 @@ void prepare_submodule_repo_env(struct argv_array *out)
DEFAULT_GIT_DIR_ENVIRONMENT);
}
+static void prepare_submodule_repo_env_in_gitdir(struct argv_array *out)
+{
+ prepare_submodule_repo_env_no_git_dir(out);
+ argv_array_pushf(out, "%s=.", GIT_DIR_ENVIRONMENT);
+}
+
/* Helper function to display the submodule header line prior to the full
* summary output. If it can locate the submodule objects directory it will
* attempt to lookup both the left and right commits and put them into the
@@ -1241,6 +1247,29 @@ static int get_fetch_recurse_config(const struct submodule *submodule,
return spf->default_option;
}
+static struct repository *get_submodule_repo_for(struct repository *r,
+ const struct submodule *sub)
+{
+ struct repository *ret = xmalloc(sizeof(*ret));
+
+ if (repo_submodule_init(ret, r, sub)) {
+ /*
+ * No entry in .gitmodules? Technically not a submodule,
+ * but historically we supported repositories that happen to be
+ * in-place where a gitlink is. Keep supporting them.
+ */
+ struct strbuf gitdir = STRBUF_INIT;
+ strbuf_repo_worktree_path(&gitdir, r, "%s/.git", sub->path);
+ if (repo_init(ret, gitdir.buf, NULL)) {
+ strbuf_release(&gitdir);
+ return NULL;
+ }
+ strbuf_release(&gitdir);
+ }
+
+ return ret;
+}
+
static int get_next_submodule(struct child_process *cp,
struct strbuf *err, void *data, void **task_cb)
{
@@ -1248,12 +1277,11 @@ static int get_next_submodule(struct child_process *cp,
struct submodule_parallel_fetch *spf = data;
for (; spf->count < spf->r->index->cache_nr; spf->count++) {
- struct strbuf submodule_path = STRBUF_INIT;
- struct strbuf submodule_git_dir = STRBUF_INIT;
struct strbuf submodule_prefix = STRBUF_INIT;
const struct cache_entry *ce = spf->r->index->cache[spf->count];
- const char *git_dir, *default_argv;
+ const char *default_argv;
const struct submodule *submodule;
+ struct repository *repo;
struct submodule default_submodule = SUBMODULE_INIT;
if (!S_ISGITLINK(ce->ce_mode))
@@ -1288,16 +1316,12 @@ static int get_next_submodule(struct child_process *cp,
continue;
}
- strbuf_repo_worktree_path(&submodule_path, spf->r, "%s", ce->name);
- strbuf_addf(&submodule_git_dir, "%s/.git", submodule_path.buf);
strbuf_addf(&submodule_prefix, "%s%s/", spf->prefix, ce->name);
- git_dir = read_gitfile(submodule_git_dir.buf);
- if (!git_dir)
- git_dir = submodule_git_dir.buf;
- if (is_directory(git_dir)) {
+ repo = get_submodule_repo_for(spf->r, submodule);
+ if (repo) {
child_process_init(cp);
- cp->dir = strbuf_detach(&submodule_path, NULL);
- prepare_submodule_repo_env(&cp->env_array);
+ prepare_submodule_repo_env_in_gitdir(&cp->env_array);
+ cp->dir = xstrdup(repo->gitdir);
cp->git_cmd = 1;
if (!spf->quiet)
strbuf_addf(err, "Fetching submodule %s%s\n",
@@ -1307,10 +1331,11 @@ static int get_next_submodule(struct child_process *cp,
argv_array_push(&cp->args, default_argv);
argv_array_push(&cp->args, "--submodule-prefix");
argv_array_push(&cp->args, submodule_prefix.buf);
+
+ repo_clear(repo);
+ free(repo);
ret = 1;
}
- strbuf_release(&submodule_path);
- strbuf_release(&submodule_git_dir);
strbuf_release(&submodule_prefix);
if (ret) {
spf->count++;
diff --git a/t/t5526-fetch-submodules.sh b/t/t5526-fetch-submodules.sh
index 6c2f9b2ba2..42692219a1 100755
--- a/t/t5526-fetch-submodules.sh
+++ b/t/t5526-fetch-submodules.sh
@@ -566,7 +566,12 @@ test_expect_success 'fetching submodule into a broken repository' '
test_must_fail git -C dst status &&
test_must_fail git -C dst diff &&
- test_must_fail git -C dst fetch --recurse-submodules
+
+ # git-fetch cannot find the git directory of the submodule,
+ # so it will do nothing, successfully, as it cannot distinguish between
+ # this broken submodule and a submodule that was just set active but
+ # not cloned yet
+ git -C dst fetch --recurse-submodules
'
test_expect_success "fetch new commits when submodule got renamed" '
--
2.19.0
^ permalink raw reply related [relevance 26%]
* [PATCH 6/9] repository: repo_submodule_init to take a submodule struct
2018-10-16 18:13 9% [PATCH 0/9] Resending sb/submodule-recursive-fetch-gets-the-tip Stefan Beller
` (3 preceding siblings ...)
2018-10-16 18:13 16% ` [PATCH 5/9] submodule.c: do not copy around submodule list Stefan Beller
@ 2018-10-16 18:13 24% ` Stefan Beller
2018-10-16 18:13 26% ` [PATCH 7/9] submodule: fetch in submodules git directory instead of in worktree Stefan Beller
` (4 subsequent siblings)
9 siblings, 0 replies; 200+ results
From: Stefan Beller @ 2018-10-16 18:13 UTC (permalink / raw)
To: gitster; +Cc: git, Stefan Beller
When constructing a struct repository for a submodule for some revision
of the superproject where the submodule is not contained in the index,
it may not be present in the working tree currently either. In that
siutation giving a 'path' argument is not useful. Upgrade the
repo_submodule_init function to take a struct submodule instead.
While we are at it, overhaul the repo_submodule_init function by renaming
the submodule repository struct, which is to be initialized, to a name
that is not confused with the struct submodule as easily.
Also move its documentation into the header file.
Signed-off-by: Stefan Beller <sbeller@google.com>
---
builtin/grep.c | 17 +++++++-----
builtin/ls-files.c | 12 +++++----
repository.c | 27 ++++++++------------
repository.h | 11 ++++++--
t/helper/test-submodule-nested-repo-config.c | 8 +++---
5 files changed, 41 insertions(+), 34 deletions(-)
diff --git a/builtin/grep.c b/builtin/grep.c
index 7da8fef31a..ba7634258a 100644
--- a/builtin/grep.c
+++ b/builtin/grep.c
@@ -418,7 +418,10 @@ static int grep_submodule(struct grep_opt *opt, struct repository *superproject,
const struct object_id *oid,
const char *filename, const char *path)
{
- struct repository submodule;
+ struct repository subrepo;
+ const struct submodule *sub = submodule_from_path(superproject,
+ &null_oid, path);
+
int hit;
/*
@@ -434,12 +437,12 @@ static int grep_submodule(struct grep_opt *opt, struct repository *superproject,
return 0;
}
- if (repo_submodule_init(&submodule, superproject, path)) {
+ if (repo_submodule_init(&subrepo, superproject, sub)) {
grep_read_unlock();
return 0;
}
- repo_read_gitmodules(&submodule);
+ repo_read_gitmodules(&subrepo);
/*
* NEEDSWORK: This adds the submodule's object directory to the list of
@@ -451,7 +454,7 @@ static int grep_submodule(struct grep_opt *opt, struct repository *superproject,
* store is no longer global and instead is a member of the repository
* object.
*/
- add_to_alternates_memory(submodule.objects->objectdir);
+ add_to_alternates_memory(subrepo.objects->objectdir);
grep_read_unlock();
if (oid) {
@@ -476,14 +479,14 @@ static int grep_submodule(struct grep_opt *opt, struct repository *superproject,
init_tree_desc(&tree, data, size);
hit = grep_tree(opt, pathspec, &tree, &base, base.len,
- object->type == OBJ_COMMIT, &submodule);
+ object->type == OBJ_COMMIT, &subrepo);
strbuf_release(&base);
free(data);
} else {
- hit = grep_cache(opt, &submodule, pathspec, 1);
+ hit = grep_cache(opt, &subrepo, pathspec, 1);
}
- repo_clear(&submodule);
+ repo_clear(&subrepo);
return hit;
}
diff --git a/builtin/ls-files.c b/builtin/ls-files.c
index 7f9919a362..4d1649c1b3 100644
--- a/builtin/ls-files.c
+++ b/builtin/ls-files.c
@@ -206,17 +206,19 @@ static void show_files(struct repository *repo, struct dir_struct *dir);
static void show_submodule(struct repository *superproject,
struct dir_struct *dir, const char *path)
{
- struct repository submodule;
+ struct repository subrepo;
+ const struct submodule *sub = submodule_from_path(superproject,
+ &null_oid, path);
- if (repo_submodule_init(&submodule, superproject, path))
+ if (repo_submodule_init(&subrepo, superproject, sub))
return;
- if (repo_read_index(&submodule) < 0)
+ if (repo_read_index(&subrepo) < 0)
die("index file corrupt");
- show_files(&submodule, dir);
+ show_files(&subrepo, dir);
- repo_clear(&submodule);
+ repo_clear(&subrepo);
}
static void show_ce(struct repository *repo, struct dir_struct *dir,
diff --git a/repository.c b/repository.c
index 5dd1486718..aabe64ee5d 100644
--- a/repository.c
+++ b/repository.c
@@ -166,30 +166,23 @@ int repo_init(struct repository *repo,
return -1;
}
-/*
- * Initialize 'submodule' as the submodule given by 'path' in parent repository
- * 'superproject'.
- * Return 0 upon success and a non-zero value upon failure.
- */
-int repo_submodule_init(struct repository *submodule,
+int repo_submodule_init(struct repository *subrepo,
struct repository *superproject,
- const char *path)
+ const struct submodule *sub)
{
- const struct submodule *sub;
struct strbuf gitdir = STRBUF_INIT;
struct strbuf worktree = STRBUF_INIT;
int ret = 0;
- sub = submodule_from_path(superproject, &null_oid, path);
if (!sub) {
ret = -1;
goto out;
}
- strbuf_repo_worktree_path(&gitdir, superproject, "%s/.git", path);
- strbuf_repo_worktree_path(&worktree, superproject, "%s", path);
+ strbuf_repo_worktree_path(&gitdir, superproject, "%s/.git", sub->path);
+ strbuf_repo_worktree_path(&worktree, superproject, "%s", sub->path);
- if (repo_init(submodule, gitdir.buf, worktree.buf)) {
+ if (repo_init(subrepo, gitdir.buf, worktree.buf)) {
/*
* If initilization fails then it may be due to the submodule
* not being populated in the superproject's worktree. Instead
@@ -201,16 +194,16 @@ int repo_submodule_init(struct repository *submodule,
strbuf_repo_git_path(&gitdir, superproject,
"modules/%s", sub->name);
- if (repo_init(submodule, gitdir.buf, NULL)) {
+ if (repo_init(subrepo, gitdir.buf, NULL)) {
ret = -1;
goto out;
}
}
- submodule->submodule_prefix = xstrfmt("%s%s/",
- superproject->submodule_prefix ?
- superproject->submodule_prefix :
- "", path);
+ subrepo->submodule_prefix = xstrfmt("%s%s/",
+ superproject->submodule_prefix ?
+ superproject->submodule_prefix :
+ "", sub->path);
out:
strbuf_release(&gitdir);
diff --git a/repository.h b/repository.h
index 9f16c42c1e..a9c7a5baa5 100644
--- a/repository.h
+++ b/repository.h
@@ -116,9 +116,16 @@ void repo_set_worktree(struct repository *repo, const char *path);
void repo_set_hash_algo(struct repository *repo, int algo);
void initialize_the_repository(void);
int repo_init(struct repository *r, const char *gitdir, const char *worktree);
-int repo_submodule_init(struct repository *submodule,
+
+/*
+ * Initialize the repository 'subrepo' as the submodule given by the
+ * struct submodule 'sub' in parent repository 'superproject'.
+ * Return 0 upon success and a non-zero value upon failure.
+ */
+struct submodule;
+int repo_submodule_init(struct repository *subrepo,
struct repository *superproject,
- const char *path);
+ const struct submodule *sub);
void repo_clear(struct repository *repo);
/*
diff --git a/t/helper/test-submodule-nested-repo-config.c b/t/helper/test-submodule-nested-repo-config.c
index a31e2a9bea..bc97929bbc 100644
--- a/t/helper/test-submodule-nested-repo-config.c
+++ b/t/helper/test-submodule-nested-repo-config.c
@@ -10,19 +10,21 @@ static void die_usage(int argc, const char **argv, const char *msg)
int cmd__submodule_nested_repo_config(int argc, const char **argv)
{
- struct repository submodule;
+ struct repository subrepo;
+ const struct submodule *sub;
if (argc < 3)
die_usage(argc, argv, "Wrong number of arguments.");
setup_git_directory();
- if (repo_submodule_init(&submodule, the_repository, argv[1])) {
+ sub = submodule_from_path(the_repository, &null_oid, argv[1]);
+ if (repo_submodule_init(&subrepo, the_repository, sub)) {
die_usage(argc, argv, "Submodule not found.");
}
/* Read the config of _child_ submodules. */
- print_config_from_gitmodules(&submodule, argv[2]);
+ print_config_from_gitmodules(&subrepo, argv[2]);
submodule_free(the_repository);
--
2.19.0
^ permalink raw reply related [relevance 24%]
* [PATCH 5/9] submodule.c: do not copy around submodule list
2018-10-16 18:13 9% [PATCH 0/9] Resending sb/submodule-recursive-fetch-gets-the-tip Stefan Beller
` (2 preceding siblings ...)
2018-10-16 18:13 16% ` [PATCH 4/9] submodule.c: move global changed_submodule_names into fetch submodule struct Stefan Beller
@ 2018-10-16 18:13 16% ` Stefan Beller
2018-10-16 18:13 24% ` [PATCH 6/9] repository: repo_submodule_init to take a submodule struct Stefan Beller
` (5 subsequent siblings)
9 siblings, 0 replies; 200+ results
From: Stefan Beller @ 2018-10-16 18:13 UTC (permalink / raw)
To: gitster; +Cc: git, Stefan Beller
'calculate_changed_submodule_paths' uses a local list to compute the
changed submodules, and then produces the result by copying appropriate
items into the result list.
Instead use the result list directly and prune items afterwards
using string_list_remove_empty_items.
By doing so we'll have access to the util pointer for longer that
contains the commits that we need to fetch, which will be
useful in a later patch.
Signed-off-by: Stefan Beller <sbeller@google.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
---
submodule.c | 19 ++++++++++---------
1 file changed, 10 insertions(+), 9 deletions(-)
diff --git a/submodule.c b/submodule.c
index 6b4cee82bf..cbefe5f54d 100644
--- a/submodule.c
+++ b/submodule.c
@@ -1142,8 +1142,7 @@ static void calculate_changed_submodule_paths(
struct submodule_parallel_fetch *spf)
{
struct argv_array argv = ARGV_ARRAY_INIT;
- struct string_list changed_submodules = STRING_LIST_INIT_DUP;
- const struct string_list_item *name;
+ struct string_list_item *name;
/* No need to check if there are no submodules configured */
if (!submodule_from_path(the_repository, NULL, NULL))
@@ -1160,9 +1159,9 @@ static void calculate_changed_submodule_paths(
* Collect all submodules (whether checked out or not) for which new
* commits have been recorded upstream in "changed_submodule_names".
*/
- collect_changed_submodules(&changed_submodules, &argv);
+ collect_changed_submodules(&spf->changed_submodule_names, &argv);
- for_each_string_list_item(name, &changed_submodules) {
+ for_each_string_list_item(name, &spf->changed_submodule_names) {
struct oid_array *commits = name->util;
const struct submodule *submodule;
const char *path = NULL;
@@ -1176,12 +1175,14 @@ static void calculate_changed_submodule_paths(
if (!path)
continue;
- if (!submodule_has_commits(path, commits))
- string_list_append(&spf->changed_submodule_names,
- name->string);
+ if (submodule_has_commits(path, commits)) {
+ oid_array_clear(commits);
+ *name->string = '\0';
+ }
}
- free_submodules_oids(&changed_submodules);
+ string_list_remove_empty_items(&spf->changed_submodule_names, 1);
+
argv_array_clear(&argv);
oid_array_clear(&ref_tips_before_fetch);
oid_array_clear(&ref_tips_after_fetch);
@@ -1377,7 +1378,7 @@ int fetch_populated_submodules(struct repository *r,
argv_array_clear(&spf.args);
out:
- string_list_clear(&spf.changed_submodule_names, 1);
+ free_submodules_oids(&spf.changed_submodule_names);
return spf.result;
}
--
2.19.0
^ permalink raw reply related [relevance 16%]
* [PATCH 4/9] submodule.c: move global changed_submodule_names into fetch submodule struct
2018-10-16 18:13 9% [PATCH 0/9] Resending sb/submodule-recursive-fetch-gets-the-tip Stefan Beller
2018-10-16 18:13 24% ` [PATCH 2/9] submodule.c: fix indentation Stefan Beller
2018-10-16 18:13 17% ` [PATCH 3/9] submodule.c: sort changed_submodule_names before searching it Stefan Beller
@ 2018-10-16 18:13 16% ` Stefan Beller
2018-10-16 18:13 16% ` [PATCH 5/9] submodule.c: do not copy around submodule list Stefan Beller
` (6 subsequent siblings)
9 siblings, 1 reply; 200+ results
From: Stefan Beller @ 2018-10-16 18:13 UTC (permalink / raw)
To: gitster; +Cc: git, Stefan Beller
The `changed_submodule_names` are only used for fetching, so let's make it
part of the struct that is passed around for fetching submodules.
Signed-off-by: Stefan Beller <sbeller@google.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
---
submodule.c | 42 +++++++++++++++++++++++-------------------
1 file changed, 23 insertions(+), 19 deletions(-)
diff --git a/submodule.c b/submodule.c
index 9fbfcfcfe1..6b4cee82bf 100644
--- a/submodule.c
+++ b/submodule.c
@@ -24,7 +24,7 @@
#include "object-store.h"
static int config_update_recurse_submodules = RECURSE_SUBMODULES_OFF;
-static struct string_list changed_submodule_names = STRING_LIST_INIT_DUP;
+
static int initialized_fetch_ref_tips;
static struct oid_array ref_tips_before_fetch;
static struct oid_array ref_tips_after_fetch;
@@ -1124,7 +1124,22 @@ void check_for_new_submodule_commits(struct object_id *oid)
oid_array_append(&ref_tips_after_fetch, oid);
}
-static void calculate_changed_submodule_paths(void)
+struct submodule_parallel_fetch {
+ int count;
+ struct argv_array args;
+ struct repository *r;
+ const char *prefix;
+ int command_line_option;
+ int default_option;
+ int quiet;
+ int result;
+
+ struct string_list changed_submodule_names;
+};
+#define SPF_INIT {0, ARGV_ARRAY_INIT, NULL, NULL, 0, 0, 0, 0, STRING_LIST_INIT_DUP }
+
+static void calculate_changed_submodule_paths(
+ struct submodule_parallel_fetch *spf)
{
struct argv_array argv = ARGV_ARRAY_INIT;
struct string_list changed_submodules = STRING_LIST_INIT_DUP;
@@ -1162,7 +1177,8 @@ static void calculate_changed_submodule_paths(void)
continue;
if (!submodule_has_commits(path, commits))
- string_list_append(&changed_submodule_names, name->string);
+ string_list_append(&spf->changed_submodule_names,
+ name->string);
}
free_submodules_oids(&changed_submodules);
@@ -1199,18 +1215,6 @@ int submodule_touches_in_range(struct object_id *excl_oid,
return ret;
}
-struct submodule_parallel_fetch {
- int count;
- struct argv_array args;
- struct repository *r;
- const char *prefix;
- int command_line_option;
- int default_option;
- int quiet;
- int result;
-};
-#define SPF_INIT {0, ARGV_ARRAY_INIT, NULL, NULL, 0, 0, 0, 0}
-
static int get_fetch_recurse_config(const struct submodule *submodule,
struct submodule_parallel_fetch *spf)
{
@@ -1271,7 +1275,7 @@ static int get_next_submodule(struct child_process *cp,
case RECURSE_SUBMODULES_ON_DEMAND:
if (!submodule ||
!string_list_lookup(
- &changed_submodule_names,
+ &spf->changed_submodule_names,
submodule->name))
continue;
default_argv = "on-demand";
@@ -1363,8 +1367,8 @@ int fetch_populated_submodules(struct repository *r,
argv_array_push(&spf.args, "--recurse-submodules-default");
/* default value, "--submodule-prefix" and its value are added later */
- calculate_changed_submodule_paths();
- string_list_sort(&changed_submodule_names);
+ calculate_changed_submodule_paths(&spf);
+ string_list_sort(&spf.changed_submodule_names);
run_processes_parallel(max_parallel_jobs,
get_next_submodule,
fetch_start_failure,
@@ -1373,7 +1377,7 @@ int fetch_populated_submodules(struct repository *r,
argv_array_clear(&spf.args);
out:
- string_list_clear(&changed_submodule_names, 1);
+ string_list_clear(&spf.changed_submodule_names, 1);
return spf.result;
}
--
2.19.0
^ permalink raw reply related [relevance 16%]
* [PATCH 3/9] submodule.c: sort changed_submodule_names before searching it
2018-10-16 18:13 9% [PATCH 0/9] Resending sb/submodule-recursive-fetch-gets-the-tip Stefan Beller
2018-10-16 18:13 24% ` [PATCH 2/9] submodule.c: fix indentation Stefan Beller
@ 2018-10-16 18:13 17% ` Stefan Beller
2018-10-16 18:13 16% ` [PATCH 4/9] submodule.c: move global changed_submodule_names into fetch submodule struct Stefan Beller
` (7 subsequent siblings)
9 siblings, 0 replies; 200+ results
From: Stefan Beller @ 2018-10-16 18:13 UTC (permalink / raw)
To: gitster; +Cc: git, Stefan Beller
We can string_list_insert() to maintain sorted-ness of the
list as we find new items, or we can string_list_append() to
build an unsorted list and sort it at the end just once.
As we do not rely on the sortedness while building the
list, we pick the "append and sort at the end" as it
has better worst case execution times.
Signed-off-by: Stefan Beller <sbeller@google.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
---
submodule.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/submodule.c b/submodule.c
index e145ebbb16..9fbfcfcfe1 100644
--- a/submodule.c
+++ b/submodule.c
@@ -1270,7 +1270,7 @@ static int get_next_submodule(struct child_process *cp,
case RECURSE_SUBMODULES_DEFAULT:
case RECURSE_SUBMODULES_ON_DEMAND:
if (!submodule ||
- !unsorted_string_list_lookup(
+ !string_list_lookup(
&changed_submodule_names,
submodule->name))
continue;
@@ -1364,6 +1364,7 @@ int fetch_populated_submodules(struct repository *r,
/* default value, "--submodule-prefix" and its value are added later */
calculate_changed_submodule_paths();
+ string_list_sort(&changed_submodule_names);
run_processes_parallel(max_parallel_jobs,
get_next_submodule,
fetch_start_failure,
--
2.19.0
^ permalink raw reply related [relevance 17%]
* [PATCH 2/9] submodule.c: fix indentation
2018-10-16 18:13 9% [PATCH 0/9] Resending sb/submodule-recursive-fetch-gets-the-tip Stefan Beller
@ 2018-10-16 18:13 24% ` Stefan Beller
2018-10-16 18:13 17% ` [PATCH 3/9] submodule.c: sort changed_submodule_names before searching it Stefan Beller
` (8 subsequent siblings)
9 siblings, 0 replies; 200+ results
From: Stefan Beller @ 2018-10-16 18:13 UTC (permalink / raw)
To: gitster; +Cc: git, Stefan Beller
The submodule subsystem is really bad at staying within 80 characters.
Fix it while we are here.
Signed-off-by: Stefan Beller <sbeller@google.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
---
submodule.c | 9 ++++++---
1 file changed, 6 insertions(+), 3 deletions(-)
diff --git a/submodule.c b/submodule.c
index 2b7082b2db..e145ebbb16 100644
--- a/submodule.c
+++ b/submodule.c
@@ -1258,7 +1258,8 @@ static int get_next_submodule(struct child_process *cp,
if (!submodule) {
const char *name = default_name_or_path(ce->name);
if (name) {
- default_submodule.path = default_submodule.name = name;
+ default_submodule.path = name;
+ default_submodule.name = name;
submodule = &default_submodule;
}
}
@@ -1268,8 +1269,10 @@ static int get_next_submodule(struct child_process *cp,
default:
case RECURSE_SUBMODULES_DEFAULT:
case RECURSE_SUBMODULES_ON_DEMAND:
- if (!submodule || !unsorted_string_list_lookup(&changed_submodule_names,
- submodule->name))
+ if (!submodule ||
+ !unsorted_string_list_lookup(
+ &changed_submodule_names,
+ submodule->name))
continue;
default_argv = "on-demand";
break;
--
2.19.0
^ permalink raw reply related [relevance 24%]
* [PATCH 0/9] Resending sb/submodule-recursive-fetch-gets-the-tip
@ 2018-10-16 18:13 9% Stefan Beller
2018-10-16 18:13 24% ` [PATCH 2/9] submodule.c: fix indentation Stefan Beller
` (9 more replies)
0 siblings, 10 replies; 200+ results
From: Stefan Beller @ 2018-10-16 18:13 UTC (permalink / raw)
To: gitster; +Cc: git, Stefan Beller
This is based on ao/submodule-wo-gitmodules-checked-out.
This resends origin/sb/submodule-recursive-fetch-gets-the-tip, resolving
the issues pointed out via origin/xxx/sb-submodule-recursive-fetch-gets-the-tip-in-pu
by basing this series on origin/ao/submodule-wo-gitmodules-checked-out
A range-diff below shows how picking a different base changed the patches,
apart from that no further adjustments have been made.
Thanks,
Stefan
Stefan Beller (9):
sha1-array: provide oid_array_filter
submodule.c: fix indentation
submodule.c: sort changed_submodule_names before searching it
submodule.c: move global changed_submodule_names into fetch submodule
struct
submodule.c: do not copy around submodule list
repository: repo_submodule_init to take a submodule struct
submodule: fetch in submodules git directory instead of in worktree
fetch: retry fetching submodules if needed objects were not fetched
builtin/fetch: check for submodule updates for non branch fetches
Documentation/technical/api-oid-array.txt | 5 +
builtin/fetch.c | 14 +-
builtin/grep.c | 17 +-
builtin/ls-files.c | 12 +-
repository.c | 27 +-
repository.h | 11 +-
sha1-array.c | 17 ++
sha1-array.h | 3 +
submodule.c | 275 +++++++++++++++----
t/helper/test-submodule-nested-repo-config.c | 8 +-
t/t5526-fetch-submodules.sh | 23 +-
11 files changed, 315 insertions(+), 97 deletions(-)
git range-diff origin/xxx/sb-submodule-recursive-fetch-gets-the-tip-in-pu...
[...]
585: ac1f98a0df < -: ---------- doc: move git-rev-parse from porcelain to plumbing
586: 7cf1a0fbef = 1: a035323c49 sha1-array: provide oid_array_filter
587: 01077381d0 = 2: 30ed20b4f0 submodule.c: fix indentation
588: 4b0cdf5899 = 3: cd590ea88d submodule.c: sort changed_submodule_names before searching it
589: 78e5099ecc ! 4: ce959811ba submodule.c: move global changed_submodule_names into fetch submodule struct
@@ -12,7 +12,7 @@
--- a/submodule.c
+++ b/submodule.c
@@
- #include "commit-reach.h"
+ #include "object-store.h"
static int config_update_recurse_submodules = RECURSE_SUBMODULES_OFF;
-static struct string_list changed_submodule_names = STRING_LIST_INIT_DUP;
590: d813f18bb3 = 5: 151f9a8ad4 submodule.c: do not copy around submodule list
591: a077d63af7 ! 6: 3a97743fa2 repository: repo_submodule_init to take a submodule struct
@@ -15,7 +15,6 @@
Also move its documentation into the header file.
Signed-off-by: Stefan Beller <sbeller@google.com>
- Signed-off-by: Junio C Hamano <gitster@pobox.com>
diff --git a/builtin/grep.c b/builtin/grep.c
--- a/builtin/grep.c
@@ -31,12 +30,16 @@
+
int hit;
- if (!is_submodule_active(superproject, path))
+ /*
+@@
return 0;
+ }
-- if (repo_submodule_init(&submodule, superproject, path))
-+ if (repo_submodule_init(&subrepo, superproject, sub))
+- if (repo_submodule_init(&submodule, superproject, path)) {
++ if (repo_submodule_init(&subrepo, superproject, sub)) {
+ grep_read_unlock();
return 0;
+ }
- repo_read_gitmodules(&submodule);
+ repo_read_gitmodules(&subrepo);
@@ -44,9 +47,9 @@
/*
* NEEDSWORK: This adds the submodule's object directory to the list of
@@
+ * store is no longer global and instead is a member of the repository
* object.
*/
- grep_read_lock();
- add_to_alternates_memory(submodule.objects->objectdir);
+ add_to_alternates_memory(subrepo.objects->objectdir);
grep_read_unlock();
@@ -100,19 +103,6 @@
static void show_ce(struct repository *repo, struct dir_struct *dir,
-diff --git a/builtin/submodule--helper.c b/builtin/submodule--helper.c
---- a/builtin/submodule--helper.c
-+++ b/builtin/submodule--helper.c
-@@
- if (!sub)
- BUG("We could get the submodule handle before?");
-
-- if (repo_submodule_init(&subrepo, the_repository, path))
-+ if (repo_submodule_init(&subrepo, the_repository, sub))
- die(_("could not get a repository handle for submodule '%s'"), path);
-
- if (!repo_config_get_string(&subrepo, "core.worktree", &cw)) {
-
diff --git a/repository.c b/repository.c
--- a/repository.c
+++ b/repository.c
@@ -197,3 +187,32 @@
void repo_clear(struct repository *repo);
/*
+
+diff --git a/t/helper/test-submodule-nested-repo-config.c b/t/helper/test-submodule-nested-repo-config.c
+--- a/t/helper/test-submodule-nested-repo-config.c
++++ b/t/helper/test-submodule-nested-repo-config.c
+@@
+
+ int cmd__submodule_nested_repo_config(int argc, const char **argv)
+ {
+- struct repository submodule;
++ struct repository subrepo;
++ const struct submodule *sub;
+
+ if (argc < 3)
+ die_usage(argc, argv, "Wrong number of arguments.");
+
+ setup_git_directory();
+
+- if (repo_submodule_init(&submodule, the_repository, argv[1])) {
++ sub = submodule_from_path(the_repository, &null_oid, argv[1]);
++ if (repo_submodule_init(&subrepo, the_repository, sub)) {
+ die_usage(argc, argv, "Submodule not found.");
+ }
+
+ /* Read the config of _child_ submodules. */
+- print_config_from_gitmodules(&submodule, argv[2]);
++ print_config_from_gitmodules(&subrepo, argv[2]);
+
+ submodule_free(the_repository);
+
592: 780f6c1a92 = 7: 4e8ad61f8d submodule: fetch in submodules git directory instead of in worktree
593: a530535912 = 8: 24bac00db7 fetch: retry fetching submodules if needed objects were not fetched
594: a72bde3a8a = 9: e031182e44 builtin/fetch: check for submodule updates for non branch fetches
[...]
^ permalink raw reply [relevance 9%]
* [PATCH] submodule helper: convert relative URL to absolute URL if needed
2018-10-16 5:15 2% ` Junio C Hamano
@ 2018-10-16 17:27 21% ` Stefan Beller
2018-10-16 21:05 4% ` Jonathan Nieder
0 siblings, 1 reply; 200+ results
From: Stefan Beller @ 2018-10-16 17:27 UTC (permalink / raw)
To: gitster; +Cc: git, jrnieder, jungjw, sbeller
The submodule helper update_clone called by "git submodule update",
clones submodules if needed. As submodules used to have the URL indicating
if they were active, the step to resolve relative URLs was done in the
"submodule init" step. Nowadays submodules can be configured active without
calling an explicit init, e.g. via configuring submodule.active.
When trying to obtain submodules that are set active this way, we'll
fallback to the URL found in the .gitmodules, which may be relative to the
superproject, but we do not resolve it, yet:
git clone https://gerrit.googlesource.com/gerrit
cd gerrit && grep url .gitmodules
url = ../plugins/codemirror-editor
...
git config submodule.active .
git submodule update
fatal: repository '../plugins/codemirror-editor' does not exist
fatal: clone of '../plugins/codemirror-editor' into submodule path '/tmp/gerrit/plugins/codemirror-editor' failed
Failed to clone 'plugins/codemirror-editor'. Retry scheduled
[...]
fatal: clone of '../plugins/codemirror-editor' into submodule path '/tmp/gerrit/plugins/codemirror-editor' failed
Failed to clone 'plugins/codemirror-editor' a second time, aborting
[...]
To resolve the issue, factor out the function that resolves the relative
URLs in "git submodule init" (in the submodule helper in the init_submodule
function) and call it at the appropriate place in the update_clone helper.
Reported-by: Jaewoong Jung <jungjw@google.com>
Signed-off-by: Stefan Beller <sbeller@google.com>
---
builtin/submodule--helper.c | 51 ++++++++++++++++++++++++-------------
t/t7400-submodule-basic.sh | 24 +++++++++++++++++
2 files changed, 58 insertions(+), 17 deletions(-)
diff --git a/builtin/submodule--helper.c b/builtin/submodule--helper.c
index f6fb8991f3..13c2e4b556 100644
--- a/builtin/submodule--helper.c
+++ b/builtin/submodule--helper.c
@@ -584,6 +584,26 @@ static int module_foreach(int argc, const char **argv, const char *prefix)
return 0;
}
+static char *compute_submodule_clone_url(const char *rel_url)
+{
+ char *remoteurl, *relurl;
+ char *remote = get_default_remote();
+ struct strbuf remotesb = STRBUF_INIT;
+
+ strbuf_addf(&remotesb, "remote.%s.url", remote);
+ if (git_config_get_string(remotesb.buf, &remoteurl)) {
+ warning(_("could not look up configuration '%s'. Assuming this repository is its own authoritative upstream."), remotesb.buf);
+ remoteurl = xgetcwd();
+ }
+ relurl = relative_url(remoteurl, rel_url, NULL);
+
+ free(remote);
+ free(remoteurl);
+ strbuf_release(&remotesb);
+
+ return relurl;
+}
+
struct init_cb {
const char *prefix;
unsigned int flags;
@@ -634,21 +654,9 @@ static void init_submodule(const char *path, const char *prefix,
/* Possibly a url relative to parent */
if (starts_with_dot_dot_slash(url) ||
starts_with_dot_slash(url)) {
- char *remoteurl, *relurl;
- char *remote = get_default_remote();
- struct strbuf remotesb = STRBUF_INIT;
- strbuf_addf(&remotesb, "remote.%s.url", remote);
- free(remote);
-
- if (git_config_get_string(remotesb.buf, &remoteurl)) {
- warning(_("could not lookup configuration '%s'. Assuming this repository is its own authoritative upstream."), remotesb.buf);
- remoteurl = xgetcwd();
- }
- relurl = relative_url(remoteurl, url, NULL);
- strbuf_release(&remotesb);
- free(remoteurl);
- free(url);
- url = relurl;
+ char *oldurl = url;
+ url = compute_submodule_clone_url(oldurl);
+ free(oldurl);
}
if (git_config_set_gently(sb.buf, url))
@@ -1514,6 +1522,7 @@ static int prepare_to_clone_next_submodule(const struct cache_entry *ce,
struct strbuf sb = STRBUF_INIT;
const char *displaypath = NULL;
int needs_cloning = 0;
+ int need_free_url = 0;
if (ce_stage(ce)) {
if (suc->recursive_prefix)
@@ -1562,8 +1571,14 @@ static int prepare_to_clone_next_submodule(const struct cache_entry *ce,
strbuf_reset(&sb);
strbuf_addf(&sb, "submodule.%s.url", sub->name);
- if (repo_config_get_string_const(the_repository, sb.buf, &url))
- url = sub->url;
+ if (repo_config_get_string_const(the_repository, sb.buf, &url)) {
+ if (starts_with_dot_slash(sub->url) ||
+ starts_with_dot_dot_slash(sub->url)) {
+ url = compute_submodule_clone_url(sub->url);
+ need_free_url = 1;
+ } else
+ url = sub->url;
+ }
strbuf_reset(&sb);
strbuf_addf(&sb, "%s/.git", ce->name);
@@ -1608,6 +1623,8 @@ static int prepare_to_clone_next_submodule(const struct cache_entry *ce,
cleanup:
strbuf_reset(&displaypath_sb);
strbuf_reset(&sb);
+ if (need_free_url)
+ free((void*)url);
return needs_cloning;
}
diff --git a/t/t7400-submodule-basic.sh b/t/t7400-submodule-basic.sh
index c0ffc1022a..76a7cb0af7 100755
--- a/t/t7400-submodule-basic.sh
+++ b/t/t7400-submodule-basic.sh
@@ -1224,6 +1224,30 @@ test_expect_success 'submodule update and setting submodule.<name>.active' '
test_cmp expect actual
'
+test_expect_success 'clone active submodule without submodule url set' '
+ test_when_finished "rm -rf test/test" &&
+ mkdir test &&
+ # another dir breaks accidental relative paths still being correct
+ git clone file://"$pwd"/multisuper test/test &&
+ (
+ cd test/test &&
+ git config submodule.active "." &&
+
+ # do not pass --init flag, as the submodule is already active:
+ git submodule update &&
+ git submodule status >actual_raw &&
+
+ cut -c 1,43- actual_raw >actual &&
+ cat >expect <<-\EOF &&
+ sub0 (test2)
+ sub1 (test2)
+ sub2 (test2)
+ sub3 (test2)
+ EOF
+ test_cmp expect actual
+ )
+'
+
test_expect_success 'clone --recurse-submodules with a pathspec works' '
test_when_finished "rm -rf multisuper_clone" &&
cat >expected <<-\EOF &&
--
2.19.0
^ permalink raw reply related [relevance 21%]
* Re: [PATCH] submodule helper: convert relative URL to absolute URL if needed
2018-10-16 0:33 4% ` Jonathan Nieder
@ 2018-10-16 5:15 2% ` Junio C Hamano
2018-10-16 17:27 21% ` Stefan Beller
0 siblings, 1 reply; 200+ results
From: Junio C Hamano @ 2018-10-16 5:15 UTC (permalink / raw)
To: Jonathan Nieder; +Cc: Stefan Beller, git, Jaewoong Jung
Jonathan Nieder <jrnieder@gmail.com> writes:
>> git config submodule.active .
>> git submodule update
>> fatal: repository '../plugins/codemirror-editor' does not exist
>> fatal: clone of '../plugins/codemirror-editor' into submodule path '/tmp/gerrit/plugins/codemirror-editor' failed
>> Failed to clone 'plugins/codemirror-editor'. Retry scheduled
>> [...]
>> fatal: clone of '../plugins/codemirror-editor' into submodule path '/tmp/gerrit/plugins/codemirror-editor' failed
>> Failed to clone 'plugins/codemirror-editor' a second time, aborting
>> [...]
>
> Thanks.
>
> "git log" and other tools have the ability to rewrap lines and will get
> confused by this transcript. Can you indent it to unconfuse them?
>
>> Signed-off-by: Stefan Beller <sbeller@google.com>
>
> Please also credit the bug reporter:
>
> Reported-by: Jaewoong Jung <jungjw@google.com>
>
> ...
> nit: inconsistent vertical whitespace (extra blank line?)
> ...
> I think this would be easier to read with all the release commands
> together:
> ...
All good points.
^ permalink raw reply [relevance 2%]
* Re: [PATCH] submodule helper: convert relative URL to absolute URL if needed
2018-10-16 0:19 21% ` Stefan Beller
@ 2018-10-16 0:33 4% ` Jonathan Nieder
2018-10-16 5:15 2% ` Junio C Hamano
0 siblings, 1 reply; 200+ results
From: Jonathan Nieder @ 2018-10-16 0:33 UTC (permalink / raw)
To: Stefan Beller; +Cc: gitster, git, Jaewoong Jung
Hi,
Stefan Beller wrote:
> The submodule helper update_clone called by "git submodule update",
> clones submodules if needed. As submodules used to have the URL indicating
> if they were active, the step to resolve relative URLs was done in the
> "submodule init" step. Nowadays submodules can be configured active without
> calling an explicit init, e.g. via configuring submodule.active.
>
> When trying to obtain submodules that are set active this way, we'll
> fallback to the URL found in the .gitmodules, which may be relative to the
> superproject, but we do not resolve it, yet:
>
> git clone https://gerrit.googlesource.com/gerrit
> cd gerrit && grep url .gitmodules
> url = ../plugins/codemirror-editor
> ...
> git config submodule.active .
> git submodule update
> fatal: repository '../plugins/codemirror-editor' does not exist
> fatal: clone of '../plugins/codemirror-editor' into submodule path '/tmp/gerrit/plugins/codemirror-editor' failed
> Failed to clone 'plugins/codemirror-editor'. Retry scheduled
> [...]
> fatal: clone of '../plugins/codemirror-editor' into submodule path '/tmp/gerrit/plugins/codemirror-editor' failed
> Failed to clone 'plugins/codemirror-editor' a second time, aborting
> [...]
Thanks.
"git log" and other tools have the ability to rewrap lines and will get
confused by this transcript. Can you indent it to unconfuse them?
> Signed-off-by: Stefan Beller <sbeller@google.com>
Please also credit the bug reporter:
Reported-by: Jaewoong Jung <jungjw@google.com>
[...]
> --- a/builtin/submodule--helper.c
> +++ b/builtin/submodule--helper.c
> @@ -584,6 +584,27 @@ static int module_foreach(int argc, const char **argv, const char *prefix)
> return 0;
> }
>
> +
nit: inconsistent vertical whitespace (extra blank line?)
> +static char *compute_submodule_clone_url(const char *rel_url)
> +{
> + char *remoteurl, *relurl;
> + char *remote = get_default_remote();
> + struct strbuf remotesb = STRBUF_INIT;
> +
> + strbuf_addf(&remotesb, "remote.%s.url", remote);
> + free(remote);
> +
> + if (git_config_get_string(remotesb.buf, &remoteurl)) {
> + warning(_("could not lookup configuration '%s'. Assuming this repository is its own authoritative upstream."), remotesb.buf);
> + remoteurl = xgetcwd();
> + }
> + relurl = relative_url(remoteurl, rel_url, NULL);
> + strbuf_release(&remotesb);
> + free(remoteurl);
> +
> + return relurl;
> +}
I think this would be easier to read with all the release commands
together:
...
free(remote);
free(remoteurl);
strbuf_release(&remotesb);
return relurl;
[...]
> @@ -634,21 +655,9 @@ static void init_submodule(const char *path, const char *prefix,
[...]
> - relurl = relative_url(remoteurl, url, NULL);
> - strbuf_release(&remotesb);
> - free(remoteurl);
> - free(url);
> - url = relurl;
> + char *to_free = url;
> + url = compute_submodule_clone_url(url);
> + free(to_free);
I still think this would be easier to read with a style that makes
the ownership passing clearer:
char *oldurl = url;
url = compute_submodule_clone_url(oldurl);
free(oldurl);
With whatever subset of the above tweaks makes sense,
Reviewed-by: Jonathan Nieder <jrnieder@gmail.com>
Thanks.
^ permalink raw reply [relevance 4%]
* [PATCH] submodule helper: convert relative URL to absolute URL if needed
2018-10-12 21:53 22% [PATCH] submodule helper: convert relative URL to absolute URL if needed Stefan Beller
2018-10-12 22:27 4% ` Jonathan Nieder
@ 2018-10-16 0:19 21% ` Stefan Beller
2018-10-16 0:33 4% ` Jonathan Nieder
1 sibling, 1 reply; 200+ results
From: Stefan Beller @ 2018-10-16 0:19 UTC (permalink / raw)
To: sbeller, gitster; +Cc: git, jrnieder
The submodule helper update_clone called by "git submodule update",
clones submodules if needed. As submodules used to have the URL indicating
if they were active, the step to resolve relative URLs was done in the
"submodule init" step. Nowadays submodules can be configured active without
calling an explicit init, e.g. via configuring submodule.active.
When trying to obtain submodules that are set active this way, we'll
fallback to the URL found in the .gitmodules, which may be relative to the
superproject, but we do not resolve it, yet:
git clone https://gerrit.googlesource.com/gerrit
cd gerrit && grep url .gitmodules
url = ../plugins/codemirror-editor
...
git config submodule.active .
git submodule update
fatal: repository '../plugins/codemirror-editor' does not exist
fatal: clone of '../plugins/codemirror-editor' into submodule path '/tmp/gerrit/plugins/codemirror-editor' failed
Failed to clone 'plugins/codemirror-editor'. Retry scheduled
[...]
fatal: clone of '../plugins/codemirror-editor' into submodule path '/tmp/gerrit/plugins/codemirror-editor' failed
Failed to clone 'plugins/codemirror-editor' a second time, aborting
[...]
To resolve the issue, factor out the function that resolves the relative
URLs in "git submodule init" (in the submodule helper in the init_submodule
function) and call it at the appropriate place in the update_clone helper.
Signed-off-by: Stefan Beller <sbeller@google.com>
---
Jonathan wrote:
> Ah, this is moved code. I should have used --color-moved. ;-)
Yes, any cleanup should go on top.
I extended the commit message and made sure not to leak memory.
When rerolling origin/xxx/sb-submodule-recursive-fetch-gets-the-tip-in-pu,
there will be conflicts with this series, but I can work with that.
builtin/submodule--helper.c | 48 ++++++++++++++++++++++++-------------
t/t7400-submodule-basic.sh | 24 +++++++++++++++++++
2 files changed, 55 insertions(+), 17 deletions(-)
diff --git a/builtin/submodule--helper.c b/builtin/submodule--helper.c
index f6fb8991f3..03f5e0d03e 100644
--- a/builtin/submodule--helper.c
+++ b/builtin/submodule--helper.c
@@ -584,6 +584,27 @@ static int module_foreach(int argc, const char **argv, const char *prefix)
return 0;
}
+
+static char *compute_submodule_clone_url(const char *rel_url)
+{
+ char *remoteurl, *relurl;
+ char *remote = get_default_remote();
+ struct strbuf remotesb = STRBUF_INIT;
+
+ strbuf_addf(&remotesb, "remote.%s.url", remote);
+ free(remote);
+
+ if (git_config_get_string(remotesb.buf, &remoteurl)) {
+ warning(_("could not lookup configuration '%s'. Assuming this repository is its own authoritative upstream."), remotesb.buf);
+ remoteurl = xgetcwd();
+ }
+ relurl = relative_url(remoteurl, rel_url, NULL);
+ strbuf_release(&remotesb);
+ free(remoteurl);
+
+ return relurl;
+}
+
struct init_cb {
const char *prefix;
unsigned int flags;
@@ -634,21 +655,9 @@ static void init_submodule(const char *path, const char *prefix,
/* Possibly a url relative to parent */
if (starts_with_dot_dot_slash(url) ||
starts_with_dot_slash(url)) {
- char *remoteurl, *relurl;
- char *remote = get_default_remote();
- struct strbuf remotesb = STRBUF_INIT;
- strbuf_addf(&remotesb, "remote.%s.url", remote);
- free(remote);
-
- if (git_config_get_string(remotesb.buf, &remoteurl)) {
- warning(_("could not lookup configuration '%s'. Assuming this repository is its own authoritative upstream."), remotesb.buf);
- remoteurl = xgetcwd();
- }
- relurl = relative_url(remoteurl, url, NULL);
- strbuf_release(&remotesb);
- free(remoteurl);
- free(url);
- url = relurl;
+ char *to_free = url;
+ url = compute_submodule_clone_url(url);
+ free(to_free);
}
if (git_config_set_gently(sb.buf, url))
@@ -1562,8 +1571,13 @@ static int prepare_to_clone_next_submodule(const struct cache_entry *ce,
strbuf_reset(&sb);
strbuf_addf(&sb, "submodule.%s.url", sub->name);
- if (repo_config_get_string_const(the_repository, sb.buf, &url))
- url = sub->url;
+ if (repo_config_get_string_const(the_repository, sb.buf, &url)) {
+ if (starts_with_dot_slash(sub->url) ||
+ starts_with_dot_dot_slash(sub->url))
+ url = compute_submodule_clone_url(sub->url);
+ else
+ url = sub->url;
+ }
strbuf_reset(&sb);
strbuf_addf(&sb, "%s/.git", ce->name);
diff --git a/t/t7400-submodule-basic.sh b/t/t7400-submodule-basic.sh
index c0ffc1022a..76a7cb0af7 100755
--- a/t/t7400-submodule-basic.sh
+++ b/t/t7400-submodule-basic.sh
@@ -1224,6 +1224,30 @@ test_expect_success 'submodule update and setting submodule.<name>.active' '
test_cmp expect actual
'
+test_expect_success 'clone active submodule without submodule url set' '
+ test_when_finished "rm -rf test/test" &&
+ mkdir test &&
+ # another dir breaks accidental relative paths still being correct
+ git clone file://"$pwd"/multisuper test/test &&
+ (
+ cd test/test &&
+ git config submodule.active "." &&
+
+ # do not pass --init flag, as the submodule is already active:
+ git submodule update &&
+ git submodule status >actual_raw &&
+
+ cut -c 1,43- actual_raw >actual &&
+ cat >expect <<-\EOF &&
+ sub0 (test2)
+ sub1 (test2)
+ sub2 (test2)
+ sub3 (test2)
+ EOF
+ test_cmp expect actual
+ )
+'
+
test_expect_success 'clone --recurse-submodules with a pathspec works' '
test_when_finished "rm -rf multisuper_clone" &&
cat >expected <<-\EOF &&
--
2.19.0
^ permalink raw reply related [relevance 21%]
* Re: What's cooking in git.git (Oct 2018, #02; Sat, 13)
2018-10-13 1:03 5% ` Junio C Hamano
@ 2018-10-15 17:32 5% ` Stefan Beller
0 siblings, 0 replies; 200+ results
From: Stefan Beller @ 2018-10-15 17:32 UTC (permalink / raw)
To: Junio C Hamano; +Cc: git
On Fri, Oct 12, 2018 at 6:03 PM Junio C Hamano <gitster@pobox.com> wrote:
>
> Stefan Beller <sbeller@google.com> writes:
>
> >> * sb/submodule-recursive-fetch-gets-the-tip (2018-10-11) 9 commits
> >> . builtin/fetch: check for submodule updates for non branch fetches
> >> . fetch: retry fetching submodules if needed objects were not fetched
> >> . submodule: fetch in submodules git directory instead of in worktree
> >> . repository: repo_submodule_init to take a submodule struct
> >> . submodule.c: do not copy around submodule list
> >> . submodule.c: move global changed_submodule_names into fetch submodule struct
> >> . submodule.c: sort changed_submodule_names before searching it
> >> . submodule.c: fix indentation
> >> . sha1-array: provide oid_array_filter
> >>
> >> "git fetch --recurse-submodules" may not fetch the necessary commit
> >> that is bound to the superproject, which is getting corrected.
> >>
> >> Ejected for now, as it has fallouts in places like t/helper/.
> >
> > This is the first time I hear about that, I'll look into that.
> > The tipmost commit there is also shoddy, I'll redo that.
>
> This is the first time I saw the breakage with this series, but I
> would not be suprised, as this was rerolled recently. Who knows
> what got changed in this series and in other topics---any new
> interaction can arise and that is a normal part of distributed
> development.
I performed the same merge last week, and the range-diff indicates,
that your version of next was further ahead than mine.
The breakage itself comes from
t/helper/test-submodule-nested-repo-config.c: In function
‘cmd__submodule_nested_repo_config’:
t/helper/test-submodule-nested-repo-config.c:20:54: warning: passing
argument 3 of ‘repo_submodule_init’ from incompatible pointer type
[-Wincompatible-pointer-types]
if (repo_submodule_init(&submodule, the_repository, argv[1])) {
^~~~
In file included from ./cache.h:17:0,
from ./submodule-config.h:4,
from t/helper/test-submodule-nested-repo-config.c:2:
./repository.h:126:5: note: expected ‘const struct submodule *’ but
argument is of type ‘const char *’
int repo_submodule_init(struct repository *subrepo,
^~~~~~~~~~~~~~~~~~~
which is introduced by
commit c369da44610acc5e56213b8784d4250ae619fdb9
(origin/ao/submodule-wo-gitmodules-checked-out)
Author: Antonio Ospite <ao2@ao2.it>
Date: 2018-10-05 15:06
t/helper: add test-submodule-nested-repo-config
Add a test tool to exercise config_from_gitmodules(), in particular for
the case of nested submodules.
Add also a test to document that reading the submoudles config of nested
submodules does not work yet when the .gitmodules file is not in the
working tree but it still in the index.
This is because the git API does not always make it possible access the
object store of an arbitrary repository (see get_oid() usage in
config_from_gitmodules()).
When this git limitation gets fixed the aforementioned use case will be
supported too.
Signed-off-by: Antonio Ospite <ao2@ao2.it>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
My resend will take that into account, building on Antonios series.
Thanks,
Stefan
^ permalink raw reply [relevance 5%]
* Git Test Coverage Report (Monday, Oct 15)
@ 2018-10-15 16:24 1% Derrick Stolee
0 siblings, 0 replies; 200+ results
From: Derrick Stolee @ 2018-10-15 16:24 UTC (permalink / raw)
To: Git List
In an effort to ensure new code is reasonably covered by the test suite,
we now have contrib/coverage-diff.sh to combine the gcov output from
'make coverage-test ; make coverage-report' with the output from 'git
diff A B' to discover _new_ lines of code that are not covered. This
report ignores lines including "BUG(".
This report takes the output of these results after running on four
branches:
pu: 78bfaf7bc1efe6892fe4dbedf9ed80f9dd48146c
jch: d67e018e0f57ebbb4fa0354a58a0a6d47e25a948
next: 152ad8e3369ac77026886a2910e3a407c281df35
master: 5a0cc8aca797dbd7d2be3b67458ff880ed45cddf
maint: 04861070400d3ca9d487bd0d736ca818b9ffe371
I ran the test suite on each of these branches on an Ubuntu Linux VM,
and I'm missing some dependencies (like apache, svn, and perforce) so
not all tests are run.
I submit this output without comment. I'm taking a look especially at my
own lines to see where coverage can be improved.
Thanks,
-Stolee
Uncovered code in 'pu' (12227c8) and not in 'jch' (22f2f0f)
-----------------------------------------------------------
builtin/blame.c
74e8221b52 builtin/blame.c 924) blame_date_width = sizeof("Thu Oct 19
16:00");
74e8221b52 builtin/blame.c 925) break;
builtin/branch.c
ba8ba9df26 builtin/branch.c 452) die(_("could not resolve HEAD"));
ba8ba9df26 builtin/branch.c 458) die(_("HEAD (%s) points outside of
refs/heads/"), refname);
builtin/grep.c
a8ace17bd4 builtin/grep.c 439) grep_read_unlock();
builtin/help.c
e6e76baaf4 builtin/help.c 429) if (!exclude_guides || alias[0] == '!') {
e6e76baaf4 builtin/help.c 430) printf_ln(_("'%s' is aliased to '%s'"),
cmd, alias);
e6e76baaf4 builtin/help.c 431) free(alias);
e6e76baaf4 builtin/help.c 432) exit(0);
e6e76baaf4 builtin/help.c 441) fprintf_ln(stderr, _("'%s' is aliased to
'%s'"), cmd, alias);
e6e76baaf4 builtin/help.c 442) count = split_cmdline(alias, &argv);
e6e76baaf4 builtin/help.c 443) if (count < 0)
e6e76baaf4 builtin/help.c 444) die(_("bad alias.%s string: %s"), cmd,
e6e76baaf4 builtin/help.c 446) free(argv);
e6e76baaf4 builtin/help.c 448) return alias;
builtin/rebase--interactive.c
6424061be4 builtin/rebase--interactive.c 23) return
error_errno(_("could not read '%s'."), todo_file);
6424061be4 builtin/rebase--interactive.c 28) return
error_errno(_("could not write '%s'"), todo_file);
7ccfac40bc builtin/rebase--interactive.c 43) return
error_errno(_("could not read '%s'."), todo_file);
7ccfac40bc builtin/rebase--interactive.c 46)
todo_list_release(&todo_list);
7ccfac40bc builtin/rebase--interactive.c 47) return error(_("unusable
todo list: '%s'"), todo_file);
9787d17d40 builtin/rebase--interactive.c 294) ret =
rearrange_squash_in_todo_file();
date.c
74e8221b52 113) die("Timestamp too large for this system: %"PRItime, time);
74e8221b52 216) if (tm->tm_mon == human_tm->tm_mon) {
74e8221b52 217) if (tm->tm_mday > human_tm->tm_mday) {
74e8221b52 219) } else if (tm->tm_mday == human_tm->tm_mday) {
74e8221b52 220) hide.date = hide.wday = 1;
74e8221b52 221) } else if (tm->tm_mday + 5 > human_tm->tm_mday) {
74e8221b52 223) hide.date = 1;
74e8221b52 231) gettimeofday(&now, NULL);
74e8221b52 232) show_date_relative(time, tz, &now, buf);
74e8221b52 233) return;
74e8221b52 246) hide.seconds = 1;
74e8221b52 247) hide.tz |= !hide.date;
74e8221b52 248) hide.wday = hide.time = !hide.year;
74e8221b52 262) strbuf_rtrim(buf);
74e8221b52 287) gettimeofday(&now, NULL);
74e8221b52 290) human_tz = local_time_tzoffset(now.tv_sec, &human_tm);
74e8221b52 886) static int auto_date_style(void)
74e8221b52 888) return (isatty(1) || pager_in_use()) ? DATE_HUMAN :
DATE_NORMAL;
74e8221b52 909) return DATE_HUMAN;
74e8221b52 911) return auto_date_style();
git.c
a9a60b94cc 322) fprintf_ln(stderr, _("'%s' is aliased to '%s'"),
rebase-interactive.c
b74a37a5a7 26) warning(_("unrecognized setting %s for option "
6424061be4 107) return error_errno(_("could not write '%s''"), todo_file);
6424061be4 110) return error(_("could not copy '%s' to '%s'."), todo_file,
b74a37a5a7 174) goto leave_check;
sequencer.c
b5d6062402 4425) strbuf_insert(buf, todo_list->items[insert].offset_in_buf +
b5d6062402 4437) int sequencer_add_exec_commands(const char *commands)
06d8136126 4444) return error_errno(_("could not read '%s'."), todo_file);
b5d6062402 4446) if (todo_list_parse_insn_buffer(todo_list.buf.buf,
&todo_list)) {
b5d6062402 4451) todo_list_add_exec_commands(&todo_list, commands);
b5d6062402 4452) res = write_message(todo_list.buf.buf,
todo_list.buf.len, todo_file, 0);
0cce4a2756 4453) todo_list_release(&todo_list);
b5d6062402 4455) return res;
b74a37a5a7 4515) goto out;
b74a37a5a7 4520) goto out;
b8dac44d10 4660) todo_list_release(&new_todo);
009173ed7b 4665) todo_list_release(&new_todo);
009173ed7b 4666) return error_errno(_("could not write '%s'"), todo_file);
9787d17d40 4859) int rearrange_squash_in_todo_file(void)
9787d17d40 4861) const char *todo_file = rebase_path_todo();
9787d17d40 4862) struct todo_list todo_list = TODO_LIST_INIT;
9787d17d40 4863) int res = 0;
9787d17d40 4865) if (strbuf_read_file_or_whine(&todo_list.buf,
todo_file) < 0)
9787d17d40 4866) return -1;
9787d17d40 4867) if (todo_list_parse_insn_buffer(todo_list.buf.buf,
&todo_list) < 0) {
9787d17d40 4868) todo_list_release(&todo_list);
9787d17d40 4869) return -1;
9787d17d40 4872) res = todo_list_rearrange_squash(&todo_list);
9787d17d40 4873) if (!res)
9787d17d40 4874) res = rewrite_file(todo_file, todo_list.buf.buf,
todo_list.buf.len);
9787d17d40 4876) todo_list_release(&todo_list);
submodule-config.c
bcbc780d14 739) return CONFIG_INVALID_KEY;
45f5ef3d77 754) warning(_("Could not update .gitmodules entry %s"), key);
wrapper.c
5efde212fc 70) die("Out of memory, malloc failed (tried to allocate %"
PRIuMAX " bytes)",
5efde212fc 73) error("Out of memory, malloc failed (tried to allocate
%" PRIuMAX " bytes)",
Commits introducing uncovered code:
Alban Gruin 009173ed7: sequencer: change complete_action() to use
the refactored functions
Alban Gruin 06d813612: sequencer: fix a call to error() in
transform_todo_file()
Alban Gruin 6424061be: rebase-interactive: rewrite edit_todo_list()
to handle the initial edit
Alban Gruin 7ccfac40b: rebase--interactive: move
transform_todo_file() to rebase--interactive.c
Alban Gruin 9787d17d4: sequencer: refactor rearrange_squash() to
work on a todo_list
Alban Gruin b5d606240: sequencer: refactor
sequencer_add_exec_commands() to work on a todo_list
Alban Gruin b74a37a5a: sequencer: refactor check_todo_list() to
work on a todo_list
Alban Gruin b8dac44d1: sequencer: refactor skip_unnecessary_picks()
to work on a todo_list
Antonio Ospite 45f5ef3d7: submodule: factor out a
config_set_in_gitmodules_file_gently function
Antonio Ospite a8ace17bd: submodule: support reading .gitmodules
when it's not in the working tree
Antonio Ospite bcbc780d1: submodule: add a
print_config_from_gitmodules() helper
Daniels Umanovskis ba8ba9df2: branch: introduce --show-current
display option
Liam Beguin 0cce4a275: rebase -i -x: add exec commands via the
rebase--helper
Linus Torvalds 74e8221b5: Add 'human' date format
Martin Koegler 5efde212f: zlib.c: use size_t for size
Rasmus Villemoes a9a60b94c: git.c: handle_alias: prepend alias info
when first argument is -h
Rasmus Villemoes e6e76baaf: help: redirect to aliased commands for
"git cmd --help"
Uncovered code in 'jch' (22f2f0f) and not in 'next' (152ad8e)
-------------------------------------------------------------
apply.c
eccb5a5f3d 4071) return get_oid_hex(p->old_oid_prefix, oid);
builtin/fetch.c
7bbc53a589 builtin/fetch.c 385) continue; /* can this happen??? */
builtin/rebase--interactive.c
53bbcfbde7 builtin/rebase--interactive2.c 24) return error(_("no HEAD?"));
53bbcfbde7 builtin/rebase--interactive2.c 51) return
error_errno(_("could not create temporary %s"), path_state_dir());
53bbcfbde7 builtin/rebase--interactive2.c 57) return
error_errno(_("could not mark as interactive"));
53bbcfbde7 builtin/rebase--interactive2.c 77) return -1;
53bbcfbde7 builtin/rebase--interactive2.c 81) return -1;
53bbcfbde7 builtin/rebase--interactive2.c 87) free(revisions);
53bbcfbde7 builtin/rebase--interactive2.c 88) free(shortrevisions);
53bbcfbde7 builtin/rebase--interactive2.c 90) return -1;
53bbcfbde7 builtin/rebase--interactive2.c 98) free(revisions);
53bbcfbde7 builtin/rebase--interactive2.c 99) free(shortrevisions);
53bbcfbde7 builtin/rebase--interactive2.c 101) return
error_errno(_("could not open %s"), rebase_path_todo());
53bbcfbde7 builtin/rebase--interactive2.c 106)
argv_array_push(&make_script_args, restrict_revision);
53bbcfbde7 builtin/rebase--interactive2.c 114) error(_("could not
generate todo list"));
53bbcfbde7 builtin/rebase--interactive2.c 206)
usage_with_options(builtin_rebase_interactive_usage, options);
53bbcfbde7 builtin/rebase--interactive2.c 220)
warning(_("--[no-]rebase-cousins has no effect without "
0af129b2ed builtin/rebase--interactive2.c 226) die(_("a base commit must
be provided with --upstream or --onto"));
34b47315d9 builtin/rebase--interactive.c 261) ret = rearrange_squash();
34b47315d9 builtin/rebase--interactive.c 262) break;
34b47315d9 builtin/rebase--interactive.c 264) ret =
sequencer_add_exec_commands(cmd);
34b47315d9 builtin/rebase--interactive.c 265) break;
builtin/rebase.c
55071ea248 61) strbuf_trim(&out);
55071ea248 62) ret = !strcmp("true", out.buf);
55071ea248 63) strbuf_release(&out);
002ee2fe68 115) die(_("%s requires an interactive rebase"), option);
f95736288a 148) return error_errno(_("could not read '%s'"), path);
f95736288a 162) return -1;
f95736288a 167) return error(_("could not get 'onto': '%s'"), buf.buf);
f95736288a 178) return -1;
f95736288a 179) } else if (read_one(state_dir_path("head", opts), &buf))
f95736288a 180) return -1;
f95736288a 182) return error(_("invalid orig-head: '%s'"), buf.buf);
f95736288a 186) return -1;
f95736288a 188) opts->flags &= ~REBASE_NO_QUIET;
73d51ed0a5 196) opts->signoff = 1;
73d51ed0a5 197) opts->flags |= REBASE_FORCE;
ead98c111b 204) return -1;
12026a412c 219) return -1;
ba1905a5fe 227) return -1;
ba1905a5fe 235) return -1;
6defce2b02 255) return error(_("Could not read '%s'"), path);
6defce2b02 271) res = error(_("Cannot store %s"), autostash.buf);
6defce2b02 275) return res;
bc24382c2b 373) argv_array_pushf(&child.args,
bc24382c2b 375) oid_to_hex(&opts->restrict_revision->object.oid));
ac7f467fef 507) struct strbuf dir = STRBUF_INIT;
6defce2b02 509) apply_autostash(opts);
ac7f467fef 510) strbuf_addstr(&dir, opts->state_dir);
ac7f467fef 511) remove_dir_recursively(&dir, 0);
ac7f467fef 512) strbuf_release(&dir);
ac7f467fef 513) die("Nothing to do");
ac7f467fef 543) return -1;
ac7f467fef 547) rollback_lock_file(&lock);
ac7f467fef 548) return error(_("could not determine HEAD revision"));
ac7f467fef 565) rollback_lock_file(&lock);
ac7f467fef 566) return error(_("could not read index"));
ac7f467fef 570) error(_("failed to find tree of %s"), oid_to_hex(oid));
ac7f467fef 571) rollback_lock_file(&lock);
ac7f467fef 572) free((void *)desc.buffer);
ac7f467fef 573) return -1;
ac7f467fef 586) ret = error(_("could not write index"));
ac7f467fef 590) return ret;
ac7f467fef 606) } else if (old_orig)
ac7f467fef 607) delete_ref(NULL, "ORIG_HEAD", old_orig, 0);
bff014dac7 635) opts->flags &= !REBASE_DIFFSTAT;
9a48a615b4 669) return 1;
9a48a615b4 685) return 0;
55071ea248 893) const char *path = mkpath("%s/git-legacy-rebase",
55071ea248 896) if (sane_execvp(path, (char **)argv) < 0)
55071ea248 897) die_errno(_("could not exec %s"), path);
0eabf4b95c 915) die(_("It looks like 'git am' is in progress. Cannot
rebase."));
f28d40d3a9 952) usage_with_options(builtin_rebase_usage,
f95736288a 972) die(_("Cannot read HEAD"));
f95736288a 976) die(_("could not read index"));
f95736288a 990) exit(1);
122420c295 1002) die(_("could not discard worktree changes"));
122420c295 1004) exit(1);
5e5d96197c 1015) exit(1);
5e5d96197c 1018) die(_("could not move back to %s"),
5a61494539 1028) die(_("could not remove '%s'"), options.state_dir);
c54dacb50e 1047) const char *last_slash = strrchr(options.state_dir, '/');
c54dacb50e 1048) const char *state_dir_base =
c54dacb50e 1049) last_slash ? last_slash + 1 : options.state_dir;
c54dacb50e 1050) const char *cmd_live_rebase =
c54dacb50e 1052) strbuf_reset(&buf);
c54dacb50e 1053) strbuf_addf(&buf, "rm -fr \"%s\"", options.state_dir);
c54dacb50e 1054) die(_("It seems that there is already a %s directory,
and\n"
53f9e5be94 1078) strbuf_addstr(&options.git_am_opt, " --ignore-date");
53f9e5be94 1079) options.flags |= REBASE_FORCE;
7998dbe1ec 1091) strbuf_addf(&options.git_am_opt, " -C%d", opt_c);
3c3588c7d3 1123) else if (strcmp("no-rebase-cousins", rebase_merges))
3c3588c7d3 1124) die(_("Unknown mode: %s"), rebase_merges);
ba1905a5fe 1146) die(_("--strategy requires --merge or --interactive"));
cda614e489 1164) strbuf_addstr(&options.git_format_patch_opt, "
--progress");
ac7f467fef 1173) options.state_dir = apply_dir();
ac7f467fef 1174) break;
ac7f467fef 1251) die(_("invalid upstream '%s'"), options.upstream_name);
9dba809a69 1257) die(_("Could not create new root commit"));
e65123a71d 1307) die(_("fatal: no such branch/commit '%s'"),
ac7f467fef 1315) die(_("No such ref: %s"), "HEAD");
ac7f467fef 1327) die(_("Could not resolve HEAD to a revision"));
e0333e5c63 1340) die(_("could not read index"));
6defce2b02 1367) die(_("Cannot autostash"));
6defce2b02 1370) die(_("Unexpected stash response: '%s'"),
6defce2b02 1376) die(_("Could not create directory for '%s'"),
6defce2b02 1382) die(_("could not reset --hard"));
e65123a71d 1426) ret = !!error(_("could not parse '%s'"),
e65123a71d 1428) goto cleanup;
e65123a71d 1437) ret = !!error(_("could not switch to "
1ed9c14ff2 1447) resolve_ref_unsafe("HEAD", 0, NULL, &flag))
1ed9c14ff2 1448) puts(_("HEAD is up to date."));
9a48a615b4 1457) resolve_ref_unsafe("HEAD", 0, NULL, &flag))
9a48a615b4 1458) puts(_("HEAD is up to date, rebase forced."));
builtin/repack.c
2f0c9e9a9b 239) die("repack: Expecting full hex object ID lines only
from pack-objects.");
2f0c9e9a9b 411) die("repack: Expecting full hex object ID lines only
from pack-objects.");
builtin/stash.c
3d5ec65ce8 builtin/stash--helper.c 126) error(_("'%s' is not a
stash-like commit"), revision);
3d5ec65ce8 builtin/stash--helper.c 127) free_stash_info(info);
3d5ec65ce8 builtin/stash--helper.c 128) exit(128);
3d5ec65ce8 builtin/stash--helper.c 161) free_stash_info(info);
3d5ec65ce8 builtin/stash--helper.c 162) fprintf_ln(stderr, _("No stash
entries found."));
3d5ec65ce8 builtin/stash--helper.c 163) return -1;
3d5ec65ce8 builtin/stash--helper.c 198) free_stash_info(info);
7005771171 builtin/stash--helper.c 225) return error(_("git stash clear
with parameters is "
3d5ec65ce8 builtin/stash--helper.c 241) return -1;
3d5ec65ce8 builtin/stash--helper.c 249) return -1;
3d5ec65ce8 builtin/stash--helper.c 262) return -1;
3d5ec65ce8 builtin/stash--helper.c 265) return error(_("unable to write
new index file"));
3d5ec65ce8 builtin/stash--helper.c 377) remove_path(stash_index_path.buf);
3d5ec65ce8 builtin/stash--helper.c 378) return -1;
3d5ec65ce8 builtin/stash--helper.c 405) return -1;
3d5ec65ce8 builtin/stash--helper.c 408) return error(_("cannot apply a
stash in the middle of a merge"));
3d5ec65ce8 builtin/stash--helper.c 418) strbuf_release(&out);
3d5ec65ce8 builtin/stash--helper.c 419) return error(_("Could not
generate diff %s^!."),
3d5ec65ce8 builtin/stash--helper.c 426) return error(_("Conflicts in
index."
3d5ec65ce8 builtin/stash--helper.c 432) return error(_("Could not save
index tree"));
3d5ec65ce8 builtin/stash--helper.c 439) return error(_("could not
restore untracked files from stash"));
3d5ec65ce8 builtin/stash--helper.c 470) return -1;
3d5ec65ce8 builtin/stash--helper.c 475) strbuf_release(&out);
3d5ec65ce8 builtin/stash--helper.c 480) strbuf_release(&out);
3d5ec65ce8 builtin/stash--helper.c 481) return -1;
7005771171 builtin/stash--helper.c 557) return error(_("%s: Could not
drop stash entry"),
5bf62a19c0 builtin/stash--helper.c 632) printf_ln(_("The stash entry is
kept in case "
104eb50d14 builtin/stash--helper.c 766) free_stash_info(&info);
193c3e3516 builtin/stash.c 767)
usage_with_options(git_stash_show_usage, options);
813904a0ce builtin/stash--helper.c 783) stash_msg = "Created via \"git
stash store\".";
813904a0ce builtin/stash--helper.c 789) if (!quiet) {
813904a0ce builtin/stash--helper.c 790) fprintf_ln(stderr, _("Cannot
update %s with %s"),
813904a0ce builtin/stash--helper.c 793) return -1;
813904a0ce builtin/stash--helper.c 817) if (!quiet)
813904a0ce builtin/stash--helper.c 818) fprintf_ln(stderr, _("\"git
stash store\" requires one "
813904a0ce builtin/stash--helper.c 820) return -1;
9f630e7480 builtin/stash--helper.c 902) return -1;
9f630e7480 builtin/stash--helper.c 962) ret = -1;
9f630e7480 builtin/stash--helper.c 963) goto done;
9f630e7480 builtin/stash--helper.c 968) ret = -1;
9f630e7480 builtin/stash--helper.c 969) goto done;
9f630e7480 builtin/stash--helper.c 974) ret = -1;
9f630e7480 builtin/stash--helper.c 975) goto done;
9f630e7480 builtin/stash--helper.c 1001) ret = -1;
9f630e7480 builtin/stash--helper.c 1002) goto done;
9f630e7480 builtin/stash--helper.c 1013) ret = -1;
9f630e7480 builtin/stash--helper.c 1014) goto done;
9f630e7480 builtin/stash--helper.c 1020) ret = -1;
9f630e7480 builtin/stash--helper.c 1021) goto done;
9f630e7480 builtin/stash--helper.c 1028) ret = -1;
9f630e7480 builtin/stash--helper.c 1029) goto done;
9f630e7480 builtin/stash--helper.c 1054) ret = -1;
9f630e7480 builtin/stash--helper.c 1055) goto done;
9f630e7480 builtin/stash--helper.c 1067) ret = -1;
9f630e7480 builtin/stash--helper.c 1068) goto done;
9f630e7480 builtin/stash--helper.c 1074) ret = -1;
9f630e7480 builtin/stash--helper.c 1075) goto done;
9f630e7480 builtin/stash--helper.c 1086) ret = -1;
9f630e7480 builtin/stash--helper.c 1087) goto done;
9f630e7480 builtin/stash--helper.c 1092) ret = -1;
9f630e7480 builtin/stash--helper.c 1093) goto done;
c2cc69f192 builtin/stash--helper.c 1128) fprintf_ln(stderr, _("You do
not have "
9f630e7480 builtin/stash--helper.c 1137) ret = 1;
9f630e7480 builtin/stash--helper.c 1138) goto done;
c2cc69f192 builtin/stash--helper.c 1154) if (!quiet)
c2cc69f192 builtin/stash--helper.c 1155) fprintf_ln(stderr, _("Cannot
save the current "
9f630e7480 builtin/stash--helper.c 1157) ret = -1;
9f630e7480 builtin/stash--helper.c 1158) goto done;
c2cc69f192 builtin/stash--helper.c 1163) if (!quiet)
c2cc69f192 builtin/stash--helper.c 1164) fprintf_ln(stderr, _("Cannot save "
9f630e7480 builtin/stash--helper.c 1166) ret = -1;
9f630e7480 builtin/stash--helper.c 1167) goto done;
c2cc69f192 builtin/stash--helper.c 1174) if (!quiet)
c2cc69f192 builtin/stash--helper.c 1175) fprintf_ln(stderr, _("Cannot
save the current "
9f630e7480 builtin/stash--helper.c 1177) goto done;
c2cc69f192 builtin/stash--helper.c 1183) if (!quiet)
c2cc69f192 builtin/stash--helper.c 1184) fprintf_ln(stderr, _("Cannot
save the current "
9f630e7480 builtin/stash--helper.c 1186) ret = -1;
9f630e7480 builtin/stash--helper.c 1187) goto done;
c2cc69f192 builtin/stash--helper.c 1213) if (!quiet)
c2cc69f192 builtin/stash--helper.c 1214) fprintf_ln(stderr, _("Cannot
record "
9f630e7480 builtin/stash--helper.c 1216) ret = -1;
9f630e7480 builtin/stash--helper.c 1217) goto done;
1a0f0409a7 builtin/stash--helper.c 1289) ret = -1;
1a0f0409a7 builtin/stash--helper.c 1290) goto done;
1a0f0409a7 builtin/stash--helper.c 1300) ret = -1;
c2cc69f192 builtin/stash--helper.c 1301) if (!quiet)
c2cc69f192 builtin/stash--helper.c 1302) fprintf_ln(stderr, _("Cannot
initialize stash"));
1a0f0409a7 builtin/stash--helper.c 1303) goto done;
1a0f0409a7 builtin/stash--helper.c 1313) ret = -1;
c2cc69f192 builtin/stash--helper.c 1314) if (!quiet)
c2cc69f192 builtin/stash--helper.c 1315) fprintf_ln(stderr, _("Cannot
save the current status"));
1a0f0409a7 builtin/stash--helper.c 1316) goto done;
1a0f0409a7 builtin/stash--helper.c 1333) ret = -1;
1a0f0409a7 builtin/stash--helper.c 1352) ret = -1;
1a0f0409a7 builtin/stash--helper.c 1353) goto done;
1a0f0409a7 builtin/stash--helper.c 1362) ret = -1;
1a0f0409a7 builtin/stash--helper.c 1363) goto done;
1a0f0409a7 builtin/stash--helper.c 1371) ret = -1;
1a0f0409a7 builtin/stash--helper.c 1380) ret = -1;
1a0f0409a7 builtin/stash--helper.c 1391) ret = -1;
1a0f0409a7 builtin/stash--helper.c 1392) goto done;
1a0f0409a7 builtin/stash--helper.c 1401) ret = -1;
1a0f0409a7 builtin/stash--helper.c 1402) goto done;
1a0f0409a7 builtin/stash--helper.c 1410) ret = -1;
1a0f0409a7 builtin/stash--helper.c 1436) ret = -1;
193c3e3516 builtin/stash.c 1568)
usage_msg_opt(xstrfmt(_("unknown subcommand: %s"), argv[0]),
193c3e3516 builtin/stash.c 1596) continue;
packfile.c
1127a98cce 117) return error("index file %s is too small", path);
1127a98cce 119) return error("empty data");
prio-queue.c
2d181390f3 94) return queue->array[queue->nr - 1].data;
rebase-interactive.c
64a43cbd5d 62) return error_errno(_("could not read '%s'."), todo_file);
64a43cbd5d 66) strbuf_release(&buf);
64a43cbd5d 67) return -1;
a9f5476fbc 75) return error_errno(_("could not read '%s'."), todo_file);
a9f5476fbc 79) strbuf_release(&buf);
a9f5476fbc 80) return -1;
64a43cbd5d 86) return -1;
revision.c
4943d28849 2931) return;
4943d28849 2934) return;
4943d28849 2937) c->object.flags |= UNINTERESTING;
4943d28849 2940) return;
4943d28849 2943) mark_parents_uninteresting(c);
4943d28849 2966) return;
4943d28849 2969) return;
4943d28849 2974) return;
4943d28849 3022) init_author_date_slab(&info->author_date);
4943d28849 3023) info->topo_queue.compare = compare_commits_by_author_date;
4943d28849 3024) info->topo_queue.cb_data = &info->author_date;
4943d28849 3025) break;
4943d28849 3038) continue;
4943d28849 3048) record_author_date(&info->author_date, c);
6c04ff3001 3086) if (!revs->ignore_missing_links)
6c04ff3001 3087) die("Failed to traverse parents of commit %s",
4943d28849 3088) oid_to_hex(&commit->object.oid));
4943d28849 3096) continue;
sequencer.c
65850686cf 2278) return;
65850686cf 2375) write_file(rebase_path_quiet(), "%s\n", quiet);
2c58483a59 3373) return error(_("could not checkout %s"), commit);
4df66c40b0 3387) return error(_("%s: not a valid OID"), orig_head);
71f82465b1 3407) fprintf(stderr, _("Stopped at HEAD\n"));
b97e187364 4771) return -1;
b97e187364 4774) return -1;
b97e187364 4780) return error_errno(_("could not read '%s'."), todo_file);
b97e187364 4783) todo_list_release(&todo_list);
b97e187364 4784) return error(_("unusable todo list: '%s'"), todo_file);
b97e187364 4803) todo_list_release(&todo_list);
b97e187364 4804) return -1;
b97e187364 4808) return error(_("could not copy '%s' to '%s'."), todo_file,
b97e187364 4812) return error(_("could not transform the todo list"));
b97e187364 4841) return error(_("could not transform the todo list"));
b97e187364 4844) return error(_("could not skip unnecessary pick
commands"));
b97e187364 4850) return -1;
split-index.c
e3d837989e 335) ce->ce_flags |= CE_UPDATE_IN_BASE;
strbuf.c
f95736288a 127) --sb->len;
Commits introducing uncovered code:
Alban Gruin 0af129b2e: rebase--interactive2: rewrite the submodes
of interactive rebase in C
Alban Gruin 2c58483a5: rebase -i: rewrite setup_reflog_action() in C
Alban Gruin 34b47315d: rebase -i: move rebase--helper modes to
rebase--interactive
Alban Gruin 4df66c40b: rebase -i: rewrite checkout_onto() in C
Alban Gruin 53bbcfbde: rebase -i: implement the main part of
interactive rebase as a builtin
Alban Gruin 64a43cbd5: rebase -i: rewrite the edit-todo
functionality in C
Alban Gruin 65850686c: rebase -i: rewrite write_basic_state() in C
Alban Gruin a9f5476fb: sequencer: refactor append_todo_help() to
write its message to a buffer
Alban Gruin b97e18736: rebase -i: rewrite complete_action() in C
brian m. carlson 2f0c9e9a9: builtin/repack: replace hard-coded
constants
brian m. carlson eccb5a5f3: apply: rename new_sha1_prefix and
old_sha1_prefix
Derrick Stolee 2d181390f: prio-queue: add 'peek' operation
Derrick Stolee 4943d2884: revision.c: refactor basic topo-order logic
Derrick Stolee 6c04ff300: revision.c: begin refactoring
--topo-order logic
Joel Teichroeb 3d5ec65ce: stash: convert apply to builtin
Joel Teichroeb 5bf62a19c: stash: convert pop to builtin
Joel Teichroeb 700577117: stash: convert drop and clear to builtin
Johannes Schindelin 71f82465b: rebase -i: introduce the 'break' command
Johannes Schindelin bc24382c2: builtin rebase: prepare for builtin
rebase -i
Josh Steadmon 1127a98cc: fuzz: add fuzz testing for packfile indices.
Junio C Hamano 7bbc53a58: fetch: replace string-list used as a
look-up table with a hashmap
Paul-Sebastian Ungureanu 104eb50d1: stash: convert show to builtin
Paul-Sebastian Ungureanu 193c3e351: stash: convert
`stash--helper.c` into `stash.c`
Paul-Sebastian Ungureanu 1a0f0409a: stash: convert push to builtin
Paul-Sebastian Ungureanu 813904a0c: stash: convert store to builtin
Paul-Sebastian Ungureanu 9f630e748: stash: convert create to builtin
Paul-Sebastian Ungureanu c2cc69f19: stash: make push -q quiet
Pratik Karki 002ee2fe6: builtin rebase: support `keep-empty` option
Pratik Karki 0eabf4b95: builtin rebase: stop if `git am` is in progress
Pratik Karki 12026a412: builtin rebase: support `--gpg-sign` option
Pratik Karki 122420c29: builtin rebase: support --skip
Pratik Karki 1ed9c14ff: builtin rebase: support --force-rebase
Pratik Karki 3c3588c7d: builtin rebase: support
--rebase-merges[=[no-]rebase-cousins]
Pratik Karki 53f9e5be9: builtin rebase: support `ignore-date` option
Pratik Karki 55071ea24: rebase: start implementing it as a builtin
Pratik Karki 5a6149453: builtin rebase: support --quit
Pratik Karki 5e5d96197: builtin rebase: support --abort
Pratik Karki 6defce2b0: builtin rebase: support `--autostash` option
Pratik Karki 73d51ed0a: builtin rebase: support --signoff
Pratik Karki 7998dbe1e: builtin rebase: support `-C` and
`--whitespace=<type>`
Pratik Karki 9a48a615b: builtin rebase: try to fast forward when
possible
Pratik Karki 9dba809a6: builtin rebase: support --root
Pratik Karki ac7f467fe: builtin/rebase: support running "git rebase
<upstream>"
Pratik Karki ba1905a5f: builtin rebase: add support for custom
merge strategies
Pratik Karki bff014dac: builtin rebase: support the `verbose` and
`diffstat` options
Pratik Karki c54dacb50: builtin rebase: start a new rebase only if
none is in progress
Pratik Karki cda614e48: builtin rebase: show progress when
connected to a terminal
Pratik Karki e0333e5c6: builtin rebase: require a clean worktree
Pratik Karki e65123a71: builtin rebase: support `git rebase
<upstream> <switch-to>`
Pratik Karki ead98c111: builtin rebase: support --rerere-autoupdate
Pratik Karki f28d40d3a: builtin rebase: support --onto
Pratik Karki f95736288: builtin rebase: support --continue
SZEDER Gábor e3d837989: split-index: don't compare cached data of
entries already marked for split index
Uncovered code in 'next' (152ad8e) and not in 'master' (5a0cc8a)
----------------------------------------------------------------
blame.c
a470beea39 113) !strcmp(r->index->cache[-1 - pos]->name, path))
a470beea39 272) int pos = index_name_pos(r->index, path, len);
a470beea39 274) mode = r->index->cache[pos]->ce_mode;
builtin/am.c
2abf350385 1414) repo_init_revisions(the_repository, &rev_info, NULL);
builtin/archive.c
e001fd3a50 builtin/archive.c 78) die(_("git archive: expected ACK/NAK,
got a flush packet"));
e001fd3a50 builtin/archive.c 80) if (starts_with(reader.line, "NACK "))
e001fd3a50 builtin/archive.c 81) die(_("git archive: NACK %s"),
reader.line + 5);
e001fd3a50 builtin/archive.c 82) if (starts_with(reader.line, "ERR "))
e001fd3a50 builtin/archive.c 83) die(_("remote error: %s"), reader.line
+ 4);
e001fd3a50 builtin/archive.c 84) die(_("git archive: protocol error"));
e001fd3a50 builtin/archive.c 89) die(_("git archive: expected a flush"));
fb19d32f05 builtin/archive.c 99) if (version != discover_version(&reader))
fb19d32f05 builtin/archive.c 100) die(_("git archive: received different
protocol versions in subsequent requests"));
builtin/gc.c
3029970275 builtin/gc.c 461) ret = error_errno(_("cannot stat '%s'"),
gc_log_path);
3029970275 builtin/gc.c 470) ret = error_errno(_("cannot read '%s'"),
gc_log_path);
fec2ed2187 builtin/gc.c 495) die(FAILED_RUN, pack_refs_cmd.argv[0]);
fec2ed2187 builtin/gc.c 498) die(FAILED_RUN, reflog.argv[0]);
3029970275 builtin/gc.c 585) exit(128);
fec2ed2187 builtin/gc.c 637) die(FAILED_RUN, repack.argv[0]);
fec2ed2187 builtin/gc.c 647) die(FAILED_RUN, prune.argv[0]);
fec2ed2187 builtin/gc.c 654) die(FAILED_RUN, prune_worktrees.argv[0]);
fec2ed2187 builtin/gc.c 658) die(FAILED_RUN, rerere.argv[0]);
builtin/pack-objects.c
2fa233a554 builtin/pack-objects.c 1512) hashcpy(base_oid.hash, base_sha1);
2fa233a554 builtin/pack-objects.c 1513) if
(!in_same_island(&delta->idx.oid, &base_oid))
2fa233a554 builtin/pack-objects.c 1514) return 0;
builtin/rev-list.c
7c0fe330d5 builtin/rev-list.c 227) die("unexpected missing %s object '%s'",
7c0fe330d5 builtin/rev-list.c 228) type_name(obj->type),
oid_to_hex(&obj->oid));
builtin/upload-archive.c
e001fd3a50 builtin/upload-archive.c 111) if (version == protocol_v0 ||
version == protocol_v1)
e001fd3a50 builtin/upload-archive.c 112) packet_write_fmt(1, "NACK
unable to spawn subprocess\n");
e001fd3a50 builtin/upload-archive.c 113) else if (version == protocol_v2)
e001fd3a50 builtin/upload-archive.c 114) error_clnt("unable to spawn
subprocess\n");
commit-graph.c
5cef295f28 67) return 0;
20fd6d5799 79) return 0;
config.c
c780b9cfe8 2298) return val;
c780b9cfe8 2301) if (is_bool)
c780b9cfe8 2302) return val ? 0 : 1;
c780b9cfe8 2304) return val;
diff.c
b78ea5fc35 4130) add_external_diff_name(o->repo, &argv, other, two);
help.c
26c7d06783 help.c 500) static int get_alias(const char *var,
const char *value, void *data)
26c7d06783 help.c 502) struct string_list *list = data;
26c7d06783 help.c 504) if (skip_prefix(var, "alias.", &var))
26c7d06783 help.c 505) string_list_append(list, var)->util =
xstrdup(value);
26c7d06783 help.c 507) return 0;
26c7d06783 help.c 530) printf("\n%s\n", _("Command aliases"));
26c7d06783 help.c 531) ALLOC_ARRAY(aliases, alias_list.nr + 1);
26c7d06783 help.c 532) for (i = 0; i < alias_list.nr; i++) {
26c7d06783 help.c 533) aliases[i].name = alias_list.items[i].string;
26c7d06783 help.c 534) aliases[i].help = alias_list.items[i].util;
26c7d06783 help.c 535) aliases[i].category = 1;
26c7d06783 help.c 537) aliases[alias_list.nr].name = NULL;
26c7d06783 help.c 538) print_command_list(aliases, 1, longest);
26c7d06783 help.c 539) free(aliases);
http-backend.c
fb19d32f05 646) argv[1] = ".";
fb19d32f05 647) argv[2] = NULL;
list-objects-filter-options.c
bc5975d24f 55) if (errbuf) {
bc5975d24f 56) strbuf_addstr(
bc5975d24f 60) return 1;
cc0b05a4cc 86) if (errbuf)
list-objects-filter.c
list-objects.c
f447a499db 197) ctx->show_object(obj, base->buf, ctx->show_data);
midx.c
e43d2dcce1 285) if (hasheq(oid.hash,
e43d2dcce1 286) p->bad_object_sha1 + the_hash_algo->rawsz * i))
oidset.c
8b2f8cbcb1 29) kh_del_oid(&set->set, pos);
8b2f8cbcb1 30) return 1;
preload-index.c
ae9af12287 73) struct progress_data *pd = p->progress;
ae9af12287 75) pthread_mutex_lock(&pd->mutex);
ae9af12287 76) pd->n += last_nr - nr;
ae9af12287 77) display_progress(pd->progress, pd->n);
ae9af12287 78) pthread_mutex_unlock(&pd->mutex);
ae9af12287 79) last_nr = nr;
ae9af12287 93) struct progress_data *pd = p->progress;
ae9af12287 95) pthread_mutex_lock(&pd->mutex);
ae9af12287 96) display_progress(pd->progress, pd->n + last_nr);
ae9af12287 97) pthread_mutex_unlock(&pd->mutex);
ae9af12287 128) pd.progress = start_delayed_progress(_("Refreshing
index"), index->cache_nr);
ae9af12287 129) pthread_mutex_init(&pd.mutex, NULL);
ae9af12287 140) p->progress = &pd;
read-cache.c
ae9af12287 1490) progress = start_delayed_progress(_("Refresh index"),
ae9af12287 1491) istate->cache_nr);
ae9af12287 1533) display_progress(progress, i);
ae9af12287 1566) display_progress(progress, istate->cache_nr);
ae9af12287 1567) stop_progress(&progress);
252d079cbd 1778) const unsigned char *cp = (const unsigned char *)name;
252d079cbd 1782) strip_len = decode_varint(&cp);
77ff1127a4 1783) if (previous_ce) {
77ff1127a4 1784) previous_len = previous_ce->ce_namelen;
77ff1127a4 1785) if (previous_len < strip_len)
252d079cbd 1786) die(_("malformed name field in the index, near path '%s'"),
77ff1127a4 1787) previous_ce->name);
77ff1127a4 1788) copy_len = previous_len - strip_len;
77ff1127a4 1790) copy_len = 0;
252d079cbd 1792) name = (const char *)cp;
252d079cbd 1798) len += copy_len;
252d079cbd 1819) if (copy_len)
252d079cbd 1820) memcpy(ce->name, previous_ce->name, copy_len);
252d079cbd 1821) memcpy(ce->name + copy_len, name, len + 1 - copy_len);
252d079cbd 1822) *ent_size = (name - ((char *)ondisk)) + len + 1 - copy_len;
abb4bb8384 1959) munmap((void *)p->mmap, p->mmap_size);
abb4bb8384 1960) die(_("index file corrupt"));
77ff1127a4 2001) mem_pool_init(&istate->ce_mem_pool,
77ff1127a4 2041) static void *load_cache_entries_thread(void *_data)
77ff1127a4 2043) struct load_cache_entries_thread_data *p = _data;
77ff1127a4 2047) for (i = p->ieot_start; i < p->ieot_start +
p->ieot_blocks; i++) {
77ff1127a4 2048) p->consumed += load_cache_entry_block(p->istate,
p->ce_mem_pool,
77ff1127a4 2049) p->offset, p->ieot->entries[i].nr, p->mmap,
p->ieot->entries[i].offset, NULL);
77ff1127a4 2050) p->offset += p->ieot->entries[i].nr;
77ff1127a4 2052) return NULL;
77ff1127a4 2055) static unsigned long load_cache_entries_threaded(struct
index_state *istate, const char *mmap, size_t mmap_size,
77ff1127a4 2060) unsigned long consumed = 0;
77ff1127a4 2063) if (istate->name_hash_initialized)
77ff1127a4 2066) mem_pool_init(&istate->ce_mem_pool, 0);
77ff1127a4 2069) if (nr_threads > ieot->nr)
77ff1127a4 2070) nr_threads = ieot->nr;
77ff1127a4 2071) data = xcalloc(nr_threads, sizeof(*data));
77ff1127a4 2073) offset = ieot_start = 0;
77ff1127a4 2074) ieot_blocks = DIV_ROUND_UP(ieot->nr, nr_threads);
77ff1127a4 2075) for (i = 0; i < nr_threads; i++) {
77ff1127a4 2076) struct load_cache_entries_thread_data *p = &data[i];
77ff1127a4 2079) if (ieot_start + ieot_blocks > ieot->nr)
77ff1127a4 2080) ieot_blocks = ieot->nr - ieot_start;
77ff1127a4 2082) p->istate = istate;
77ff1127a4 2083) p->offset = offset;
77ff1127a4 2084) p->mmap = mmap;
77ff1127a4 2085) p->ieot = ieot;
77ff1127a4 2086) p->ieot_start = ieot_start;
77ff1127a4 2087) p->ieot_blocks = ieot_blocks;
77ff1127a4 2090) nr = 0;
77ff1127a4 2091) for (j = p->ieot_start; j < p->ieot_start +
p->ieot_blocks; j++)
77ff1127a4 2092) nr += p->ieot->entries[j].nr;
77ff1127a4 2093) if (istate->version == 4) {
77ff1127a4 2094) mem_pool_init(&p->ce_mem_pool,
77ff1127a4 2097) mem_pool_init(&p->ce_mem_pool,
77ff1127a4 2101) err = pthread_create(&p->pthread, NULL,
load_cache_entries_thread, p);
77ff1127a4 2102) if (err)
77ff1127a4 2103) die(_("unable to create load_cache_entries thread:
%s"), strerror(err));
77ff1127a4 2106) for (j = 0; j < ieot_blocks; j++)
77ff1127a4 2107) offset += ieot->entries[ieot_start + j].nr;
77ff1127a4 2108) ieot_start += ieot_blocks;
77ff1127a4 2111) for (i = 0; i < nr_threads; i++) {
77ff1127a4 2112) struct load_cache_entries_thread_data *p = &data[i];
77ff1127a4 2114) err = pthread_join(p->pthread, NULL);
77ff1127a4 2115) if (err)
77ff1127a4 2116) die(_("unable to join load_cache_entries thread: %s"),
strerror(err));
77ff1127a4 2117) mem_pool_combine(istate->ce_mem_pool, p->ce_mem_pool);
77ff1127a4 2118) consumed += p->consumed;
77ff1127a4 2121) free(data);
77ff1127a4 2123) return consumed;
77ff1127a4 2192) nr_threads = cpus;
abb4bb8384 2196) extension_offset = read_eoie_extension(mmap, mmap_size);
abb4bb8384 2197) if (extension_offset) {
abb4bb8384 2200) p.src_offset = extension_offset;
abb4bb8384 2201) err = pthread_create(&p.pthread, NULL,
load_index_extensions, &p);
abb4bb8384 2202) if (err)
abb4bb8384 2203) die(_("unable to create load_index_extensions thread:
%s"), strerror(err));
abb4bb8384 2205) nr_threads--;
77ff1127a4 2214) ieot = read_ieot_extension(mmap, mmap_size,
extension_offset);
77ff1127a4 2217) src_offset += load_cache_entries_threaded(istate, mmap,
mmap_size, src_offset, nr_threads, ieot);
77ff1127a4 2218) free(ieot);
abb4bb8384 2232) int ret = pthread_join(p.pthread, NULL);
abb4bb8384 2233) if (ret)
abb4bb8384 2234) die(_("unable to join load_index_extensions thread:
%s"), strerror(ret));
3255089ada 2775) ieot_blocks = nr_threads;
77ff1127a4 2776) if (ieot_blocks > istate->cache_nr)
77ff1127a4 2777) ieot_blocks = istate->cache_nr;
3255089ada 2785) ieot = xcalloc(1, sizeof(struct index_entry_offset_table)
3255089ada 2786) + (ieot_blocks * sizeof(struct index_entry_offset)));
77ff1127a4 2787) ieot_entries = DIV_ROUND_UP(entries, ieot_blocks);
3255089ada 2794) free(ieot);
3b1d9e045e 2795) return -1;
3255089ada 2821) ieot->entries[ieot->nr].nr = nr;
3255089ada 2822) ieot->entries[ieot->nr].offset = offset;
3255089ada 2823) ieot->nr++;
3255089ada 2829) if (previous_name)
3255089ada 2830) previous_name->buf[0] = 0;
3255089ada 2831) nr = 0;
3255089ada 2832) offset = lseek(newfd, 0, SEEK_CUR);
3255089ada 2833) if (offset < 0) {
3255089ada 2834) free(ieot);
3255089ada 2835) return -1;
3255089ada 2837) offset += write_buffer_len;
3255089ada 2847) ieot->entries[ieot->nr].nr = nr;
3255089ada 2848) ieot->entries[ieot->nr].offset = offset;
3255089ada 2849) ieot->nr++;
3255089ada 2861) free(ieot);
3b1d9e045e 2862) return -1;
3255089ada 2876) struct strbuf sb = STRBUF_INIT;
3255089ada 2878) write_ieot_extension(&sb, ieot);
3255089ada 2879) err = write_index_ext_header(&c, &eoie_c, newfd,
CACHE_EXT_INDEXENTRYOFFSETTABLE, sb.len) < 0
3255089ada 2880) || ce_write(&c, newfd, sb.buf, sb.len) < 0;
3255089ada 2881) strbuf_release(&sb);
3255089ada 2882) free(ieot);
3255089ada 2883) if (err)
3255089ada 2884) return -1;
3b1d9e045e 3372) static size_t read_eoie_extension(const char *mmap,
size_t mmap_size)
3b1d9e045e 3390) if (mmap_size < sizeof(struct cache_header) +
EOIE_SIZE_WITH_HEADER + the_hash_algo->rawsz)
3b1d9e045e 3391) return 0;
3b1d9e045e 3394) index = eoie = mmap + mmap_size - EOIE_SIZE_WITH_HEADER
- the_hash_algo->rawsz;
3b1d9e045e 3395) if (CACHE_EXT(index) != CACHE_EXT_ENDOFINDEXENTRIES)
3b1d9e045e 3396) return 0;
3b1d9e045e 3397) index += sizeof(uint32_t);
3b1d9e045e 3400) extsize = get_be32(index);
3b1d9e045e 3401) if (extsize != EOIE_SIZE)
3b1d9e045e 3402) return 0;
3b1d9e045e 3403) index += sizeof(uint32_t);
3b1d9e045e 3409) offset = get_be32(index);
3b1d9e045e 3410) if (mmap + offset < mmap + sizeof(struct cache_header))
3b1d9e045e 3411) return 0;
3b1d9e045e 3412) if (mmap + offset >= eoie)
3b1d9e045e 3413) return 0;
3b1d9e045e 3414) index += sizeof(uint32_t);
3b1d9e045e 3425) src_offset = offset;
3b1d9e045e 3426) the_hash_algo->init_fn(&c);
3b1d9e045e 3427) while (src_offset < mmap_size - the_hash_algo->rawsz -
EOIE_SIZE_WITH_HEADER) {
3b1d9e045e 3435) memcpy(&extsize, mmap + src_offset + 4, 4);
3b1d9e045e 3436) extsize = ntohl(extsize);
3b1d9e045e 3439) if (src_offset + 8 + extsize < src_offset)
3b1d9e045e 3440) return 0;
3b1d9e045e 3442) the_hash_algo->update_fn(&c, mmap + src_offset, 8);
3b1d9e045e 3444) src_offset += 8;
3b1d9e045e 3445) src_offset += extsize;
3b1d9e045e 3447) the_hash_algo->final_fn(hash, &c);
3b1d9e045e 3448) if (!hasheq(hash, (const unsigned char *)index))
3b1d9e045e 3449) return 0;
3b1d9e045e 3452) if (src_offset != mmap_size - the_hash_algo->rawsz -
EOIE_SIZE_WITH_HEADER)
3b1d9e045e 3453) return 0;
3b1d9e045e 3455) return offset;
3255089ada 3475) static struct index_entry_offset_table
*read_ieot_extension(const char *mmap, size_t mmap_size, size_t offset)
3255089ada 3477) const char *index = NULL;
3255089ada 3483) if (!offset)
3255089ada 3484) return NULL;
3255089ada 3485) while (offset <= mmap_size -
the_hash_algo->rawsz - 8) {
3255089ada 3486) extsize = get_be32(mmap + offset + 4);
3255089ada 3487) if (CACHE_EXT((mmap + offset)) ==
CACHE_EXT_INDEXENTRYOFFSETTABLE) {
3255089ada 3488) index = mmap + offset + 4 + 4;
3255089ada 3489) break;
3255089ada 3491) offset += 8;
3255089ada 3492) offset += extsize;
3255089ada 3494) if (!index)
3255089ada 3495) return NULL;
3255089ada 3498) ext_version = get_be32(index);
3255089ada 3499) if (ext_version != IEOT_VERSION) {
3255089ada 3500) error("invalid IEOT version %d", ext_version);
3255089ada 3501) return NULL;
3255089ada 3503) index += sizeof(uint32_t);
3255089ada 3506) nr = (extsize - sizeof(uint32_t)) /
(sizeof(uint32_t) + sizeof(uint32_t));
3255089ada 3507) if (!nr) {
3255089ada 3508) error("invalid number of IEOT entries %d", nr);
3255089ada 3509) return NULL;
3255089ada 3511) ieot = xmalloc(sizeof(struct
index_entry_offset_table)
3255089ada 3512) + (nr * sizeof(struct index_entry_offset)));
3255089ada 3513) ieot->nr = nr;
3255089ada 3514) for (i = 0; i < nr; i++) {
3255089ada 3515) ieot->entries[i].offset = get_be32(index);
3255089ada 3516) index += sizeof(uint32_t);
3255089ada 3517) ieot->entries[i].nr = get_be32(index);
3255089ada 3518) index += sizeof(uint32_t);
3255089ada 3521) return ieot;
3255089ada 3524) static void write_ieot_extension(struct strbuf *sb,
struct index_entry_offset_table *ieot)
3255089ada 3530) put_be32(&buffer, IEOT_VERSION);
3255089ada 3531) strbuf_add(sb, &buffer, sizeof(uint32_t));
3255089ada 3534) for (i = 0; i < ieot->nr; i++) {
3255089ada 3537) put_be32(&buffer, ieot->entries[i].offset);
3255089ada 3538) strbuf_add(sb, &buffer, sizeof(uint32_t));
3255089ada 3541) put_be32(&buffer, ieot->entries[i].nr);
3255089ada 3542) strbuf_add(sb, &buffer, sizeof(uint32_t));
3255089ada 3544) }
refs.c
4a6067cda5 1405) return 0;
revision.c
2abf350385 1525) if (ce_path_match(istate, ce, &revs->prune_data, NULL)) {
2abf350385 1531) while ((i+1 < istate->cache_nr) &&
2abf350385 1532) ce_same_name(ce, istate->cache[i+1]))
transport-helper.c
fb19d32f05 643) if (!data->connect && !data->stateless_connect)
transport.c
wt-status.c
f3bd35fa0d 671) s->committable = 1;
73ba5d78b4 1953) if (s->state.rebase_in_progress ||
73ba5d78b4 1954) s->state.rebase_interactive_in_progress)
73ba5d78b4 1955) branch_name = s->state.onto;
73ba5d78b4 1956) else if (s->state.detached_from)
73ba5d78b4 1957) branch_name = s->state.detached_from;
Commits introducing uncovered code:
Ben Peart 3255089ad: ieot: add Index Entry Offset Table (IEOT)
extension
Ben Peart 3b1d9e045: eoie: add End of Index Entry (EOIE) extension
Ben Peart 77ff1127a: read-cache: load cache entries on worker threads
Ben Peart abb4bb838: read-cache: load cache extensions on a worker
thread
Ben Peart c780b9cfe: config: add new index.threads config setting
Derrick Stolee 20fd6d579: commit-graph: not compatible with grafts
Derrick Stolee 5cef295f2: commit-graph: not compatible with
uninitialized repo
Jeff King 2fa233a55: pack-objects: handle island check for
"external" delta base
Jeff King e43d2dcce: more oideq/hasheq conversions
Jonathan Nieder 302997027: gc: do not return error for prior errors
in daemonized mode
Jonathan Nieder fec2ed218: gc: exit with status 128 on failure
Josh Steadmon e001fd3a5: archive: implement protocol v2 archive command
Josh Steadmon fb19d32f0: archive: allow archive over HTTP(S) with
proto v2
Matthew DeVore 7c0fe330d: rev-list: handle missing tree objects
properly
Matthew DeVore bc5975d24: list-objects-filter: implement filter tree:0
Matthew DeVore cc0b05a4c: list-objects-filter-options: do not
over-strbuf_init
Matthew DeVore f447a499d: list-objects: store common func args in
struct
Nguyễn Thái Ngọc Duy 252d079cb: read-cache.c: optimize reading
index format v4
Nguyễn Thái Ngọc Duy 26c7d0678: help -a: improve and make --verbose
default
Nguyễn Thái Ngọc Duy 2abf35038: revision.c: remove implicit
dependency on the_index
Nguyễn Thái Ngọc Duy a470beea3: blame.c: rename "repo" argument to "r"
Nguyễn Thái Ngọc Duy ae9af1228: status: show progress bar if
refreshing the index takes too long
Nguyễn Thái Ngọc Duy b78ea5fc3: diff.c: reduce implicit dependency
on the_index
René Scharfe 8b2f8cbcb: oidset: use khash
Stefan Beller 4a6067cda: refs.c: migrate internal ref iteration to
pass thru repository argument
Stephen P. Smith 73ba5d78b: roll wt_status_state into wt_status and
populate in the collect phase
Stephen P. Smith f3bd35fa0: wt-status.c: set the committable flag
in the collect phase
Uncovered code in 'master' (5a0cc8a) and not in (fe8321ec05)
-----------------------------------------------------------------
builtin/fsck.c
66ec0390e7 builtin/fsck.c 862) midx_argv[2] = "--object-dir";
66ec0390e7 builtin/fsck.c 863) midx_argv[3] = alt->path;
66ec0390e7 builtin/fsck.c 864) if (run_command(&midx_verify))
66ec0390e7 builtin/fsck.c 865) errors_found |= ERROR_COMMIT_GRAPH;
fsck.c
fb8952077d 214) die_errno("Could not read '%s'", path);
midx.c
56ee7ff156 949) return 0;
cc6af73c02 990) midx_report(_("failed to load pack-index for packfile %s"),
cc6af73c02 991) e.p->pack_name);
cc6af73c02 992) break;
Commits introducing uncovered code:
Derrick Stolee 56ee7ff15: multi-pack-index: add 'verify' verb
Derrick Stolee 66ec0390e: fsck: verify multi-pack-index
Derrick Stolee cc6af73c0: multi-pack-index: verify object offsets
René Scharfe fb8952077: fsck: use strbuf_getline() to read skiplist
file
[1] https://dev.azure.com/git/git/_build/results?buildId=184&view=results
Build running coverage-test for 'pu' and coverage-diff.sh against 'jch'
[2] https://dev.azure.com/git/git/_build/results?buildId=185&view=results
Build running coverage-test for 'jch' and coverage-diff.sh against
'next'
[3] https://dev.azure.com/git/git/_build/results?buildId=186&view=results
Build running coverage-test for 'next' and coverage-diff.sh against
'master'
[4] https://dev.azure.com/git/git/_build/results?buildId=187&view=logs
Build running coverage-test for 'master' and coverage-diff.sh
against 'master@{1}'
^ permalink raw reply [relevance 1%]
* [PATCH v2 00/15] Hash function transition part 15
@ 2018-10-15 0:01 5% brian m. carlson
0 siblings, 0 replies; 200+ results
From: brian m. carlson @ 2018-10-15 0:01 UTC (permalink / raw)
To: git
Cc: Jeff King, Eric Sunshine, Nguyễn Thái Ngọc Duy,
Stefan Beller
This is the fifteenth series in the ongoing hash function transition.
This series includes several conversions to use the_hash_algo, combined
with some use of parse_oid_hex and GIT_MAX_RAWSZ.
Changes from v1:
* Fix several other substantially similar issues in builtin/repack.
* Fix a comment style issue.
* Improve commit message as suggested by Stefan.
* Pull in Gábor's patch and place it at the beginning of the series.
SZEDER Gábor (1):
object_id.cocci: match only expressions of type 'struct object_id'
brian m. carlson (14):
pack-bitmap-write: use GIT_MAX_RAWSZ for allocation
builtin/repack: replace hard-coded constants
builtin/mktree: remove hard-coded constant
builtin/fetch-pack: remove constants with parse_oid_hex
pack-revindex: express constants in terms of the_hash_algo
packfile: express constants in terms of the_hash_algo
refs/packed-backend: express constants using the_hash_algo
upload-pack: express constants in terms of the_hash_algo
transport: use parse_oid_hex instead of a constant
tag: express constant in terms of the_hash_algo
apply: replace hard-coded constants
apply: rename new_sha1_prefix and old_sha1_prefix
submodule: make zero-oid comparison hash function agnostic
rerere: convert to use the_hash_algo
apply.c | 50 ++++++------
builtin/fetch-pack.c | 13 ++--
builtin/mktree.c | 2 +-
builtin/repack.c | 13 ++--
contrib/coccinelle/object_id.cocci | 117 ++++++++++++++++-------------
git-submodule.sh | 7 +-
pack-bitmap-write.c | 2 +-
pack-revindex.c | 10 ++-
packfile.c | 5 +-
refs/packed-backend.c | 14 ++--
rerere.c | 81 ++++++++++----------
tag.c | 2 +-
transport.c | 7 +-
upload-pack.c | 13 ++--
14 files changed, 181 insertions(+), 155 deletions(-)
Range-diff against v1:
-: ---------- > 1: 35d9cefd9a object_id.cocci: match only expressions of type 'struct object_id'
1: 223d6f6695 ! 2: bb2a15a6e9 pack-bitmap-write: use GIT_MAX_RAWSZ for allocation
@@ -2,6 +2,7 @@
pack-bitmap-write: use GIT_MAX_RAWSZ for allocation
+ Reviewed-by: Stefan Beller <sbeller@google.com>
Signed-off-by: brian m. carlson <sandals@crustytoothpaste.net>
diff --git a/pack-bitmap-write.c b/pack-bitmap-write.c
2: bbdb147d8d < -: ---------- builtin/repack: replace hard-coded constant
-: ---------- > 3: 95bdac161e builtin/repack: replace hard-coded constants
3: bf7b55fe4e = 4: 2a28675cfc builtin/mktree: remove hard-coded constant
4: 155451f608 = 5: 7941bb0060 builtin/fetch-pack: remove constants with parse_oid_hex
5: 9ec7065d62 ! 6: 2ec9a22ea7 pack-revindex: express constants in terms of the_hash_algo
@@ -3,7 +3,9 @@
pack-revindex: express constants in terms of the_hash_algo
Express the various constants used in terms of the_hash_algo.
+ While we're at it, fix a comment style issue as well.
+ Reviewed-by: Stefan Beller <sbeller@google.com>
Signed-off-by: brian m. carlson <sandals@crustytoothpaste.net>
diff --git a/pack-revindex.c b/pack-revindex.c
@@ -37,7 +39,8 @@
}
- /* This knows the pack format -- the 20-byte trailer
-+ /* This knows the pack format -- the hash trailer
++ /*
++ * This knows the pack format -- the hash trailer
* follows immediately after the last object data.
*/
- p->revindex[num_ent].offset = p->pack_size - 20;
6: 5e8576c6e4 ! 7: 3ccb2b7217 packfile: express constants in terms of the_hash_algo
@@ -5,6 +5,11 @@
Replace uses of GIT_SHA1_RAWSZ with references to the_hash_algo to avoid
dependence on a particular hash length.
+ It's likely that in the future, we'll update the pack format to indicate
+ what hash algorithm it uses, and then this code will change. However,
+ at least on an interim basis, make it easier to develop on a pure
+ SHA-256 Git by using the_hash_algo here.
+
Signed-off-by: brian m. carlson <sandals@crustytoothpaste.net>
diff --git a/packfile.c b/packfile.c
7: 4c7b6471db = 8: 39eb7e1069 refs/packed-backend: express constants using the_hash_algo
8: 8318dcb630 = 9: 55269d5fc2 upload-pack: express constants in terms of the_hash_algo
9: 16916c9fa2 = 10: d30536a3bd transport: use parse_oid_hex instead of a constant
10: 9d8e2bc4ae = 11: 2139fe1fe1 tag: express constant in terms of the_hash_algo
11: 58f91f2167 = 12: af815c4215 apply: replace hard-coded constants
12: 6899dfc4af = 13: f651b226c8 apply: rename new_sha1_prefix and old_sha1_prefix
13: cc974651cb = 14: bf1d450aa5 submodule: make zero-oid comparison hash function agnostic
14: b373f16c12 = 15: d984036c1c rerere: convert to use the_hash_algo
^ permalink raw reply [relevance 5%]
* Re: What's cooking in git.git (Oct 2018, #02; Sat, 13)
2018-10-12 19:44 2% ` Stefan Beller
2018-10-12 23:37 7% ` Stefan Beller
@ 2018-10-13 1:03 5% ` Junio C Hamano
2018-10-15 17:32 5% ` Stefan Beller
1 sibling, 1 reply; 200+ results
From: Junio C Hamano @ 2018-10-13 1:03 UTC (permalink / raw)
To: Stefan Beller; +Cc: git
Stefan Beller <sbeller@google.com> writes:
>> * sb/submodule-recursive-fetch-gets-the-tip (2018-10-11) 9 commits
>> . builtin/fetch: check for submodule updates for non branch fetches
>> . fetch: retry fetching submodules if needed objects were not fetched
>> . submodule: fetch in submodules git directory instead of in worktree
>> . repository: repo_submodule_init to take a submodule struct
>> . submodule.c: do not copy around submodule list
>> . submodule.c: move global changed_submodule_names into fetch submodule struct
>> . submodule.c: sort changed_submodule_names before searching it
>> . submodule.c: fix indentation
>> . sha1-array: provide oid_array_filter
>>
>> "git fetch --recurse-submodules" may not fetch the necessary commit
>> that is bound to the superproject, which is getting corrected.
>>
>> Ejected for now, as it has fallouts in places like t/helper/.
>
> This is the first time I hear about that, I'll look into that.
> The tipmost commit there is also shoddy, I'll redo that.
This is the first time I saw the breakage with this series, but I
would not be suprised, as this was rerolled recently. Who knows
what got changed in this series and in other topics---any new
interaction can arise and that is a normal part of distributed
development.
The xx/sb-submodule-recursive-fetch-gets-the-tip-in-pu branch at
git://github.com/gitster/git.git has a merge of this into 'pu', with
textual conflicts all resolved.
At least t/helper/test-submodule-nested-repo-config.c fails to
build; I didn't check if there are other breakages.
^ permalink raw reply [relevance 5%]
* Re: [PATCH 17/19] submodule: use submodule repos for object lookup
@ 2018-10-13 0:20 8% ` Stefan Beller
2018-10-16 19:30 8% ` Stefan Beller
1 sibling, 0 replies; 200+ results
From: Stefan Beller @ 2018-10-13 0:20 UTC (permalink / raw)
To: Jonathan Tan; +Cc: git
On Thu, Oct 11, 2018 at 3:41 PM Jonathan Tan <jonathantanmy@google.com> wrote:
>
> > +/*
> > + * Initialize 'out' based on the provided submodule path.
> > + *
> > + * Unlike repo_submodule_init, this tolerates submodules not present
> > + * in .gitmodules. NEEDSWORK: The repo_submodule_init behavior is
> > + * preferrable. This function exists only to preserve historical behavior.
>
> What do you mean by "The repo_submodule_init behavior is preferable"?
Preferable in the sense that it follows the definition of a submodule, but this
here works for "any repo" that happens to be at the gitlink.
> If
> we need to preserve historical behavior, then it seems that the most
> preferable one is something that meets our needs (like open_submodule()
> in this patch).
Yes, I'll reword to drop the preferrable, but still state the difference.
I wonder if instead we'd want to introduce a
repo_submodule_init(struct repository *submodule \
struct repository *superproject \
const char *path, \
int tolerate_lookalikes)
Another patch proposes to replace the path
by a struct submodule, but for lookalikes, we do not have
a struct submodule to begin with (though in the other
patches we cook up a fake entry in the submodule config)
> > +static int open_submodule(struct repository *out, const char *path)
> > +{
> > + struct strbuf sb = STRBUF_INIT;
> > +
> > + if (submodule_to_gitdir(&sb, path) || repo_init(out, sb.buf, NULL)) {
> > + strbuf_release(&sb);
> > + return -1;
> > + }
> > +
> > + out->submodule_prefix = xstrdup(path);
>
> Do we need to set submodule_prefix?
Good point! Thanks for catching this.
>
> > @@ -507,7 +533,7 @@ static void show_submodule_header(struct diff_options *o, const char *path,
> > else if (is_null_oid(two))
> > message = "(submodule deleted)";
> >
> > - if (add_submodule_odb(path)) {
> > + if (open_submodule(sub, path) < 0) {
>
> This function, as a side effect, writes the open repository to "sub" for
> its caller to use. I think it's better if its callers open "sub" and
> then pass it to show_submodule_header() if successful. If opening the
> submodule is expected to fail sometimes (like it seems here), then we
> can allow callers to also pass NULL, and document what happens when NULL
> is passed.
That looks intriguing, I'll take a look. Note that we also pass
in **left and **right to have it assigned in there.
>
> Also, repositories open in this way should also be freed.
Yes, thanks!
^ permalink raw reply [relevance 8%]
* Re: What's cooking in git.git (Oct 2018, #02; Sat, 13)
2018-10-12 19:44 2% ` Stefan Beller
@ 2018-10-12 23:37 7% ` Stefan Beller
2018-10-13 1:03 5% ` Junio C Hamano
1 sibling, 0 replies; 200+ results
From: Stefan Beller @ 2018-10-12 23:37 UTC (permalink / raw)
To: Junio C Hamano; +Cc: git
On Fri, Oct 12, 2018 at 12:44 PM Stefan Beller <sbeller@google.com> wrote:
> > * sb/submodule-recursive-fetch-gets-the-tip (2018-10-11) 9 commits
> > . builtin/fetch: check for submodule updates for non branch fetches
> > . fetch: retry fetching submodules if needed objects were not fetched
> > . submodule: fetch in submodules git directory instead of in worktree
> > . repository: repo_submodule_init to take a submodule struct
> > . submodule.c: do not copy around submodule list
> > . submodule.c: move global changed_submodule_names into fetch submodule struct
> > . submodule.c: sort changed_submodule_names before searching it
> > . submodule.c: fix indentation
> > . sha1-array: provide oid_array_filter
> >
> > "git fetch --recurse-submodules" may not fetch the necessary commit
> > that is bound to the superproject, which is getting corrected.
> >
> > Ejected for now, as it has fallouts in places like t/helper/.
>
> This is the first time I hear about that, I'll look into that.
I looked into that, and merging with origin/next only
had one merge conflict in submodule.c, easy to resolve.
It builds and tests cleanly after that.
What fallouts did you observe?
> The tipmost commit there is also shoddy, I'll redo that.
I confused this series with sb/submodule-update-in-C
which got merged already, I may send a fixup commit there.
So it seems to me that this series is still good.
Thanks,
Stefan
^ permalink raw reply [relevance 7%]
* Re: [PATCH] submodule helper: convert relative URL to absolute URL if needed
2018-10-12 21:53 22% [PATCH] submodule helper: convert relative URL to absolute URL if needed Stefan Beller
@ 2018-10-12 22:27 4% ` Jonathan Nieder
2018-10-16 0:19 21% ` Stefan Beller
1 sibling, 0 replies; 200+ results
From: Jonathan Nieder @ 2018-10-12 22:27 UTC (permalink / raw)
To: Stefan Beller; +Cc: git, Jaewoong Jung
Hi,
Stefan Beller wrote:
> The submodule helper update_clone called by "git submodule update",
> clones submodules if needed. As submodules used to have the URL indicating
> if they were active, the step to resolve relative URLs was done in the
> "submodule init" step. Nowadays submodules can be configured active without
> calling an explicit init, e.g. via configuring submodule.active.
>
> Then we'll fallback to the URL found in the .gitmodules, which may be
> relative to the superproject, but we do not resolve it, yet.
Oh! Good catch.
> To do so, factor out the function that resolves the relative URLs in
> "git submodule init" (in the submodule helper in the init_submodule
> function) and cal it at the appropriate place in the update_clone helper.
s/cal/call/
> Signed-off-by: Stefan Beller <sbeller@google.com>
> ---
> builtin/submodule--helper.c | 48 ++++++++++++++++++++++++-------------
> t/t7400-submodule-basic.sh | 24 +++++++++++++++++++
> 2 files changed, 55 insertions(+), 17 deletions(-)
What is the symptom when this fails?
> diff --git a/builtin/submodule--helper.c b/builtin/submodule--helper.c
> index f6fb8991f3..a9a3ac38be 100644
> --- a/builtin/submodule--helper.c
> +++ b/builtin/submodule--helper.c
> @@ -584,6 +584,27 @@ static int module_foreach(int argc, const char **argv, const char *prefix)
> return 0;
> }
>
> +
> +char *compute_submodule_clone_url(const char *rel_url)
Should be static.
Is the caller responsible for freeing the returned buffer?
> +{
> + char *remoteurl, *relurl;
> + char *remote = get_default_remote();
> + struct strbuf remotesb = STRBUF_INIT;
optional: could rename, something like
struct strbuf key = STRBUF_INIT;
remote = get_default_remote();
strbuf_addf(&key, "remote.%s.url", remote);
[...]
strbuf_release(&key);
free(remote);
return result;
> +
> + strbuf_addf(&remotesb, "remote.%s.url", remote);
> + free(remote);
> +
> + if (git_config_get_string(remotesb.buf, &remoteurl)) {
> + warning(_("could not lookup configuration '%s'. Assuming this repository is its own authoritative upstream."), remotesb.buf);
nit: lookup is the noun, "look up" is the verb
> + remoteurl = xgetcwd();
> + }
> + relurl = relative_url(remoteurl, rel_url, NULL);
> + strbuf_release(&remotesb);
> + free(remoteurl);
> +
> + return relurl;
> +}
> +
> struct init_cb {
> const char *prefix;
> unsigned int flags;
> @@ -634,21 +655,9 @@ static void init_submodule(const char *path, const char *prefix,
> /* Possibly a url relative to parent */
> if (starts_with_dot_dot_slash(url) ||
> starts_with_dot_slash(url)) {
> - char *remoteurl, *relurl;
> - char *remote = get_default_remote();
> - struct strbuf remotesb = STRBUF_INIT;
> - strbuf_addf(&remotesb, "remote.%s.url", remote);
> - free(remote);
> -
> - if (git_config_get_string(remotesb.buf, &remoteurl)) {
> - warning(_("could not lookup configuration '%s'. Assuming this repository is its own authoritative upstream."), remotesb.buf);
> - remoteurl = xgetcwd();
> - }
> - relurl = relative_url(remoteurl, url, NULL);
> - strbuf_release(&remotesb);
> - free(remoteurl);
> - free(url);
> - url = relurl;
Ah, this is moved code. I should have used --color-moved. ;-)
> + char *to_free = url;
> + url = compute_submodule_clone_url(url);
> + free(to_free);
Maybe:
char *old_url = url;
url = ...(old_url);
free(old_url);
> }
>
> if (git_config_set_gently(sb.buf, url))
> @@ -1562,8 +1571,13 @@ static int prepare_to_clone_next_submodule(const struct cache_entry *ce,
>
> strbuf_reset(&sb);
> strbuf_addf(&sb, "submodule.%s.url", sub->name);
> - if (repo_config_get_string_const(the_repository, sb.buf, &url))
> - url = sub->url;
> + if (repo_config_get_string_const(the_repository, sb.buf, &url)) {
> + if (starts_with_dot_slash(sub->url) ||
> + starts_with_dot_dot_slash(sub->url))
> + url = compute_submodule_clone_url(sub->url);
> + else
> + url = sub->url;
> + }
Nice.
>
> strbuf_reset(&sb);
> strbuf_addf(&sb, "%s/.git", ce->name);
> diff --git a/t/t7400-submodule-basic.sh b/t/t7400-submodule-basic.sh
> index c0ffc1022a..3f5dd5e4ef 100755
> --- a/t/t7400-submodule-basic.sh
> +++ b/t/t7400-submodule-basic.sh
> @@ -1224,6 +1224,30 @@ test_expect_success 'submodule update and setting submodule.<name>.active' '
> test_cmp expect actual
> '
>
> +test_expect_success 'clone active submodule without submodule url set' '
Thanks for the test \o/.
> + test_when_finished "rm -rf test/test" &&
> + mkdir test &&
> + # another dir breaks accidental relative paths still being correct
> + git clone file://"$pwd"/multisuper test/test &&
> + (
> + cd test/test &&
> + git config submodule.active "." &&
> +
> + # do not pass --init flag, as it is already active:
What does "it" refer to here?
> + git submodule update &&
> + git submodule status >actual_raw &&
> +
> + cut -c 1,43- actual_raw >actual &&
> + cat >expect <<-\EOF &&
> + sub0 (test2)
> + sub1 (test2)
> + sub2 (test2)
> + sub3 (test2)
> + EOF
> + test_cmp expect actual
> + )
> +'
> +
> test_expect_success 'clone --recurse-submodules with a pathspec works' '
> test_when_finished "rm -rf multisuper_clone" &&
> cat >expected <<-\EOF &&
Thanks for the quick fix.
Jonathan
^ permalink raw reply [relevance 4%]
* [PATCH] submodule helper: convert relative URL to absolute URL if needed
@ 2018-10-12 21:53 22% Stefan Beller
2018-10-12 22:27 4% ` Jonathan Nieder
2018-10-16 0:19 21% ` Stefan Beller
0 siblings, 2 replies; 200+ results
From: Stefan Beller @ 2018-10-12 21:53 UTC (permalink / raw)
To: git; +Cc: Stefan Beller
The submodule helper update_clone called by "git submodule update",
clones submodules if needed. As submodules used to have the URL indicating
if they were active, the step to resolve relative URLs was done in the
"submodule init" step. Nowadays submodules can be configured active without
calling an explicit init, e.g. via configuring submodule.active.
Then we'll fallback to the URL found in the .gitmodules, which may be
relative to the superproject, but we do not resolve it, yet.
To do so, factor out the function that resolves the relative URLs in
"git submodule init" (in the submodule helper in the init_submodule
function) and cal it at the appropriate place in the update_clone helper.
Signed-off-by: Stefan Beller <sbeller@google.com>
---
builtin/submodule--helper.c | 48 ++++++++++++++++++++++++-------------
t/t7400-submodule-basic.sh | 24 +++++++++++++++++++
2 files changed, 55 insertions(+), 17 deletions(-)
diff --git a/builtin/submodule--helper.c b/builtin/submodule--helper.c
index f6fb8991f3..a9a3ac38be 100644
--- a/builtin/submodule--helper.c
+++ b/builtin/submodule--helper.c
@@ -584,6 +584,27 @@ static int module_foreach(int argc, const char **argv, const char *prefix)
return 0;
}
+
+char *compute_submodule_clone_url(const char *rel_url)
+{
+ char *remoteurl, *relurl;
+ char *remote = get_default_remote();
+ struct strbuf remotesb = STRBUF_INIT;
+
+ strbuf_addf(&remotesb, "remote.%s.url", remote);
+ free(remote);
+
+ if (git_config_get_string(remotesb.buf, &remoteurl)) {
+ warning(_("could not lookup configuration '%s'. Assuming this repository is its own authoritative upstream."), remotesb.buf);
+ remoteurl = xgetcwd();
+ }
+ relurl = relative_url(remoteurl, rel_url, NULL);
+ strbuf_release(&remotesb);
+ free(remoteurl);
+
+ return relurl;
+}
+
struct init_cb {
const char *prefix;
unsigned int flags;
@@ -634,21 +655,9 @@ static void init_submodule(const char *path, const char *prefix,
/* Possibly a url relative to parent */
if (starts_with_dot_dot_slash(url) ||
starts_with_dot_slash(url)) {
- char *remoteurl, *relurl;
- char *remote = get_default_remote();
- struct strbuf remotesb = STRBUF_INIT;
- strbuf_addf(&remotesb, "remote.%s.url", remote);
- free(remote);
-
- if (git_config_get_string(remotesb.buf, &remoteurl)) {
- warning(_("could not lookup configuration '%s'. Assuming this repository is its own authoritative upstream."), remotesb.buf);
- remoteurl = xgetcwd();
- }
- relurl = relative_url(remoteurl, url, NULL);
- strbuf_release(&remotesb);
- free(remoteurl);
- free(url);
- url = relurl;
+ char *to_free = url;
+ url = compute_submodule_clone_url(url);
+ free(to_free);
}
if (git_config_set_gently(sb.buf, url))
@@ -1562,8 +1571,13 @@ static int prepare_to_clone_next_submodule(const struct cache_entry *ce,
strbuf_reset(&sb);
strbuf_addf(&sb, "submodule.%s.url", sub->name);
- if (repo_config_get_string_const(the_repository, sb.buf, &url))
- url = sub->url;
+ if (repo_config_get_string_const(the_repository, sb.buf, &url)) {
+ if (starts_with_dot_slash(sub->url) ||
+ starts_with_dot_dot_slash(sub->url))
+ url = compute_submodule_clone_url(sub->url);
+ else
+ url = sub->url;
+ }
strbuf_reset(&sb);
strbuf_addf(&sb, "%s/.git", ce->name);
diff --git a/t/t7400-submodule-basic.sh b/t/t7400-submodule-basic.sh
index c0ffc1022a..3f5dd5e4ef 100755
--- a/t/t7400-submodule-basic.sh
+++ b/t/t7400-submodule-basic.sh
@@ -1224,6 +1224,30 @@ test_expect_success 'submodule update and setting submodule.<name>.active' '
test_cmp expect actual
'
+test_expect_success 'clone active submodule without submodule url set' '
+ test_when_finished "rm -rf test/test" &&
+ mkdir test &&
+ # another dir breaks accidental relative paths still being correct
+ git clone file://"$pwd"/multisuper test/test &&
+ (
+ cd test/test &&
+ git config submodule.active "." &&
+
+ # do not pass --init flag, as it is already active:
+ git submodule update &&
+ git submodule status >actual_raw &&
+
+ cut -c 1,43- actual_raw >actual &&
+ cat >expect <<-\EOF &&
+ sub0 (test2)
+ sub1 (test2)
+ sub2 (test2)
+ sub3 (test2)
+ EOF
+ test_cmp expect actual
+ )
+'
+
test_expect_success 'clone --recurse-submodules with a pathspec works' '
test_when_finished "rm -rf multisuper_clone" &&
cat >expected <<-\EOF &&
--
2.19.0
^ permalink raw reply related [relevance 22%]
* Re: What's cooking in git.git (Oct 2018, #02; Sat, 13)
@ 2018-10-12 19:44 2% ` Stefan Beller
2018-10-12 23:37 7% ` Stefan Beller
2018-10-13 1:03 5% ` Junio C Hamano
0 siblings, 2 replies; 200+ results
From: Stefan Beller @ 2018-10-12 19:44 UTC (permalink / raw)
To: Junio C Hamano; +Cc: git
>
> * sb/strbuf-h-update (2018-09-29) 1 commit
> - strbuf.h: format according to coding guidelines
>
> Code clean-up to serve as a BCP example.
>
> What's the status of this one after the discussion thread stopped here?
> cf. <CAGZ79kbV6QjsFKcD2uG_P9j1AvzSNQSi-_jXGQ9w0YU9fjhEGg@mail.gmail.com>
I started rewriting the documentation according to your proposal of having
parameters with name, then referred to as NAME in the docs.
After a few examples, I must admit I do not like that style,
so I would just want to do the minimal part that would get this patch landed,
i.e.
(a) convince you that the patch is good as-is or
(b) resend with fewer parameters made explicit if
we desire to be concise instead.
> * nd/the-index (2018-09-21) 23 commits
> (merged to 'next' on 2018-10-10 at 16e2e2e947)
[...]
> (this branch is used by sb/more-repo-in-api.)
>
> Various codepaths in the core-ish part learn to work on an
> arbitrary in-core index structure, not necessarily the default
> instance "the_index".
>
> Will merge to 'master'.
Cool!
sb/more-repo-in-api is not part of this message,
but it requires at least one resend, so I'll do that.
> * sb/submodule-recursive-fetch-gets-the-tip (2018-10-11) 9 commits
> . builtin/fetch: check for submodule updates for non branch fetches
> . fetch: retry fetching submodules if needed objects were not fetched
> . submodule: fetch in submodules git directory instead of in worktree
> . repository: repo_submodule_init to take a submodule struct
> . submodule.c: do not copy around submodule list
> . submodule.c: move global changed_submodule_names into fetch submodule struct
> . submodule.c: sort changed_submodule_names before searching it
> . submodule.c: fix indentation
> . sha1-array: provide oid_array_filter
>
> "git fetch --recurse-submodules" may not fetch the necessary commit
> that is bound to the superproject, which is getting corrected.
>
> Ejected for now, as it has fallouts in places like t/helper/.
This is the first time I hear about that, I'll look into that.
The tipmost commit there is also shoddy, I'll redo that.
^ permalink raw reply [relevance 2%]
* Git Test Coverage Report (Friday, Oct 12)
@ 2018-10-12 12:59 2% Derrick Stolee
0 siblings, 0 replies; 200+ results
From: Derrick Stolee @ 2018-10-12 12:59 UTC (permalink / raw)
To: Git List
In an effort to ensure new code is reasonably covered by the test suite,
we now have contrib/coverage-diff.sh to combine the gcov output from
'make coverage-test ; make coverage-report' with the output from 'git
diff A B' to discover _new_ lines of code that are not covered.
This report takes the output of these results after running on four
branches:
pu: 13cafb433ca0dd31b3ea229a79cc8507aa89ee19 (tests are broken,
so coverage not reported)
jch: fda196ac82002ede984896861cf28a354044d1a0
next: 76f2f5c1e34c4dbef1029e2984c2892894c444ce
master: 04861070400d3ca9d487bd0d736ca818b9ffe371
maint: cae598d9980661a978e2df4fb338518f7bf09572
I ran the test suite on each of these branches on an Ubuntu Linux VM,
and I'm missing some dependencies (like apache, svn, and perforce) so
not all tests are run.
I submit this output without comment. I'm taking a look especially at my
own lines to see where coverage can be improved.
Thanks,
-Stolee
Uncovered code new in jch, compared to next. Build logs at [1]
--------------------------------------------------------------
builtin/archive.c
e001fd3a50 builtin/archive.c 78) die(_("git archive: expected ACK/NAK,
got a flush packet"));
e001fd3a50 builtin/archive.c 80) if (starts_with(reader.line, "NACK "))
e001fd3a50 builtin/archive.c 81) die(_("git archive: NACK %s"),
reader.line + 5);
e001fd3a50 builtin/archive.c 82) if (starts_with(reader.line, "ERR "))
e001fd3a50 builtin/archive.c 83) die(_("remote error: %s"), reader.line
+ 4);
e001fd3a50 builtin/archive.c 84) die(_("git archive: protocol error"));
e001fd3a50 builtin/archive.c 89) die(_("git archive: expected a flush"));
fb19d32f05 builtin/archive.c 99) if (version != discover_version(&reader))
fb19d32f05 builtin/archive.c 100) die(_("git archive: received different
protocol versions in subsequent requests"));
builtin/fetch.c
7bbc53a589 builtin/fetch.c 385) continue; /* can this happen??? */
builtin/rebase--interactive.c
53bbcfbde7 builtin/rebase--interactive2.c 24) return error(_("no HEAD?"));
53bbcfbde7 builtin/rebase--interactive2.c 51) return
error_errno(_("could not create temporary %s"), path_state_dir());
53bbcfbde7 builtin/rebase--interactive2.c 57) return
error_errno(_("could not mark as interactive"));
53bbcfbde7 builtin/rebase--interactive2.c 77) return -1;
53bbcfbde7 builtin/rebase--interactive2.c 81) return -1;
53bbcfbde7 builtin/rebase--interactive2.c 87) free(revisions);
53bbcfbde7 builtin/rebase--interactive2.c 88) free(shortrevisions);
53bbcfbde7 builtin/rebase--interactive2.c 90) return -1;
53bbcfbde7 builtin/rebase--interactive2.c 98) free(revisions);
53bbcfbde7 builtin/rebase--interactive2.c 99) free(shortrevisions);
53bbcfbde7 builtin/rebase--interactive2.c 101) return
error_errno(_("could not open %s"), rebase_path_todo());
53bbcfbde7 builtin/rebase--interactive2.c 106)
argv_array_push(&make_script_args, restrict_revision);
53bbcfbde7 builtin/rebase--interactive2.c 114) error(_("could not
generate todo list"));
53bbcfbde7 builtin/rebase--interactive2.c 206)
usage_with_options(builtin_rebase_interactive_usage, options);
53bbcfbde7 builtin/rebase--interactive2.c 220)
warning(_("--[no-]rebase-cousins has no effect without "
0af129b2ed builtin/rebase--interactive2.c 226) die(_("a base commit must
be provided with --upstream or --onto"));
34b47315d9 builtin/rebase--interactive.c 261) ret = rearrange_squash();
34b47315d9 builtin/rebase--interactive.c 262) break;
34b47315d9 builtin/rebase--interactive.c 264) ret =
sequencer_add_exec_commands(cmd);
34b47315d9 builtin/rebase--interactive.c 265) break;
0af129b2ed builtin/rebase--interactive2.c 267) BUG("invalid command
'%d'", command);
builtin/rebase.c
55071ea248 61) strbuf_trim(&out);
55071ea248 62) ret = !strcmp("true", out.buf);
55071ea248 63) strbuf_release(&out);
002ee2fe68 115) die(_("%s requires an interactive rebase"), option);
f95736288a 148) return error_errno(_("could not read '%s'"), path);
f95736288a 162) return -1;
f95736288a 167) return error(_("could not get 'onto': '%s'"), buf.buf);
f95736288a 178) return -1;
f95736288a 179) } else if (read_one(state_dir_path("head", opts), &buf))
f95736288a 180) return -1;
f95736288a 182) return error(_("invalid orig-head: '%s'"), buf.buf);
f95736288a 186) return -1;
f95736288a 188) opts->flags &= ~REBASE_NO_QUIET;
73d51ed0a5 196) opts->signoff = 1;
73d51ed0a5 197) opts->flags |= REBASE_FORCE;
ead98c111b 204) return -1;
12026a412c 219) return -1;
ba1905a5fe 227) return -1;
ba1905a5fe 235) return -1;
6defce2b02 255) return error(_("Could not read '%s'"), path);
6defce2b02 271) res = error(_("Cannot store %s"), autostash.buf);
6defce2b02 275) return res;
bc24382c2b 373) argv_array_pushf(&child.args,
bc24382c2b 375) oid_to_hex(&opts->restrict_revision->object.oid));
ac7f467fef 488) BUG("Unhandled rebase type %d", opts->type);
ac7f467fef 507) struct strbuf dir = STRBUF_INIT;
6defce2b02 509) apply_autostash(opts);
ac7f467fef 510) strbuf_addstr(&dir, opts->state_dir);
ac7f467fef 511) remove_dir_recursively(&dir, 0);
ac7f467fef 512) strbuf_release(&dir);
ac7f467fef 513) die("Nothing to do");
d4c569f8f4 540) BUG("Not a fully qualified branch: '%s'",
switch_to_branch);
ac7f467fef 543) return -1;
ac7f467fef 547) rollback_lock_file(&lock);
ac7f467fef 548) return error(_("could not determine HEAD revision"));
ac7f467fef 565) rollback_lock_file(&lock);
ac7f467fef 566) return error(_("could not read index"));
ac7f467fef 570) error(_("failed to find tree of %s"), oid_to_hex(oid));
ac7f467fef 571) rollback_lock_file(&lock);
ac7f467fef 572) free((void *)desc.buffer);
ac7f467fef 573) return -1;
ac7f467fef 586) ret = error(_("could not write index"));
ac7f467fef 590) return ret;
ac7f467fef 606) } else if (old_orig)
ac7f467fef 607) delete_ref(NULL, "ORIG_HEAD", old_orig, 0);
bff014dac7 635) opts->flags &= !REBASE_DIFFSTAT;
9a48a615b4 669) return 1;
9a48a615b4 685) return 0;
55071ea248 893) const char *path = mkpath("%s/git-legacy-rebase",
55071ea248 896) if (sane_execvp(path, (char **)argv) < 0)
55071ea248 897) die_errno(_("could not exec %s"), path);
55071ea248 899) BUG("sane_execvp() returned???");
0eabf4b95c 915) die(_("It looks like 'git am' is in progress. Cannot
rebase."));
f28d40d3a9 952) usage_with_options(builtin_rebase_usage,
f95736288a 972) die(_("Cannot read HEAD"));
f95736288a 976) die(_("could not read index"));
f95736288a 990) exit(1);
122420c295 1002) die(_("could not discard worktree changes"));
122420c295 1004) exit(1);
5e5d96197c 1015) exit(1);
5e5d96197c 1018) die(_("could not move back to %s"),
5a61494539 1028) die(_("could not remove '%s'"), options.state_dir);
51e9ea6da7 1042) BUG("action: %d", action);
c54dacb50e 1047) const char *last_slash = strrchr(options.state_dir, '/');
c54dacb50e 1048) const char *state_dir_base =
c54dacb50e 1049) last_slash ? last_slash + 1 : options.state_dir;
c54dacb50e 1050) const char *cmd_live_rebase =
c54dacb50e 1052) strbuf_reset(&buf);
c54dacb50e 1053) strbuf_addf(&buf, "rm -fr \"%s\"", options.state_dir);
c54dacb50e 1054) die(_("It seems that there is already a %s directory,
and\n"
53f9e5be94 1078) strbuf_addstr(&options.git_am_opt, " --ignore-date");
53f9e5be94 1079) options.flags |= REBASE_FORCE;
7998dbe1ec 1091) strbuf_addf(&options.git_am_opt, " -C%d", opt_c);
3c3588c7d3 1123) else if (strcmp("no-rebase-cousins", rebase_merges))
3c3588c7d3 1124) die(_("Unknown mode: %s"), rebase_merges);
ba1905a5fe 1146) die(_("--strategy requires --merge or --interactive"));
ba1905a5fe 1156) BUG("unhandled rebase type (%d)", options.type);
cda614e489 1164) strbuf_addstr(&options.git_format_patch_opt, "
--progress");
ac7f467fef 1173) options.state_dir = apply_dir();
ac7f467fef 1174) break;
ac7f467fef 1251) die(_("invalid upstream '%s'"), options.upstream_name);
9dba809a69 1257) die(_("Could not create new root commit"));
e65123a71d 1307) die(_("fatal: no such branch/commit '%s'"),
ac7f467fef 1315) die(_("No such ref: %s"), "HEAD");
ac7f467fef 1327) die(_("Could not resolve HEAD to a revision"));
e65123a71d 1329) BUG("unexpected number of arguments left to parse");
e0333e5c63 1340) die(_("could not read index"));
6defce2b02 1367) die(_("Cannot autostash"));
6defce2b02 1370) die(_("Unexpected stash response: '%s'"),
6defce2b02 1376) die(_("Could not create directory for '%s'"),
6defce2b02 1382) die(_("could not reset --hard"));
e65123a71d 1426) ret = !!error(_("could not parse '%s'"),
e65123a71d 1428) goto cleanup;
e65123a71d 1437) ret = !!error(_("could not switch to "
1ed9c14ff2 1447) resolve_ref_unsafe("HEAD", 0, NULL, &flag))
1ed9c14ff2 1448) puts(_("HEAD is up to date."));
9a48a615b4 1457) resolve_ref_unsafe("HEAD", 0, NULL, &flag))
9a48a615b4 1458) puts(_("HEAD is up to date, rebase forced."));
builtin/rev-list.c
7c0fe330d5 builtin/rev-list.c 227) die("unexpected missing %s object '%s'",
7c0fe330d5 builtin/rev-list.c 228) type_name(obj->type),
oid_to_hex(&obj->oid));
250edfa8c8 builtin/rev-list.c 434) bisect_flags |= BISECT_FIND_ALL;
builtin/stash.c
93871263d1 builtin/stash--helper.c 130) free_stash_info(info);
93871263d1 builtin/stash--helper.c 131) error(_("'%s' is not a
stash-like commit"), revision);
93871263d1 builtin/stash--helper.c 132) exit(128);
93871263d1 builtin/stash--helper.c 165) free_stash_info(info);
93871263d1 builtin/stash--helper.c 166) fprintf_ln(stderr, _("No stash
entries found."));
93871263d1 builtin/stash--helper.c 167) return -1;
93871263d1 builtin/stash--helper.c 202) free_stash_info(info);
31f109a361 builtin/stash--helper.c 229) return error(_("git stash clear
with parameters is unimplemented"));
93871263d1 builtin/stash--helper.c 244) return -1;
93871263d1 builtin/stash--helper.c 252) return -1;
93871263d1 builtin/stash--helper.c 265) return -1;
93871263d1 builtin/stash--helper.c 268) return error(_("unable to write
new index file"));
93871263d1 builtin/stash--helper.c 374) remove_path(stash_index_path.buf);
93871263d1 builtin/stash--helper.c 375) return -1;
93871263d1 builtin/stash--helper.c 402) return -1;
93871263d1 builtin/stash--helper.c 405) return error(_("Cannot apply a
stash in the middle of a merge"));
93871263d1 builtin/stash--helper.c 415) strbuf_release(&out);
93871263d1 builtin/stash--helper.c 416) return -1;
93871263d1 builtin/stash--helper.c 422) return -1;
93871263d1 builtin/stash--helper.c 427) return -1;
93871263d1 builtin/stash--helper.c 434) return error(_("Could not
restore untracked files from stash"));
93871263d1 builtin/stash--helper.c 465) return -1;
93871263d1 builtin/stash--helper.c 470) strbuf_release(&out);
93871263d1 builtin/stash--helper.c 475) strbuf_release(&out);
93871263d1 builtin/stash--helper.c 476) return -1;
31f109a361 builtin/stash--helper.c 551) return error(_("%s: Could not
drop stash entry"),
b3513da4bd builtin/stash--helper.c 623) printf_ln(_("The stash entry is
kept in case you need it again."));
8ceb24b2c3 builtin/stash--helper.c 754) free_stash_info(&info);
129f0b0a00 builtin/stash.c 755)
usage_with_options(git_stash_show_usage, options);
0ac06fb81f builtin/stash--helper.c 808) fprintf_ln(stderr, _("\"git
stash store\" requires one <commit> argument"));
0ac06fb81f builtin/stash--helper.c 809) return -1;
f6f191b3f2 builtin/stash--helper.c 884) return 1;
f6f191b3f2 builtin/stash--helper.c 925) ret = -1;
f6f191b3f2 builtin/stash--helper.c 926) goto done;
f6f191b3f2 builtin/stash--helper.c 931) ret = -1;
f6f191b3f2 builtin/stash--helper.c 932) goto done;
f6f191b3f2 builtin/stash--helper.c 937) ret = -1;
f6f191b3f2 builtin/stash--helper.c 938) goto done;
f6f191b3f2 builtin/stash--helper.c 966) ret = -1;
f6f191b3f2 builtin/stash--helper.c 967) goto done;
f6f191b3f2 builtin/stash--helper.c 978) ret = -1;
f6f191b3f2 builtin/stash--helper.c 979) goto done;
f6f191b3f2 builtin/stash--helper.c 984) ret = -1;
f6f191b3f2 builtin/stash--helper.c 985) goto done;
f6f191b3f2 builtin/stash--helper.c 992) ret = -1;
f6f191b3f2 builtin/stash--helper.c 993) goto done;
f6f191b3f2 builtin/stash--helper.c 1018) ret = -1;
f6f191b3f2 builtin/stash--helper.c 1019) goto done;
f6f191b3f2 builtin/stash--helper.c 1031) ret = -1;
f6f191b3f2 builtin/stash--helper.c 1032) goto done;
f6f191b3f2 builtin/stash--helper.c 1037) ret = -1;
f6f191b3f2 builtin/stash--helper.c 1038) goto done;
f6f191b3f2 builtin/stash--helper.c 1049) ret = -1;
f6f191b3f2 builtin/stash--helper.c 1050) goto done;
f6f191b3f2 builtin/stash--helper.c 1055) ret = -1;
f6f191b3f2 builtin/stash--helper.c 1056) goto done;
8002b9e626 builtin/stash--helper.c 1089) fprintf_ln(stderr, _("You do
not have the initial commit yet"));
8002b9e626 builtin/stash--helper.c 1110) if (!quiet)
8002b9e626 builtin/stash--helper.c 1111) fprintf_ln(stderr, _("Cannot
save the current index state"));
f6f191b3f2 builtin/stash--helper.c 1112) ret = -1;
f6f191b3f2 builtin/stash--helper.c 1113) *stash_msg = NULL;
f6f191b3f2 builtin/stash--helper.c 1114) goto done;
8002b9e626 builtin/stash--helper.c 1119) if (!quiet)
8002b9e626 builtin/stash--helper.c 1120) fprintf_ln(stderr, _("Cannot
save the untracked files"));
f6f191b3f2 builtin/stash--helper.c 1121) ret = -1;
f6f191b3f2 builtin/stash--helper.c 1122) *stash_msg = NULL;
f6f191b3f2 builtin/stash--helper.c 1123) goto done;
8002b9e626 builtin/stash--helper.c 1131) if (!quiet)
8002b9e626 builtin/stash--helper.c 1132) fprintf_ln(stderr, _("Cannot
save the current worktree state"));
f6f191b3f2 builtin/stash--helper.c 1133) goto done;
8002b9e626 builtin/stash--helper.c 1139) if (!quiet)
8002b9e626 builtin/stash--helper.c 1140) fprintf_ln(stderr, _("Cannot
save the current worktree state"));
f6f191b3f2 builtin/stash--helper.c 1141) ret = -1;
f6f191b3f2 builtin/stash--helper.c 1142) *stash_msg = NULL;
f6f191b3f2 builtin/stash--helper.c 1143) goto done;
8002b9e626 builtin/stash--helper.c 1166) if (!quiet)
8002b9e626 builtin/stash--helper.c 1167) fprintf_ln(stderr, _("Cannot
record working tree state"));
f6f191b3f2 builtin/stash--helper.c 1168) ret = -1;
f6f191b3f2 builtin/stash--helper.c 1169) goto done;
48c061fa44 builtin/stash--helper.c 1251) return -1;
8002b9e626 builtin/stash--helper.c 1260) if (!quiet)
8002b9e626 builtin/stash--helper.c 1261) fprintf_ln(stderr, _("Cannot
initialize stash"));
48c061fa44 builtin/stash--helper.c 1262) return -1;
8002b9e626 builtin/stash--helper.c 1272) if (!quiet)
8002b9e626 builtin/stash--helper.c 1273) fprintf_ln(stderr, _("Cannot
save the current status"));
48c061fa44 builtin/stash--helper.c 1274) ret = -1;
48c061fa44 builtin/stash--helper.c 1275) goto done;
48c061fa44 builtin/stash--helper.c 1292) ret = -1;
48c061fa44 builtin/stash--helper.c 1311) ret = -1;
48c061fa44 builtin/stash--helper.c 1312) goto done;
48c061fa44 builtin/stash--helper.c 1321) ret = -1;
48c061fa44 builtin/stash--helper.c 1322) goto done;
48c061fa44 builtin/stash--helper.c 1330) ret = -1;
48c061fa44 builtin/stash--helper.c 1339) ret = -1;
48c061fa44 builtin/stash--helper.c 1350) ret = -1;
48c061fa44 builtin/stash--helper.c 1359) ret = -1;
48c061fa44 builtin/stash--helper.c 1360) goto done;
48c061fa44 builtin/stash--helper.c 1368) ret = -1;
48c061fa44 builtin/stash--helper.c 1394) ret = -1;
129f0b0a00 builtin/stash.c 1524)
usage_msg_opt(xstrfmt(_("unknown subcommand: %s"), argv[0]),
129f0b0a00 builtin/stash.c 1554) continue;
builtin/upload-archive.c
e001fd3a50 builtin/upload-archive.c 111) if (version == protocol_v0 ||
version == protocol_v1)
e001fd3a50 builtin/upload-archive.c 112) packet_write_fmt(1, "NACK
unable to spawn subprocess\n");
e001fd3a50 builtin/upload-archive.c 113) else if (version == protocol_v2)
e001fd3a50 builtin/upload-archive.c 114) error_clnt("unable to spawn
subprocess\n");
config.c
c780b9cfe8 2298) return val;
c780b9cfe8 2301) if (is_bool)
c780b9cfe8 2302) return val ? 0 : 1;
c780b9cfe8 2304) return val;
help.c
26c7d06783 help.c 500) static int get_alias(const char *var,
const char *value, void *data)
26c7d06783 help.c 502) struct string_list *list = data;
26c7d06783 help.c 504) if (skip_prefix(var, "alias.", &var))
26c7d06783 help.c 505) string_list_append(list, var)->util =
xstrdup(value);
26c7d06783 help.c 507) return 0;
26c7d06783 help.c 530) printf("\n%s\n", _("Command aliases"));
26c7d06783 help.c 531) ALLOC_ARRAY(aliases, alias_list.nr + 1);
26c7d06783 help.c 532) for (i = 0; i < alias_list.nr; i++) {
26c7d06783 help.c 533) aliases[i].name = alias_list.items[i].string;
26c7d06783 help.c 534) aliases[i].help = alias_list.items[i].util;
26c7d06783 help.c 535) aliases[i].category = 1;
26c7d06783 help.c 537) aliases[alias_list.nr].name = NULL;
26c7d06783 help.c 538) print_command_list(aliases, 1, longest);
26c7d06783 help.c 539) free(aliases);
http-backend.c
fb19d32f05 646) argv[1] = ".";
fb19d32f05 647) argv[2] = NULL;
list-objects-filter-options.c
bc5975d24f 55) if (errbuf) {
bc5975d24f 56) strbuf_addstr(
bc5975d24f 60) return 1;
cc0b05a4cc 86) if (errbuf)
list-objects-filter.c
696aa73905 47) BUG("unknown filter_situation: %d", filter_situation);
bc5975d24f 101) BUG("unknown filter_situation: %d", filter_situation);
696aa73905 152) BUG("unknown filter_situation: %d", filter_situation);
696aa73905 257) BUG("unknown filter_situation: %d", filter_situation);
696aa73905 438) BUG("invalid list-objects filter choice: %d",
list-objects.c
f447a499db 197) ctx->show_object(obj, base->buf, ctx->show_data);
midx.c
e43d2dcce1 285) if (hasheq(oid.hash,
e43d2dcce1 286) p->bad_object_sha1 + the_hash_algo->rawsz * i))
oidset.c
8b2f8cbcb1 29) kh_del_oid(&set->set, pos);
8b2f8cbcb1 30) return 1;
packfile.c
fb571c5fa7 117) return error("index file %s is too small", path);
fb571c5fa7 119) return error("empty data");
preload-index.c
ae9af12287 73) struct progress_data *pd = p->progress;
ae9af12287 75) pthread_mutex_lock(&pd->mutex);
ae9af12287 76) pd->n += last_nr - nr;
ae9af12287 77) display_progress(pd->progress, pd->n);
ae9af12287 78) pthread_mutex_unlock(&pd->mutex);
ae9af12287 79) last_nr = nr;
ae9af12287 93) struct progress_data *pd = p->progress;
ae9af12287 95) pthread_mutex_lock(&pd->mutex);
ae9af12287 96) display_progress(pd->progress, pd->n + last_nr);
ae9af12287 97) pthread_mutex_unlock(&pd->mutex);
ae9af12287 128) pd.progress = start_delayed_progress(_("Refreshing
index"), index->cache_nr);
ae9af12287 129) pthread_mutex_init(&pd.mutex, NULL);
ae9af12287 140) p->progress = &pd;
prio-queue.c
2d181390f3 94) return queue->array[queue->nr - 1].data;
read-cache.c
ae9af12287 1490) progress = start_delayed_progress(_("Refresh index"),
ae9af12287 1491) istate->cache_nr);
ae9af12287 1533) display_progress(progress, i);
ae9af12287 1566) display_progress(progress, istate->cache_nr);
ae9af12287 1567) stop_progress(&progress);
252d079cbd 1778) const unsigned char *cp = (const unsigned char *)name;
252d079cbd 1782) strip_len = decode_varint(&cp);
77ff1127a4 1783) if (previous_ce) {
77ff1127a4 1784) previous_len = previous_ce->ce_namelen;
77ff1127a4 1785) if (previous_len < strip_len)
252d079cbd 1786) die(_("malformed name field in the index, near path '%s'"),
77ff1127a4 1787) previous_ce->name);
77ff1127a4 1788) copy_len = previous_len - strip_len;
77ff1127a4 1790) copy_len = 0;
252d079cbd 1792) name = (const char *)cp;
252d079cbd 1798) len += copy_len;
252d079cbd 1819) if (copy_len)
252d079cbd 1820) memcpy(ce->name, previous_ce->name, copy_len);
252d079cbd 1821) memcpy(ce->name + copy_len, name, len + 1 - copy_len);
252d079cbd 1822) *ent_size = (name - ((char *)ondisk)) + len + 1 - copy_len;
abb4bb8384 1959) munmap((void *)p->mmap, p->mmap_size);
abb4bb8384 1960) die(_("index file corrupt"));
77ff1127a4 2001) mem_pool_init(&istate->ce_mem_pool,
77ff1127a4 2041) static void *load_cache_entries_thread(void *_data)
77ff1127a4 2043) struct load_cache_entries_thread_data *p = _data;
77ff1127a4 2047) for (i = p->ieot_start; i < p->ieot_start +
p->ieot_blocks; i++) {
77ff1127a4 2048) p->consumed += load_cache_entry_block(p->istate,
p->ce_mem_pool,
77ff1127a4 2049) p->offset, p->ieot->entries[i].nr, p->mmap,
p->ieot->entries[i].offset, NULL);
77ff1127a4 2050) p->offset += p->ieot->entries[i].nr;
77ff1127a4 2052) return NULL;
77ff1127a4 2055) static unsigned long load_cache_entries_threaded(struct
index_state *istate, const char *mmap, size_t mmap_size,
77ff1127a4 2060) unsigned long consumed = 0;
77ff1127a4 2063) if (istate->name_hash_initialized)
77ff1127a4 2064) BUG("the name hash isn't thread safe");
77ff1127a4 2066) mem_pool_init(&istate->ce_mem_pool, 0);
77ff1127a4 2069) if (nr_threads > ieot->nr)
77ff1127a4 2070) nr_threads = ieot->nr;
77ff1127a4 2071) data = xcalloc(nr_threads, sizeof(*data));
77ff1127a4 2073) offset = ieot_start = 0;
77ff1127a4 2074) ieot_blocks = DIV_ROUND_UP(ieot->nr, nr_threads);
77ff1127a4 2075) for (i = 0; i < nr_threads; i++) {
77ff1127a4 2076) struct load_cache_entries_thread_data *p = &data[i];
77ff1127a4 2079) if (ieot_start + ieot_blocks > ieot->nr)
77ff1127a4 2080) ieot_blocks = ieot->nr - ieot_start;
77ff1127a4 2082) p->istate = istate;
77ff1127a4 2083) p->offset = offset;
77ff1127a4 2084) p->mmap = mmap;
77ff1127a4 2085) p->ieot = ieot;
77ff1127a4 2086) p->ieot_start = ieot_start;
77ff1127a4 2087) p->ieot_blocks = ieot_blocks;
77ff1127a4 2090) nr = 0;
77ff1127a4 2091) for (j = p->ieot_start; j < p->ieot_start +
p->ieot_blocks; j++)
77ff1127a4 2092) nr += p->ieot->entries[j].nr;
77ff1127a4 2093) if (istate->version == 4) {
77ff1127a4 2094) mem_pool_init(&p->ce_mem_pool,
77ff1127a4 2097) mem_pool_init(&p->ce_mem_pool,
77ff1127a4 2101) err = pthread_create(&p->pthread, NULL,
load_cache_entries_thread, p);
77ff1127a4 2102) if (err)
77ff1127a4 2103) die(_("unable to create load_cache_entries thread:
%s"), strerror(err));
77ff1127a4 2106) for (j = 0; j < ieot_blocks; j++)
77ff1127a4 2107) offset += ieot->entries[ieot_start + j].nr;
77ff1127a4 2108) ieot_start += ieot_blocks;
77ff1127a4 2111) for (i = 0; i < nr_threads; i++) {
77ff1127a4 2112) struct load_cache_entries_thread_data *p = &data[i];
77ff1127a4 2114) err = pthread_join(p->pthread, NULL);
77ff1127a4 2115) if (err)
77ff1127a4 2116) die(_("unable to join load_cache_entries thread: %s"),
strerror(err));
77ff1127a4 2117) mem_pool_combine(istate->ce_mem_pool, p->ce_mem_pool);
77ff1127a4 2118) consumed += p->consumed;
77ff1127a4 2121) free(data);
77ff1127a4 2123) return consumed;
77ff1127a4 2192) nr_threads = cpus;
abb4bb8384 2196) extension_offset = read_eoie_extension(mmap, mmap_size);
abb4bb8384 2197) if (extension_offset) {
abb4bb8384 2200) p.src_offset = extension_offset;
abb4bb8384 2201) err = pthread_create(&p.pthread, NULL,
load_index_extensions, &p);
abb4bb8384 2202) if (err)
abb4bb8384 2203) die(_("unable to create load_index_extensions thread:
%s"), strerror(err));
abb4bb8384 2205) nr_threads--;
77ff1127a4 2214) ieot = read_ieot_extension(mmap, mmap_size,
extension_offset);
77ff1127a4 2217) src_offset += load_cache_entries_threaded(istate, mmap,
mmap_size, src_offset, nr_threads, ieot);
77ff1127a4 2218) free(ieot);
abb4bb8384 2232) int ret = pthread_join(p.pthread, NULL);
abb4bb8384 2233) if (ret)
abb4bb8384 2234) die(_("unable to join load_index_extensions thread:
%s"), strerror(ret));
3255089ada 2775) ieot_blocks = nr_threads;
77ff1127a4 2776) if (ieot_blocks > istate->cache_nr)
77ff1127a4 2777) ieot_blocks = istate->cache_nr;
3255089ada 2785) ieot = xcalloc(1, sizeof(struct index_entry_offset_table)
3255089ada 2786) + (ieot_blocks * sizeof(struct index_entry_offset)));
77ff1127a4 2787) ieot_entries = DIV_ROUND_UP(entries, ieot_blocks);
3255089ada 2794) free(ieot);
3b1d9e045e 2795) return -1;
3255089ada 2821) ieot->entries[ieot->nr].nr = nr;
3255089ada 2822) ieot->entries[ieot->nr].offset = offset;
3255089ada 2823) ieot->nr++;
3255089ada 2829) if (previous_name)
3255089ada 2830) previous_name->buf[0] = 0;
3255089ada 2831) nr = 0;
3255089ada 2832) offset = lseek(newfd, 0, SEEK_CUR);
3255089ada 2833) if (offset < 0) {
3255089ada 2834) free(ieot);
3255089ada 2835) return -1;
3255089ada 2837) offset += write_buffer_len;
3255089ada 2847) ieot->entries[ieot->nr].nr = nr;
3255089ada 2848) ieot->entries[ieot->nr].offset = offset;
3255089ada 2849) ieot->nr++;
3255089ada 2861) free(ieot);
3b1d9e045e 2862) return -1;
3255089ada 2876) struct strbuf sb = STRBUF_INIT;
3255089ada 2878) write_ieot_extension(&sb, ieot);
3255089ada 2879) err = write_index_ext_header(&c, &eoie_c, newfd,
CACHE_EXT_INDEXENTRYOFFSETTABLE, sb.len) < 0
3255089ada 2880) || ce_write(&c, newfd, sb.buf, sb.len) < 0;
3255089ada 2881) strbuf_release(&sb);
3255089ada 2882) free(ieot);
3255089ada 2883) if (err)
3255089ada 2884) return -1;
3b1d9e045e 3372) static size_t read_eoie_extension(const char *mmap,
size_t mmap_size)
3b1d9e045e 3390) if (mmap_size < sizeof(struct cache_header) +
EOIE_SIZE_WITH_HEADER + the_hash_algo->rawsz)
3b1d9e045e 3391) return 0;
3b1d9e045e 3394) index = eoie = mmap + mmap_size - EOIE_SIZE_WITH_HEADER
- the_hash_algo->rawsz;
3b1d9e045e 3395) if (CACHE_EXT(index) != CACHE_EXT_ENDOFINDEXENTRIES)
3b1d9e045e 3396) return 0;
3b1d9e045e 3397) index += sizeof(uint32_t);
3b1d9e045e 3400) extsize = get_be32(index);
3b1d9e045e 3401) if (extsize != EOIE_SIZE)
3b1d9e045e 3402) return 0;
3b1d9e045e 3403) index += sizeof(uint32_t);
3b1d9e045e 3409) offset = get_be32(index);
3b1d9e045e 3410) if (mmap + offset < mmap + sizeof(struct cache_header))
3b1d9e045e 3411) return 0;
3b1d9e045e 3412) if (mmap + offset >= eoie)
3b1d9e045e 3413) return 0;
3b1d9e045e 3414) index += sizeof(uint32_t);
3b1d9e045e 3425) src_offset = offset;
3b1d9e045e 3426) the_hash_algo->init_fn(&c);
3b1d9e045e 3427) while (src_offset < mmap_size - the_hash_algo->rawsz -
EOIE_SIZE_WITH_HEADER) {
3b1d9e045e 3435) memcpy(&extsize, mmap + src_offset + 4, 4);
3b1d9e045e 3436) extsize = ntohl(extsize);
3b1d9e045e 3439) if (src_offset + 8 + extsize < src_offset)
3b1d9e045e 3440) return 0;
3b1d9e045e 3442) the_hash_algo->update_fn(&c, mmap + src_offset, 8);
3b1d9e045e 3444) src_offset += 8;
3b1d9e045e 3445) src_offset += extsize;
3b1d9e045e 3447) the_hash_algo->final_fn(hash, &c);
3b1d9e045e 3448) if (!hasheq(hash, (const unsigned char *)index))
3b1d9e045e 3449) return 0;
3b1d9e045e 3452) if (src_offset != mmap_size - the_hash_algo->rawsz -
EOIE_SIZE_WITH_HEADER)
3b1d9e045e 3453) return 0;
3b1d9e045e 3455) return offset;
3255089ada 3475) static struct index_entry_offset_table
*read_ieot_extension(const char *mmap, size_t mmap_size, size_t offset)
3255089ada 3477) const char *index = NULL;
3255089ada 3483) if (!offset)
3255089ada 3484) return NULL;
3255089ada 3485) while (offset <= mmap_size -
the_hash_algo->rawsz - 8) {
3255089ada 3486) extsize = get_be32(mmap + offset + 4);
3255089ada 3487) if (CACHE_EXT((mmap + offset)) ==
CACHE_EXT_INDEXENTRYOFFSETTABLE) {
3255089ada 3488) index = mmap + offset + 4 + 4;
3255089ada 3489) break;
3255089ada 3491) offset += 8;
3255089ada 3492) offset += extsize;
3255089ada 3494) if (!index)
3255089ada 3495) return NULL;
3255089ada 3498) ext_version = get_be32(index);
3255089ada 3499) if (ext_version != IEOT_VERSION) {
3255089ada 3500) error("invalid IEOT version %d", ext_version);
3255089ada 3501) return NULL;
3255089ada 3503) index += sizeof(uint32_t);
3255089ada 3506) nr = (extsize - sizeof(uint32_t)) /
(sizeof(uint32_t) + sizeof(uint32_t));
3255089ada 3507) if (!nr) {
3255089ada 3508) error("invalid number of IEOT entries %d", nr);
3255089ada 3509) return NULL;
3255089ada 3511) ieot = xmalloc(sizeof(struct
index_entry_offset_table)
3255089ada 3512) + (nr * sizeof(struct index_entry_offset)));
3255089ada 3513) ieot->nr = nr;
3255089ada 3514) for (i = 0; i < nr; i++) {
3255089ada 3515) ieot->entries[i].offset = get_be32(index);
3255089ada 3516) index += sizeof(uint32_t);
3255089ada 3517) ieot->entries[i].nr = get_be32(index);
3255089ada 3518) index += sizeof(uint32_t);
3255089ada 3521) return ieot;
3255089ada 3524) static void write_ieot_extension(struct strbuf *sb,
struct index_entry_offset_table *ieot)
3255089ada 3530) put_be32(&buffer, IEOT_VERSION);
3255089ada 3531) strbuf_add(sb, &buffer, sizeof(uint32_t));
3255089ada 3534) for (i = 0; i < ieot->nr; i++) {
3255089ada 3537) put_be32(&buffer, ieot->entries[i].offset);
3255089ada 3538) strbuf_add(sb, &buffer, sizeof(uint32_t));
3255089ada 3541) put_be32(&buffer, ieot->entries[i].nr);
3255089ada 3542) strbuf_add(sb, &buffer, sizeof(uint32_t));
3255089ada 3544) }
rebase-interactive.c
64a43cbd5d 61) return error_errno(_("could not read '%s'."), todo_file);
64a43cbd5d 65) strbuf_release(&buf);
64a43cbd5d 66) return -1;
a9f5476fbc 74) return error_errno(_("could not read '%s'."), todo_file);
a9f5476fbc 78) strbuf_release(&buf);
a9f5476fbc 79) return -1;
64a43cbd5d 85) return -1;
revision.c
4943d28849 2931) return;
4943d28849 2934) return;
4943d28849 2937) c->object.flags |= UNINTERESTING;
4943d28849 2940) return;
4943d28849 2943) mark_parents_uninteresting(c);
4943d28849 2966) return;
4943d28849 2969) return;
4943d28849 2974) return;
4943d28849 3019) info->topo_queue.compare = compare_commits_by_commit_date;
4943d28849 3020) break;
4943d28849 3022) init_author_date_slab(&info->author_date);
4943d28849 3023) info->topo_queue.compare = compare_commits_by_author_date;
4943d28849 3024) info->topo_queue.cb_data = &info->author_date;
4943d28849 3025) break;
4943d28849 3038) continue;
4943d28849 3048) record_author_date(&info->author_date, c);
6c04ff3001 3086) if (!revs->ignore_missing_links)
6c04ff3001 3087) die("Failed to traverse parents of commit %s",
4943d28849 3088) oid_to_hex(&commit->object.oid));
4943d28849 3096) continue;
sequencer.c
65850686cf 2276) return;
65850686cf 2373) write_file(rebase_path_quiet(), "%s\n", quiet);
2c58483a59 3371) return error(_("could not checkout %s"), commit);
4df66c40b0 3385) return error(_("%s: not a valid OID"), orig_head);
b97e187364 4748) return -1;
b97e187364 4751) return -1;
b97e187364 4757) return error_errno(_("could not read '%s'."), todo_file);
b97e187364 4760) todo_list_release(&todo_list);
b97e187364 4761) return error(_("unusable todo list: '%s'"), todo_file);
b97e187364 4780) todo_list_release(&todo_list);
b97e187364 4781) return -1;
b97e187364 4785) return error(_("could not copy '%s' to '%s'."), todo_file,
b97e187364 4789) return error(_("could not transform the todo list"));
b97e187364 4818) return error(_("could not transform the todo list"));
b97e187364 4821) return error(_("could not skip unnecessary pick
commands"));
b97e187364 4827) return -1;
split-index.c
568f3a6073 310) const unsigned int ondisk_flags =
568f3a6073 314) ce_flags = ce->ce_flags;
568f3a6073 315) base_flags = base->ce_flags;
568f3a6073 317) ce->ce_flags &= ondisk_flags;
568f3a6073 318) base->ce_flags &= ondisk_flags;
568f3a6073 319) ret = memcmp(&ce->ce_stat_data, &base->ce_stat_data,
568f3a6073 322) ce->ce_flags = ce_flags;
568f3a6073 323) base->ce_flags = base_flags;
568f3a6073 324) if (ret)
568f3a6073 325) ce->ce_flags |= CE_UPDATE_IN_BASE;
strbuf.c
f95736288a 127) --sb->len;
transport-helper.c
fb19d32f05 643) if (!data->connect && !data->stateless_connect)
transport.c
99bcb883cb 299) BUG("buffer must be empty at the end of handshake()");
Commits introducing uncovered code:
Alban Gruin 0af129b2e: rebase--interactive2: rewrite the submodes
of interactive rebase in C
Alban Gruin 2c58483a5: rebase -i: rewrite setup_reflog_action() in C
Alban Gruin 34b47315d: rebase -i: move rebase--helper modes to
rebase--interactive
Alban Gruin 4df66c40b: rebase -i: rewrite checkout_onto() in C
Alban Gruin 53bbcfbde: rebase -i: implement the main part of
interactive rebase as a builtin
Alban Gruin 64a43cbd5: rebase -i: rewrite the edit-todo
functionality in C
Alban Gruin 65850686c: rebase -i: rewrite write_basic_state() in C
Alban Gruin a9f5476fb: sequencer: refactor append_todo_help() to
write its message to a buffer
Alban Gruin b97e18736: rebase -i: rewrite complete_action() in C
Ben Peart 3255089ad: ieot: add Index Entry Offset Table (IEOT)
extension
Ben Peart 3b1d9e045: eoie: add End of Index Entry (EOIE) extension
Ben Peart 77ff1127a: read-cache: load cache entries on worker threads
Ben Peart abb4bb838: read-cache: load cache extensions on a worker
thread
Ben Peart c780b9cfe: config: add new index.threads config setting
Derrick Stolee 2d181390f: prio-queue: add 'peek' operation
Derrick Stolee 4943d2884: revision.c: refactor basic topo-order logic
Derrick Stolee 6c04ff300: revision.c: begin refactoring
--topo-order logic
Harald Nordgren 250edfa8c: bisect: create 'bisect_flags' parameter
in find_bisection()
Jeff King e43d2dcce: more oideq/hasheq conversions
Joel Teichroeb 31f109a36: stash: convert drop and clear to builtin
Joel Teichroeb 93871263d: stash: convert apply to builtin
Joel Teichroeb b3513da4b: stash: convert pop to builtin
Johannes Schindelin bc24382c2: builtin rebase: prepare for builtin
rebase -i
Jonathan Tan 99bcb883c: transport: allow skipping of ref listing
Josh Steadmon e001fd3a5: archive: implement protocol v2 archive command
Josh Steadmon fb19d32f0: archive: allow archive over HTTP(S) with
proto v2
Josh Steadmon fb571c5fa: fuzz: add fuzz testing for packfile indices
Junio C Hamano 7bbc53a58: fetch: replace string-list used as a
look-up table with a hashmap
Matthew DeVore 696aa7390: list-objects-filter: use BUG rather than die
Matthew DeVore 7c0fe330d: rev-list: handle missing tree objects
properly
Matthew DeVore bc5975d24: list-objects-filter: implement filter tree:0
Matthew DeVore cc0b05a4c: list-objects-filter-options: do not
over-strbuf_init
Matthew DeVore f447a499d: list-objects: store common func args in
struct
Nguyễn Thái Ngọc Duy 252d079cb: read-cache.c: optimize reading
index format v4
Nguyễn Thái Ngọc Duy 26c7d0678: help -a: improve and make --verbose
default
Nguyễn Thái Ngọc Duy ae9af1228: status: show progress bar if
refreshing the index takes too long
Paul-Sebastian Ungureanu 0ac06fb81: stash: convert store to builtin
Paul-Sebastian Ungureanu 129f0b0a0: stash: convert
`stash--helper.c` into `stash.c`
Paul-Sebastian Ungureanu 48c061fa4: stash: convert push to builtin
Paul-Sebastian Ungureanu 8002b9e62: stash: make push -q quiet
Paul-Sebastian Ungureanu 8ceb24b2c: stash: convert show to builtin
Paul-Sebastian Ungureanu f6f191b3f: stash: convert create to builtin
Pratik Karki 002ee2fe6: builtin rebase: support `keep-empty` option
Pratik Karki 0eabf4b95: builtin rebase: stop if `git am` is in progress
Pratik Karki 12026a412: builtin rebase: support `--gpg-sign` option
Pratik Karki 122420c29: builtin rebase: support --skip
Pratik Karki 1ed9c14ff: builtin rebase: support --force-rebase
Pratik Karki 3c3588c7d: builtin rebase: support
--rebase-merges[=[no-]rebase-cousins]
Pratik Karki 51e9ea6da: builtin rebase: support --edit-todo and
--show-current-patch
Pratik Karki 53f9e5be9: builtin rebase: support `ignore-date` option
Pratik Karki 55071ea24: rebase: start implementing it as a builtin
Pratik Karki 5a6149453: builtin rebase: support --quit
Pratik Karki 5e5d96197: builtin rebase: support --abort
Pratik Karki 6defce2b0: builtin rebase: support `--autostash` option
Pratik Karki 73d51ed0a: builtin rebase: support --signoff
Pratik Karki 7998dbe1e: builtin rebase: support `-C` and
`--whitespace=<type>`
Pratik Karki 9a48a615b: builtin rebase: try to fast forward when
possible
Pratik Karki 9dba809a6: builtin rebase: support --root
Pratik Karki ac7f467fe: builtin/rebase: support running "git rebase
<upstream>"
Pratik Karki ba1905a5f: builtin rebase: add support for custom
merge strategies
Pratik Karki bff014dac: builtin rebase: support the `verbose` and
`diffstat` options
Pratik Karki c54dacb50: builtin rebase: start a new rebase only if
none is in progress
Pratik Karki cda614e48: builtin rebase: show progress when
connected to a terminal
Pratik Karki d4c569f8f: builtin rebase: only store fully-qualified
refs in `options.head_name`
Pratik Karki e0333e5c6: builtin rebase: require a clean worktree
Pratik Karki e65123a71: builtin rebase: support `git rebase
<upstream> <switch-to>`
Pratik Karki ead98c111: builtin rebase: support --rerere-autoupdate
Pratik Karki f28d40d3a: builtin rebase: support --onto
Pratik Karki f95736288: builtin rebase: support --continue
René Scharfe 8b2f8cbcb: oidset: use khash
SZEDER Gábor 568f3a607: split-index: don't compare stat data of
entries already marked for split index
Uncovered code new in next, compared to master. Build logs at [2]
-----------------------------------------------------------------
blame.c
a470beea39 113) !strcmp(r->index->cache[-1 - pos]->name, path))
a470beea39 272) int pos = index_name_pos(r->index, path, len);
a470beea39 274) mode = r->index->cache[pos]->ce_mode;
builtin/am.c
2abf350385 1414) repo_init_revisions(the_repository, &rev_info, NULL);
builtin/gc.c
3029970275 builtin/gc.c 461) ret = error_errno(_("cannot stat '%s'"),
gc_log_path);
3029970275 builtin/gc.c 470) ret = error_errno(_("cannot read '%s'"),
gc_log_path);
fec2ed2187 builtin/gc.c 495) die(FAILED_RUN, pack_refs_cmd.argv[0]);
fec2ed2187 builtin/gc.c 498) die(FAILED_RUN, reflog.argv[0]);a
3029970275 builtin/gc.c 585) exit(128);
fec2ed2187 builtin/gc.c 637) die(FAILED_RUN, repack.argv[0]);
fec2ed2187 builtin/gc.c 647) die(FAILED_RUN, prune.argv[0]);
fec2ed2187 builtin/gc.c 654) die(FAILED_RUN, prune_worktrees.argv[0]);
fec2ed2187 builtin/gc.c 658) die(FAILED_RUN, rerere.argv[0]);
builtin/pack-objects.c
2fa233a554 builtin/pack-objects.c 1512) hashcpy(base_oid.hash, base_sha1);
2fa233a554 builtin/pack-objects.c 1513) if
(!in_same_island(&delta->idx.oid, &base_oid))
2fa233a554 builtin/pack-objects.c 1514) return 0;
commit-graph.c
20fd6d5799 79) return 0;
diff.c
b78ea5fc35 4117) add_external_diff_name(o->repo, &argv, other, two);
refs.c
4a6067cda5 1405) return 0;
revision.c
2abf350385 1526) if (ce_path_match(istate, ce, &revs->prune_data, NULL)) {
2abf350385 1532) while ((i+1 < istate->cache_nr) &&
2abf350385 1533) ce_same_name(ce, istate->cache[i+1]))
wt-status.c
f3bd35fa0d 671) s->committable = 1;
73ba5d78b4 1953) if (s->state.rebase_in_progress ||
73ba5d78b4 1954) s->state.rebase_interactive_in_progress)
73ba5d78b4 1955) branch_name = s->state.onto;
73ba5d78b4 1956) else if (s->state.detached_from)
73ba5d78b4 1957) branch_name = s->state.detached_from;
Commits introducing uncovered code:
Derrick Stolee 20fd6d579: commit-graph: not compatible with grafts
Jeff King 2fa233a55: pack-objects: handle island check for
"external" delta base
Jonathan Nieder 302997027: gc: do not return error for prior errors
in daemonized mode
Jonathan Nieder fec2ed218: gc: exit with status 128 on failure
Nguyễn Thái Ngọc Duy 2abf35038: revision.c: remove implicit
dependency on the_index
Nguyễn Thái Ngọc Duy a470beea3: blame.c: rename "repo" argument to "r"
Nguyễn Thái Ngọc Duy b78ea5fc3: diff.c: reduce implicit dependency
on the_index
Stefan Beller 4a6067cda: refs.c: migrate internal ref iteration to
pass thru repository argument
Stephen P. Smith 73ba5d78b: roll wt_status_state into wt_status and
populate in the collect phase
Stephen P. Smith f3bd35fa0: wt-status.c: set the committable flag
in the collect phase
Uncovered code new in master, compared to maint. Build logs at [3]
------------------------------------------------------------------
builtin/checkout.c
fa655d8411 builtin/checkout.c 537) return 0;
fa655d8411 builtin/checkout.c 951) return error(_("index file corrupt"));
builtin/commit.c
859fdc0c3c builtin/commit.c 1657)
write_commit_graph_reachable(get_object_directory(), 0);
builtin/difftool.c
4a7e27e957 441) if (oideq(&loid, &roid))
builtin/fsck.c
454ea2e4d7 builtin/fsck.c 743) for (p = get_all_packs(the_repository); p;
66ec0390e7 builtin/fsck.c 862) midx_argv[2] = "--object-dir";
66ec0390e7 builtin/fsck.c 863) midx_argv[3] = alt->path;
66ec0390e7 builtin/fsck.c 864) if (run_command(&midx_verify))
66ec0390e7 builtin/fsck.c 865) errors_found |= ERROR_COMMIT_GRAPH;
builtin/log.c
2e6fd71a52 builtin/log.c 1461) die(_("failed to infer range-diff ranges"));
ee6cbf712e builtin/log.c 1807) die(_("--interdiff requires
--cover-letter or single patch"));
8631bf1cdd builtin/log.c 1817) else if (!rdiff_prev)
8631bf1cdd builtin/log.c 1818) die(_("--creation-factor requires
--range-diff"));
40ce41604d builtin/log.c 1822) die(_("--range-diff requires
--cover-letter or single patch"));
builtin/multi-pack-index.c
6d68e6a461 35) usage_with_options(builtin_multi_pack_index_usage,
6d68e6a461 39) die(_("too many arguments"));
6d68e6a461 48) die(_("unrecognized verb: %s"), argv[0]);
builtin/pack-objects.c
6a22d52126 builtin/pack-objects.c 1079) if (fill_midx_entry(oid, &e, m)) {
6a22d52126 builtin/pack-objects.c 1080) struct packed_git *p = e.p;
6a22d52126 builtin/pack-objects.c 1083) if (p == *found_pack)
6a22d52126 builtin/pack-objects.c 1084) offset = *found_offset;
6a22d52126 builtin/pack-objects.c 1086) offset =
find_pack_entry_one(oid->hash, p);
6a22d52126 builtin/pack-objects.c 1088) if (offset) {
6a22d52126 builtin/pack-objects.c 1089) if (!*found_pack) {
6a22d52126 builtin/pack-objects.c 1090) if (!is_pack_valid(p))
6a22d52126 builtin/pack-objects.c 1091) continue;
6a22d52126 builtin/pack-objects.c 1092) *found_offset = offset;
6a22d52126 builtin/pack-objects.c 1093) *found_pack = p;
6a22d52126 builtin/pack-objects.c 1095) want =
want_found_object(exclude, p);
6a22d52126 builtin/pack-objects.c 1096) if (want != -1)
6a22d52126 builtin/pack-objects.c 1097) return want;
28b8a73080 builtin/pack-objects.c 2778) depth++;
108f530385 builtin/pack-objects.c 2782) oe_set_tree_depth(&to_pack, ent,
depth);
454ea2e4d7 builtin/pack-objects.c 2966) p = get_all_packs(the_repository);
builtin/pack-redundant.c
454ea2e4d7 builtin/pack-redundant.c 580) struct packed_git *p =
get_all_packs(the_repository);
454ea2e4d7 builtin/pack-redundant.c 595) struct packed_git *p =
get_all_packs(the_repository);
builtin/remote.c
5025425dff builtin/remote.c 864) return error(_("No such remote:
'%s'"), name);
builtin/repack.c
16d75fa48d 48) use_delta_islands = git_config_bool(var, value);
16d75fa48d 49) return 0;
builtin/rerere.c
2373b65059 builtin/rerere.c 78) warning(_("'git rerere forget' without
paths is deprecated"));
2373b65059 builtin/rerere.c 110) die(_("unable to generate diff for
'%s'"), rerere_path(id, NULL));
builtin/show-branch.c
9001dc2a74 builtin/show-branch.c 430) if (get_oid(refname + ofs, &tmp)
|| !oideq(&tmp, oid))
builtin/submodule--helper.c
ee69b2a90c 1462) die(_("Invalid update mode '%s' for submodule path '%s'"),
ee69b2a90c 1466) die(_("Invalid update mode '%s' configured for
submodule path '%s'"),
ee69b2a90c 1469) trace_printf("loaded thing");
ee69b2a90c 1470) out->type = sub->update_strategy.type;
ee69b2a90c 1471) out->command = sub->update_strategy.command;
ee69b2a90c 1491) die("submodule--helper update-module-clone expects
<just-cloned> <path> [<update>]");
74d4731da1 2033) BUG("submodule--helper connect-gitdir-workingtree
<name> <path>");
74d4731da1 2039) BUG("We could get the submodule handle before?");
74d4731da1 2042) die(_("could not get a repository handle for submodule
'%s'"), path);
builtin/unpack-objects.c
4a7e27e957 builtin/unpack-objects.c 306) if (oideq(&info->base_oid,
&obj_list[nr].oid) ||
builtin/update-index.c
4a7e27e957 builtin/update-index.c 672) if (oideq(&ce_2->oid, &ce_3->oid) &&
builtin/worktree.c
e5353bef55 60) error_errno(_("failed to delete '%s'"), sb.buf);
e19831c94f 251) die(_("unable to re-add worktree '%s'"), path);
68a6b3a1bd 793) die(_("cannot move a locked working tree, lock reason:
%s\nuse 'move -f -f' to override or unlock first"),
f4143101cb 906) die(_("cannot remove a locked working tree, lock reason:
%s\nuse 'remove -f -f' to override or unlock first"),
cache-tree.c
4592e6080f 762) BUG("%s with flags 0x%x should not be in cache-tree",
4592e6080f 770) BUG("bad subtree '%.*s'", entlen, name);
4592e6080f 785) BUG("cache-tree for path %.*s does not match. "
commit-graph.c
6cc017431c 247) return 0;
commit-reach.c
5227c38566 63) BUG("bad generation skip %8x > %8x at %s",
5227c38566 134) return ret;
5227c38566 282) return 1;
5227c38566 314) return ret;
5227c38566 317) return ret;
5227c38566 323) return ret;
1d614d41e5 395) return 0;
1d614d41e5 401) return 0;
1d614d41e5 405) return 0;
920f93ca1c 459) return CONTAINS_NO;
920f93ca1c 484) cutoff = c->generation;
b67f6b26e3 559) continue;
b67f6b26e3 569) from->objects[i].item->flags |= assign_flag;
b67f6b26e3 570) continue;
b67f6b26e3 576) result = 0;
b67f6b26e3 577) goto cleanup;
delta-islands.c
c8d521faf7 53) memcpy(b, old, size);
c8d521faf7 73) return 1;
c8d521faf7 118) return 0;
c8d521faf7 130) return 0;
c8d521faf7 187) b->refcount--;
c8d521faf7 188) b = kh_value(island_marks, pos) = island_bitmap_new(b);
c8d521faf7 202) continue;
c8d521faf7 212) obj = ((struct tag *)obj)->tagged;
c8d521faf7 213) if (obj) {
c8d521faf7 214) parse_object(the_repository, &obj->oid);
c8d521faf7 215) marks = create_or_get_island_marks(obj);
c8d521faf7 216) island_bitmap_set(marks, island_counter);
c8d521faf7 248) return;
c8d521faf7 268) progress_state = start_progress(_("Propagating island
marks"), nr);
c8d521faf7 286) die(_("bad tree object %s"), oid_to_hex(&ent->idx.oid));
c8d521faf7 293) continue;
c8d521faf7 297) continue;
c8d521faf7 321) return config_error_nonbool(k);
c8d521faf7 330) die(_("failed to load island regex for '%s': %s"), k,
re.buf);
c8d521faf7 386) warning(_("island regex from config has "
c8d521faf7 397) strbuf_addch(&island_name, '-');
c8d521faf7 433) continue;
c8d521faf7 436) list[dst] = list[src];
diff-lib.c
9001dc2a74 diff-lib.c 345) (!oideq(oid, &old_entry->oid) ||
!oideq(&old_entry->oid, &new_entry->oid))) {
dir.c
c46c406ae1 2275) trace_performance_leave("read directory %.*s", len, path);
entry.c
b878579ae7 402) static void mark_colliding_entries(const struct checkout
*state,
b878579ae7 405) int i, trust_ino = check_stat;
b878579ae7 411) ce->ce_flags |= CE_MATCHED;
b878579ae7 413) for (i = 0; i < state->istate->cache_nr; i++) {
b878579ae7 414) struct cache_entry *dup = state->istate->cache[i];
b878579ae7 416) if (dup == ce)
b878579ae7 417) break;
b878579ae7 419) if (dup->ce_flags & (CE_MATCHED | CE_VALID |
CE_SKIP_WORKTREE))
b878579ae7 420) continue;
b878579ae7 422) if ((trust_ino && dup->ce_stat_data.sd_ino == st->st_ino) ||
b878579ae7 423) (!trust_ino && !fspathcmp(ce->name, dup->name))) {
b878579ae7 424) dup->ce_flags |= CE_MATCHED;
b878579ae7 425) break;
b878579ae7 428) }
b878579ae7 488) mark_colliding_entries(state, ce, &st);
fsck.c
fb8952077d 214) die_errno("Could not read '%s'", path);
ll-merge.c
d64324cb60 379) marker_size = DEFAULT_CONFLICT_MARKER_SIZE;
log-tree.c
4a7e27e957 477) if (oideq(&parent->item->object.oid, oid))
mailinfo.c
3aa4d81f88 992) len--;
3aa4d81f88 998) handle_filter(mi, prev);
3aa4d81f88 999) strbuf_reset(prev);
3aa4d81f88 1090) handle_filter(mi, &prev);
midx.c
4d80560c54 58) error_errno(_("failed to read %s"), midx_name);
4d80560c54 59) goto cleanup_fail;
4d80560c54 65) error(_("multi-pack-index file %s is too small"),
midx_name);
4d80560c54 66) goto cleanup_fail;
0d5b3a5ef7 146) die(_("multi-pack-index missing required OID lookup
chunk"));
662148c435 148) die(_("multi-pack-index missing required object offsets
chunk"));
4d80560c54 173) munmap(midx_map, midx_size);
4d80560c54 175) close(fd);
a40498a126 188) close_pack(m->packs[i]);
a40498a126 189) free(m->packs);
3715a6335c 262) return 0;
3715a6335c 278) return 0;
c39b02ae0a 283) nth_midxed_object_oid(&oid, m, pos);
c39b02ae0a 284) for (i = 0; i < p->num_bad_objects; i++)
c39b02ae0a 285) if (!hashcmp(oid.hash,
c39b02ae0a 286) p->bad_object_sha1 + the_hash_algo->rawsz * i))
c39b02ae0a 287) return 0;
c4d25228eb 341) return 1;
396f257018 398) warning(_("failed to add packfile '%s'"),
396f257018 400) return;
fe1ed56f5e 404) warning(_("failed to open pack-index '%s'"),
fe1ed56f5e 406) close_pack(packs->list[packs->nr]);
fe1ed56f5e 407) FREE_AND_NULL(packs->list[packs->nr]);
fe1ed56f5e 408) return;
a40498a126 481) return 1;
fe1ed56f5e 498) die(_("failed to locate object %d in packfile"),
cur_object);
32f3c541e3 611) BUG("incorrect pack-file order: %s before %s",
0d5b3a5ef7 673) BUG("OIDs not in order: %s >= %s",
662148c435 700) BUG("object %s requires a large offset (%"PRIx64") but
the MIDX is not writing large offsets!",
fc59e74844 754) die_errno(_("unable to create leading directories of %s"),
a40498a126 784) goto cleanup;
32f3c541e3 843) BUG("incorrect chunk offsets: %"PRIu64" before %"PRIu64,
32f3c541e3 848) BUG("chunk offset %"PRIu64" is not properly aligned",
32f3c541e3 860) BUG("incorrect chunk offset (%"PRIu64" != %"PRIu64")
for chunk id %"PRIx32,
32f3c541e3 887) BUG("trying to write unknown chunk id %"PRIx32,
32f3c541e3 893) BUG("incorrect final offset %"PRIu64" != %"PRIu64,
525e18c04b 923) die(_("failed to clear multi-pack-index at %s"), midx);
56ee7ff156 949) return 0;
cc6af73c02 990) midx_report(_("failed to load pack-index for packfile %s"),
cc6af73c02 991) e.p->pack_name);
cc6af73c02 992) break;
pack-bitmap.c
30cdc33fba 1130) return 0;
pack-objects.c
108f530385 169) REALLOC_ARRAY(pdata->tree_depth, pdata->nr_alloc);
fe0ac2fb7f 172) REALLOC_ARRAY(pdata->layer, pdata->nr_alloc);
108f530385 189) pdata->tree_depth[pdata->nr_objects - 1] = 0;
fe0ac2fb7f 192) pdata->layer[pdata->nr_objects - 1] = 0;
packfile.c
fe1ed56f5e 205) if (open_pack_index(p))
fe1ed56f5e 206) return 0;
fe1ed56f5e 207) level1_ofs = p->index_data;
17c35c8969 479) break;
17c35c8969 483) return error("packfile %s index unavailable",
p->pack_name);
17c35c8969 537) return 0;
refs/packed-backend.c
9001dc2a74 1163) } else if (!oideq(&update->old_oid, iter->oid)) {
refs/ref-cache.c
9001dc2a74 275) if (!oideq(&ref1->u.value.oid, &ref2->u.value.oid))
rerere.c
2373b65059 216) die(_("corrupt MERGE_RR"));
2373b65059 225) die(_("corrupt MERGE_RR"));
2373b65059 228) die(_("corrupt MERGE_RR"));
2373b65059 263) die(_("unable to write rerere record"));
2373b65059 268) die(_("unable to write rerere record"));
4af32207bc 375) break;
4af32207bc 379) strbuf_addbuf(&two, &conflict);
c0f16f8e14 383) break;
c0f16f8e14 387) break;
c0f16f8e14 391) break;
2373b65059 476) return error_errno(_("could not open '%s'"), path);
2373b65059 481) error_errno(_("could not write '%s'"), output);
2373b65059 491) error(_("there were errors while writing '%s' (%s)"),
2373b65059 494) io.io.wrerror = error_errno(_("failed to flush '%s'"),
path);
2373b65059 560) return error(_("index file corrupt"));
2373b65059 593) return error(_("index file corrupt"));
2373b65059 676) warning_errno(_("failed utime() on '%s'"),
2373b65059 682) return error_errno(_("could not open '%s'"), path);
2373b65059 684) error_errno(_("could not write '%s'"), path);
2373b65059 686) return error_errno(_("writing '%s' failed"), path);
2373b65059 712) die(_("unable to write new index file"));
2373b65059 794) die_errno(_("cannot unlink stray '%s'"), path);
2373b65059 1043) error(_("failed to update conflicted state in '%s'"),
path);
2373b65059 1061) error(_("no remembered resolution for '%s'"), path);
2373b65059 1063) error_errno(_("cannot unlink '%s'"), filename);
2373b65059 1097) return error(_("index file corrupt"));
2373b65059 1185) die_errno(_("unable to open rr-cache directory"));
revision.c
4a7e27e957 3241) oideq(&p->item->object.oid, &commit->object.oid))
sha1-file.c
67947c34ae sha1-file.c 2216) if (!hasheq(expected_sha1, real_sha1)) {
sha1-name.c
8aac67a174 sha1-name.c 154) static void unique_in_midx(struct
multi_pack_index *m,
8aac67a174 sha1-name.c 157) uint32_t num, i, first = 0;
8aac67a174 sha1-name.c 158) const struct object_id *current = NULL;
8aac67a174 sha1-name.c 159) num = m->num_objects;
8aac67a174 sha1-name.c 161) if (!num)
8aac67a174 sha1-name.c 162) return;
8aac67a174 sha1-name.c 164) bsearch_midx(&ds->bin_pfx, m, &first);
8aac67a174 sha1-name.c 171) for (i = first; i < num && !ds->ambiguous;
i++) {
8aac67a174 sha1-name.c 173) current = nth_midxed_object_oid(&oid, m, i);
8aac67a174 sha1-name.c 174) if (!match_sha(ds->len, ds->bin_pfx.hash,
current->hash))
8aac67a174 sha1-name.c 175) break;
8aac67a174 sha1-name.c 176) update_candidates(ds, current);
8aac67a174 sha1-name.c 212) m = m->next)
8aac67a174 sha1-name.c 213) unique_in_midx(m, ds);
8aac67a174 sha1-name.c 573) return;
trace.c
c46c406ae1 189) now = getnanotime();
c46c406ae1 190) perf_start_times[perf_indent] = now;
c46c406ae1 191) if (perf_indent + 1 < ARRAY_SIZE(perf_start_times))
c46c406ae1 192) perf_indent++;
c46c406ae1 194) BUG("Too deep indentation");
c46c406ae1 195) return now;
c46c406ae1 211) if (perf_indent >= strlen(space))
c46c406ae1 212) BUG("Too deep indentation");
c46c406ae1 214) strbuf_addf(&buf, ":%.*s ", perf_indent, space);
c46c406ae1 317) void trace_performance_leave_fl(const char *file, int line,
c46c406ae1 323) if (perf_indent)
c46c406ae1 324) perf_indent--;
c46c406ae1 326) if (!format) /* Allow callers to leave without tracing
anything */
c46c406ae1 327) return;
c46c406ae1 329) since = perf_start_times[perf_indent];
c46c406ae1 330) va_start(ap, format);
c46c406ae1 331) trace_performance_vprintf_fl(file, line, nanos - since,
format, ap);
c46c406ae1 332) va_end(ap);
c46c406ae1 477) trace_performance_leave("git command:%s", command_line.buf);
c46c406ae1 485) if (!command_line.len)
c46c406ae1 490) trace_performance_enter();
unpack-trees.c
b878579ae7 360) string_list_append(&list, ce->name);
b878579ae7 361) ce->ce_flags &= ~CE_MATCHED;
b878579ae7 368) warning(_("the following paths have collided (e.g.
case-sensitive paths\n"
b878579ae7 372) for (i = 0; i < list.nr; i++)
b878579ae7 373) fprintf(stderr, " '%s'\n", list.items[i].string);
b4da37380b 715) BUG("This is a directory and should not exist in index");
b4da37380b 719) BUG("pos must point at the first entry in this directory");
b4da37380b 740) BUG("We need cache-tree to do this optimization");
f1e11c6510 777) free(tree_ce);
b4da37380b 778) return rc;
b4da37380b 785) printf("Unpacked %d entries from %s to %s using
cache-tree\n",
b4da37380b 787) o->src_index->cache[pos]->name,
b4da37380b 788) o->src_index->cache[pos + nr_entries - 1]->name);
b4da37380b 811) BUG("Wrong condition to get here buddy");
Commits introducing uncovered code:
Ben Peart fa655d841: checkout: optimize "git checkout -b <new_branch>"
Christian Couder 108f53038: pack-objects: move tree_depth into
'struct packing_data'
Christian Couder fe0ac2fb7: pack-objects: move 'layer' into 'struct
packing_data'
Derrick Stolee 0d5b3a5ef: midx: write object ids in a chunk
Derrick Stolee 17c35c896: packfile: skip loading index if in
multi-pack-index
Derrick Stolee 1d614d41e: commit-reach: move ref_newer from remote.c
Derrick Stolee 32f3c541e: multi-pack-index: write pack names in chunk
Derrick Stolee 3715a6335: midx: read objects from multi-pack-index
Derrick Stolee 396f25701: multi-pack-index: read packfile list
Derrick Stolee 454ea2e4d: treewide: use get_all_packs
Derrick Stolee 4d80560c5: multi-pack-index: load into memory
Derrick Stolee 5227c3856: commit-reach: move walk methods from commit.c
Derrick Stolee 525e18c04: midx: clear midx on repack
Derrick Stolee 56ee7ff15: multi-pack-index: add 'verify' verb
Derrick Stolee 662148c43: midx: write object offsets
Derrick Stolee 66ec0390e: fsck: verify multi-pack-index
Derrick Stolee 6a22d5212: pack-objects: consider packs in
multi-pack-index
Derrick Stolee 6cc017431: commit-reach: use can_all_from_reach
Derrick Stolee 6d68e6a46: multi-pack-index: provide more helpful
usage info
Derrick Stolee 859fdc0c3: commit-graph: define GIT_TEST_COMMIT_GRAPH
Derrick Stolee 8aac67a17: midx: use midx in abbreviation calculations
Derrick Stolee 920f93ca1: commit-reach: move commit_contains from
ref-filter
Derrick Stolee a40498a12: midx: use existing midx when writing new one
Derrick Stolee b67f6b26e: commit-reach: properly peel tags
Derrick Stolee c39b02ae0: midx: mark bad packed objects
Derrick Stolee c4d25228e: config: create core.multiPackIndex setting
Derrick Stolee cc6af73c0: multi-pack-index: verify object offsets
Derrick Stolee fc59e7484: midx: write header information to lockfile
Derrick Stolee fe1ed56f5: midx: sort and deduplicate objects from
packfiles
Duy Nguyen b878579ae: clone: report duplicate entries on
case-insensitive filesystems
Eric Sunshine 2e6fd71a5: format-patch: extend --range-diff to
accept revision range
Eric Sunshine 40ce41604: format-patch: allow --range-diff to apply
to a lone-patch
Eric Sunshine 68a6b3a1b: worktree: teach 'move' to override lock
when --force given twice
Eric Sunshine 8631bf1cd: format-patch: add --creation-factor tweak
for --range-diff
Eric Sunshine e19831c94: worktree: teach 'add' to respect --force
for registered but missing path
Eric Sunshine e5353bef5: worktree: move delete_git_dir() earlier in
file for upcoming new callers
Eric Sunshine ee6cbf712: format-patch: allow --interdiff to apply
to a lone-patch
Eric Sunshine f4143101c: worktree: teach 'remove' to override lock
when --force given twice
Jeff King 16d75fa48: repack: add delta-islands support
Jeff King 28b8a7308: pack-objects: add delta-islands support
Jeff King 30cdc33fb: pack-bitmap: save "have" bitmap from walk
Jeff King 4a7e27e95: convert "oidcmp() == 0" to oideq()
Jeff King 67947c34a: convert "hashcmp() != 0" to "!hasheq()"
Jeff King 9001dc2a7: convert "oidcmp() != 0" to "!oideq()"
Jeff King c8d521faf: Add delta-islands.{c,h}
Nguyễn Thái Ngọc Duy 4592e6080: cache-tree: verify valid cache-tree
in the test suite
Nguyễn Thái Ngọc Duy b4da37380: unpack-trees: optimize walking same
trees with cache-tree
Nguyễn Thái Ngọc Duy c46c406ae: trace.h: support nested performance
tracing
Nguyễn Thái Ngọc Duy f1e11c651: unpack-trees: reduce malloc in
cache-tree walk
René Scharfe 3aa4d81f8: mailinfo: support format=flowed
René Scharfe fb8952077: fsck: use strbuf_getline() to read skiplist
file
Shulhan 5025425df: builtin/remote: quote remote name on error to
display empty name
Stefan Beller 74d4731da: submodule--helper: replace
connect-gitdir-workingtree by ensure-core-worktree
Stefan Beller ee69b2a90: submodule--helper: introduce new
update-module-mode helper
Thomas Gummerer 2373b6505: rerere: mark strings for translation
Thomas Gummerer 4af32207b: rerere: teach rerere to handle nested
conflicts
Thomas Gummerer c0f16f8e1: rerere: factor out handle_conflict function
Torsten Bögershausen d64324cb6: Make git_check_attr() a void function
[1] https://dev.azure.com/git/git/_build/results?buildId=172&view=logs
Build running coverage-test for 'jch' and coverage-diff.sh against
'next'
[2] https://git.visualstudio.com/git/_build/results?buildId=166&view=logs
Build running coverage-test for 'next' and coverage-diff.sh against
'master'
[3] https://dev.azure.com/git/git/_build/results?buildId=171&view=logs
Build running coverage-test for 'master' and coverage-diff.sh
against 'maint'
^ permalink raw reply [relevance 2%]
* Re: [RFC PATCH 00/19] Bring more repository handles into our code base
2018-10-11 21:17 6% [RFC PATCH 00/19] Bring more repository handles into our code base Stefan Beller
` (2 preceding siblings ...)
2018-10-11 21:17 2% ` [PATCH 19/19] Apply semantic patches from previous patches Stefan Beller
@ 2018-10-11 23:31 2% ` Junio C Hamano
3 siblings, 0 replies; 200+ results
From: Junio C Hamano @ 2018-10-11 23:31 UTC (permalink / raw)
To: Stefan Beller; +Cc: git, jonathantanmy
Stefan Beller <sbeller@google.com> writes:
> Additionally each patch adds a semantic patch, that would port from the old to
> the new function. These semantic patches are all applied in the very last patch,
> but we could omit applying the last patch if it causes too many merge conflicts
> and trickl in the semantic patches over time when there are no merge conflicts.
That's an interesting approach ;-)
> The original goal of all these refactoring series was to remove add_submodule_odb
> in submodule.c, which was partially reached with this series.
Yup, that is a very good goalpost to keep in mind.
> remaining calls in another series, but it shows we're close to be done with these
> large refactorings as far as I am concerned.
Nice.
^ permalink raw reply [relevance 2%]
* Re: [PATCH 18/19] submodule: don't add submodule as odb for push
@ 2018-10-11 23:09 8% ` Stefan Beller
0 siblings, 0 replies; 200+ results
From: Stefan Beller @ 2018-10-11 23:09 UTC (permalink / raw)
To: Jonathan Tan; +Cc: git
> Do you know if pushing of submodules is exercised by any test?
t5531-deep-submodule-push.sh (all of them)
t5545-push-options.sh (ok 4 - push options and submodules)
^ permalink raw reply [relevance 8%]
* [PATCH 19/19] Apply semantic patches from previous patches
2018-10-11 21:17 6% [RFC PATCH 00/19] Bring more repository handles into our code base Stefan Beller
2018-10-11 21:17 18% ` [PATCH 17/19] submodule: use submodule repos for object lookup Stefan Beller
2018-10-11 21:17 14% ` [PATCH 18/19] submodule: don't add submodule as odb for push Stefan Beller
@ 2018-10-11 21:17 2% ` Stefan Beller
2018-10-11 23:31 2% ` [RFC PATCH 00/19] Bring more repository handles into our code base Junio C Hamano
3 siblings, 0 replies; 200+ results
From: Stefan Beller @ 2018-10-11 21:17 UTC (permalink / raw)
To: git; +Cc: jonathantanmy, Stefan Beller
Previous commits added some cocci rules, but did not patch the whole tree,
as to not dilute the focus for reviewing the previous patches.
This patch is generated by 'make coccicheck' and applying the resulting
diff, which was white space damaged (>8 spaces after a tab) in blame.c,
which has been fixed.
Signed-off-by: Stefan Beller <sbeller@google.com>
---
apply.c | 6 ++--
archive.c | 5 +--
bisect.c | 5 +--
blame.c | 15 +++++----
builtin/am.c | 2 +-
builtin/blame.c | 4 +--
builtin/cat-file.c | 21 +++++++-----
builtin/checkout.c | 4 +--
builtin/commit.c | 13 +++++---
builtin/describe.c | 4 +--
builtin/difftool.c | 3 +-
builtin/fast-export.c | 7 ++--
builtin/fmt-merge-msg.c | 8 +++--
builtin/grep.c | 2 +-
builtin/index-pack.c | 8 +++--
builtin/log.c | 4 +--
builtin/merge-base.c | 2 +-
builtin/merge-tree.c | 9 ++++--
builtin/mktag.c | 3 +-
builtin/name-rev.c | 2 +-
builtin/notes.c | 12 ++++---
builtin/pack-objects.c | 22 +++++++++----
builtin/reflog.c | 5 +--
builtin/replace.c | 2 +-
builtin/shortlog.c | 5 +--
builtin/show-branch.c | 4 +--
builtin/tag.c | 4 +--
builtin/unpack-file.c | 2 +-
builtin/unpack-objects.c | 3 +-
builtin/verify-commit.c | 2 +-
bundle.c | 2 +-
combine-diff.c | 2 +-
commit-graph.c | 8 ++---
commit.c | 15 +++++----
config.c | 2 +-
diff.c | 3 +-
dir.c | 2 +-
entry.c | 3 +-
fast-import.c | 7 ++--
fsck.c | 9 +++---
grep.c | 3 +-
http-push.c | 3 +-
log-tree.c | 3 +-
mailmap.c | 2 +-
match-trees.c | 4 +--
merge-blobs.c | 6 ++--
merge-recursive.c | 13 ++++----
negotiator/default.c | 6 ++--
negotiator/skipping.c | 2 +-
notes-cache.c | 5 +--
notes-merge.c | 4 +--
notes-utils.c | 2 +-
notes.c | 10 +++---
pretty.c | 5 +--
read-cache.c | 5 +--
remote-testsvn.c | 4 +--
remote.c | 2 +-
rerere.c | 5 +--
revision.c | 12 +++----
sequencer.c | 55 +++++++++++++++++++-------------
sha1-file.c | 3 +-
sha1-name.c | 9 +++---
shallow.c | 4 +--
submodule-config.c | 3 +-
t/helper/test-revision-walking.c | 3 +-
tag.c | 5 +--
tree-walk.c | 6 ++--
tree.c | 5 +--
walker.c | 2 +-
xdiff-interface.c | 2 +-
70 files changed, 254 insertions(+), 180 deletions(-)
diff --git a/apply.c b/apply.c
index fdae1d423b..5ac284b7e8 100644
--- a/apply.c
+++ b/apply.c
@@ -3187,7 +3187,8 @@ static int apply_binary(struct apply_state *state,
unsigned long size;
char *result;
- result = read_object_file(&oid, &type, &size);
+ result = repo_read_object_file(the_repository, &oid, &type,
+ &size);
if (!result)
return error(_("the necessary postimage %s for "
"'%s' cannot be read"),
@@ -3249,7 +3250,8 @@ static int read_blob_object(struct strbuf *buf, const struct object_id *oid, uns
unsigned long sz;
char *result;
- result = read_object_file(oid, &type, &sz);
+ result = repo_read_object_file(the_repository, oid, &type,
+ &sz);
if (!result)
return -1;
/* XXX read_sha1_file NUL-terminates */
diff --git a/archive.c b/archive.c
index 994495af05..70e5eed535 100644
--- a/archive.c
+++ b/archive.c
@@ -55,7 +55,8 @@ static void format_subst(const struct commit *commit,
strbuf_add(&fmt, b + 8, c - b - 8);
strbuf_add(buf, src, b - src);
- format_commit_message(commit, fmt.buf, buf, &ctx);
+ repo_format_commit_message(the_repository, commit, fmt.buf,
+ buf, &ctx);
len -= c + 1 - src;
src = c + 1;
}
@@ -73,7 +74,7 @@ void *object_file_to_archive(const struct archiver_args *args,
const struct commit *commit = args->convert ? args->commit : NULL;
path += args->baselen;
- buffer = read_object_file(oid, type, sizep);
+ buffer = repo_read_object_file(the_repository, oid, type, sizep);
if (buffer && S_ISREG(mode)) {
struct strbuf buf = STRBUF_INIT;
size_t size = 0;
diff --git a/bisect.c b/bisect.c
index 6ae5e5b49e..ae92367738 100644
--- a/bisect.c
+++ b/bisect.c
@@ -136,8 +136,9 @@ static void show_list(const char *debug, int counted, int nr,
unsigned flags = commit->object.flags;
enum object_type type;
unsigned long size;
- char *buf = read_object_file(&commit->object.oid, &type,
- &size);
+ char *buf = repo_read_object_file(the_repository,
+ &commit->object.oid, &type,
+ &size);
const char *subject_start;
int subject_len;
diff --git a/blame.c b/blame.c
index c229a10c0e..fc2764d036 100644
--- a/blame.c
+++ b/blame.c
@@ -322,8 +322,9 @@ static void fill_origin_blob(struct diff_options *opt,
&o->blob_oid, 1, &file->ptr, &file_size))
;
else
- file->ptr = read_object_file(&o->blob_oid, &type,
- &file_size);
+ file->ptr = repo_read_object_file(the_repository,
+ &o->blob_oid, &type,
+ &file_size);
file->size = file_size;
if (!file->ptr)
@@ -1455,7 +1456,7 @@ static void pass_blame(struct blame_scoreboard *sb, struct blame_origin *origin,
if (sg_origin[i])
continue;
- if (parse_commit(p))
+ if (repo_parse_commit(the_repository, p))
continue;
porigin = find(sb->repo, p, origin);
if (!porigin)
@@ -1594,7 +1595,7 @@ void assign_blame(struct blame_scoreboard *sb, int opt)
* so hold onto it in the meantime.
*/
blame_origin_incref(suspect);
- parse_commit(commit);
+ repo_parse_commit(the_repository, commit);
if (sb->reverse ||
(!(commit->object.flags & UNINTERESTING) &&
!(revs->max_age != -1 && commit->date < revs->max_age)))
@@ -1864,8 +1865,10 @@ void setup_scoreboard(struct blame_scoreboard *sb,
&sb->final_buf_size))
;
else
- sb->final_buf = read_object_file(&o->blob_oid, &type,
- &sb->final_buf_size);
+ sb->final_buf = repo_read_object_file(the_repository,
+ &o->blob_oid,
+ &type,
+ &sb->final_buf_size);
if (!sb->final_buf)
die(_("cannot read blob %s for path %s"),
diff --git a/builtin/am.c b/builtin/am.c
index 601570dbef..3c992a93fb 100644
--- a/builtin/am.c
+++ b/builtin/am.c
@@ -1360,7 +1360,7 @@ static void get_commit_info(struct am_state *state, struct commit *commit)
die(_("unable to parse commit %s"), oid_to_hex(&commit->object.oid));
state->msg = xstrdup(msg + 2);
state->msg_len = strlen(state->msg);
- unuse_commit_buffer(commit, buffer);
+ repo_unuse_commit_buffer(the_repository, commit, buffer);
}
/**
diff --git a/builtin/blame.c b/builtin/blame.c
index a443af9ee9..dcd300c0d3 100644
--- a/builtin/blame.c
+++ b/builtin/blame.c
@@ -197,7 +197,7 @@ static void get_commit_info(struct commit *commit,
&ret->author_time, &ret->author_tz);
if (!detailed) {
- unuse_commit_buffer(commit, message);
+ repo_unuse_commit_buffer(the_repository, commit, message);
return;
}
@@ -211,7 +211,7 @@ static void get_commit_info(struct commit *commit,
else
strbuf_addf(&ret->summary, "(%s)", oid_to_hex(&commit->object.oid));
- unuse_commit_buffer(commit, message);
+ repo_unuse_commit_buffer(the_repository, commit, message);
}
/*
diff --git a/builtin/cat-file.c b/builtin/cat-file.c
index 8d97c84725..afad7b1c84 100644
--- a/builtin/cat-file.c
+++ b/builtin/cat-file.c
@@ -34,7 +34,7 @@ static int filter_object(const char *path, unsigned mode,
{
enum object_type type;
- *buf = read_object_file(oid, &type, size);
+ *buf = repo_read_object_file(the_repository, oid, &type, size);
if (!*buf)
return error(_("cannot read object %s '%s'"),
oid_to_hex(oid), path);
@@ -133,7 +133,8 @@ static int cat_one_file(int opt, const char *exp_type, const char *obj_name,
if (type == OBJ_BLOB)
return stream_blob_to_fd(1, &oid, NULL, 0);
- buf = read_object_file(&oid, &type, &size);
+ buf = repo_read_object_file(the_repository, &oid, &type,
+ &size);
if (!buf)
die("Cannot read object %s", obj_name);
@@ -144,8 +145,10 @@ static int cat_one_file(int opt, const char *exp_type, const char *obj_name,
if (type_from_string(exp_type) == OBJ_BLOB) {
struct object_id blob_oid;
if (oid_object_info(the_repository, &oid, NULL) == OBJ_TAG) {
- char *buffer = read_object_file(&oid, &type,
- &size);
+ char *buffer = repo_read_object_file(the_repository,
+ &oid,
+ &type,
+ &size);
const char *target;
if (!skip_prefix(buffer, "object ", &target) ||
get_oid_hex(target, &blob_oid))
@@ -309,9 +312,10 @@ static void print_object_or_die(struct batch_options *opt, struct expand_data *d
if (!textconv_object(the_repository,
data->rest, 0100644, oid,
1, &contents, &size))
- contents = read_object_file(oid,
- &type,
- &size);
+ contents = repo_read_object_file(the_repository,
+ oid,
+ &type,
+ &size);
if (!contents)
die("could not convert '%s' %s",
oid_to_hex(oid), data->rest);
@@ -327,7 +331,8 @@ static void print_object_or_die(struct batch_options *opt, struct expand_data *d
unsigned long size;
void *contents;
- contents = read_object_file(oid, &type, &size);
+ contents = repo_read_object_file(the_repository, oid, &type,
+ &size);
if (!contents)
die("object %s disappeared", oid_to_hex(oid));
if (type != data->type)
diff --git a/builtin/checkout.c b/builtin/checkout.c
index ae28478ff8..0be05e0665 100644
--- a/builtin/checkout.c
+++ b/builtin/checkout.c
@@ -404,7 +404,7 @@ static void describe_detached_head(const char *msg, struct commit *commit)
{
struct strbuf sb = STRBUF_INIT;
- if (!parse_commit(commit))
+ if (!repo_parse_commit(the_repository, commit))
pp_commit_easy(CMIT_FMT_ONELINE, commit, &sb);
if (print_sha1_ellipsis()) {
fprintf(stderr, "%s %s... %s\n", msg,
@@ -732,7 +732,7 @@ static void describe_one_orphan(struct strbuf *sb, struct commit *commit)
strbuf_addstr(sb, " ");
strbuf_add_unique_abbrev(sb, &commit->object.oid, DEFAULT_ABBREV);
strbuf_addch(sb, ' ');
- if (!parse_commit(commit))
+ if (!repo_parse_commit(the_repository, commit))
pp_commit_easy(CMIT_FMT_ONELINE, commit, sb);
strbuf_addch(sb, '\n');
}
diff --git a/builtin/commit.c b/builtin/commit.c
index 9d8ce6cb3b..4c07dc5681 100644
--- a/builtin/commit.c
+++ b/builtin/commit.c
@@ -680,8 +680,9 @@ static int prepare_to_commit(const char *index_file, const char *prefix,
if (!c)
die(_("could not lookup commit %s"), squash_message);
ctx.output_encoding = get_commit_output_encoding();
- format_commit_message(c, "squash! %s\n\n", &sb,
- &ctx);
+ repo_format_commit_message(the_repository, c,
+ "squash! %s\n\n", &sb,
+ &ctx);
}
}
@@ -713,8 +714,9 @@ static int prepare_to_commit(const char *index_file, const char *prefix,
if (!commit)
die(_("could not lookup commit %s"), fixup_message);
ctx.output_encoding = get_commit_output_encoding();
- format_commit_message(commit, "fixup! %s\n\n",
- &sb, &ctx);
+ repo_format_commit_message(the_repository, commit,
+ "fixup! %s\n\n",
+ &sb, &ctx);
if (have_option_m)
strbuf_addbuf(&sb, &message);
hook_arg1 = "message";
@@ -997,7 +999,8 @@ static const char *find_author_by_nickname(const char *name)
struct pretty_print_context ctx = {0};
ctx.date_mode.type = DATE_NORMAL;
strbuf_release(&buf);
- format_commit_message(commit, "%aN <%aE>", &buf, &ctx);
+ repo_format_commit_message(the_repository, commit,
+ "%aN <%aE>", &buf, &ctx);
clear_mailmap(&mailmap);
return strbuf_detach(&buf, NULL);
}
diff --git a/builtin/describe.c b/builtin/describe.c
index 1fde0563fe..290218c0d9 100644
--- a/builtin/describe.c
+++ b/builtin/describe.c
@@ -254,7 +254,7 @@ static unsigned long finish_depth_computation(
best->depth++;
while (parents) {
struct commit *p = parents->item;
- parse_commit(p);
+ repo_parse_commit(the_repository, p);
if (!(p->object.flags & SEEN))
commit_list_insert_by_date(p, list);
p->object.flags |= c->object.flags;
@@ -381,7 +381,7 @@ static void describe_commit(struct object_id *oid, struct strbuf *dst)
}
while (parents) {
struct commit *p = parents->item;
- parse_commit(p);
+ repo_parse_commit(the_repository, p);
if (!(p->object.flags & SEEN))
commit_list_insert_by_date(p, &list);
p->object.flags |= c->object.flags;
diff --git a/builtin/difftool.c b/builtin/difftool.c
index e7023e3adf..23720d4d64 100644
--- a/builtin/difftool.c
+++ b/builtin/difftool.c
@@ -307,7 +307,8 @@ static char *get_symlink(const struct object_id *oid, const char *path)
} else {
enum object_type type;
unsigned long size;
- data = read_object_file(oid, &type, &size);
+ data = repo_read_object_file(the_repository, oid, &type,
+ &size);
if (!data)
die(_("could not read object %s for symlink %s"),
oid_to_hex(oid), path);
diff --git a/builtin/fast-export.c b/builtin/fast-export.c
index 9bd4a95a47..9e65697524 100644
--- a/builtin/fast-export.c
+++ b/builtin/fast-export.c
@@ -239,7 +239,7 @@ static void export_blob(const struct object_id *oid)
object = (struct object *)lookup_blob(the_repository, oid);
eaten = 0;
} else {
- buf = read_object_file(oid, &type, &size);
+ buf = repo_read_object_file(the_repository, oid, &type, &size);
if (!buf)
die("could not read blob %s", oid_to_hex(oid));
if (check_object_signature(oid, buf, size, type_name(type)) < 0)
@@ -617,7 +617,7 @@ static void handle_commit(struct commit *commit, struct rev_info *rev,
? strlen(message) : 0),
reencoded ? reencoded : message ? message : "");
free(reencoded);
- unuse_commit_buffer(commit, commit_buffer);
+ repo_unuse_commit_buffer(the_repository, commit, commit_buffer);
for (i = 0, p = commit->parents; p; p = p->next) {
int mark = get_object_mark(&p->item->object);
@@ -689,7 +689,8 @@ static void handle_tag(const char *name, struct tag *tag)
return;
}
- buf = read_object_file(&tag->object.oid, &type, &size);
+ buf = repo_read_object_file(the_repository, &tag->object.oid, &type,
+ &size);
if (!buf)
die("could not read tag %s", oid_to_hex(&tag->object.oid));
message = memmem(buf, size, "\n\n", 2);
diff --git a/builtin/fmt-merge-msg.c b/builtin/fmt-merge-msg.c
index 1adc84ed87..4248e3982d 100644
--- a/builtin/fmt-merge-msg.c
+++ b/builtin/fmt-merge-msg.c
@@ -264,7 +264,7 @@ static void record_person(int which, struct string_list *people,
{
const char *buffer = get_commit_buffer(commit, NULL);
record_person_from_buf(which, people, buffer);
- unuse_commit_buffer(commit, buffer);
+ repo_unuse_commit_buffer(the_repository, commit, buffer);
}
static int cmp_string_list_util_as_integral(const void *a_, const void *b_)
@@ -375,7 +375,8 @@ static void shortlog(const char *name,
if (subjects.nr > limit)
continue;
- format_commit_message(commit, "%s", &sb, &ctx);
+ repo_format_commit_message(the_repository, commit, "%s", &sb,
+ &ctx);
strbuf_ltrim(&sb);
if (!sb.len)
@@ -493,7 +494,8 @@ static void fmt_merge_msg_sigs(struct strbuf *out)
struct object_id *oid = origins.items[i].util;
enum object_type type;
unsigned long size, len;
- char *buf = read_object_file(oid, &type, &size);
+ char *buf = repo_read_object_file(the_repository, oid, &type,
+ &size);
struct strbuf sig = STRBUF_INIT;
if (!buf || type != OBJ_TAG)
diff --git a/builtin/grep.c b/builtin/grep.c
index 0c3527242e..06f7711330 100644
--- a/builtin/grep.c
+++ b/builtin/grep.c
@@ -308,7 +308,7 @@ static void *lock_and_read_oid_file(const struct object_id *oid, enum object_typ
void *data;
grep_read_lock();
- data = read_object_file(oid, type, size);
+ data = repo_read_object_file(the_repository, oid, type, size);
grep_read_unlock();
return data;
}
diff --git a/builtin/index-pack.c b/builtin/index-pack.c
index 9582ead950..887c44e111 100644
--- a/builtin/index-pack.c
+++ b/builtin/index-pack.c
@@ -817,7 +817,8 @@ static void sha1_object(const void *data, struct object_entry *obj_entry,
die(_("cannot read existing object info %s"), oid_to_hex(oid));
if (has_type != type || has_size != size)
die(_("SHA1 COLLISION FOUND WITH %s !"), oid_to_hex(oid));
- has_data = read_object_file(oid, &has_type, &has_size);
+ has_data = repo_read_object_file(the_repository, oid,
+ &has_type, &has_size);
read_unlock();
if (!data)
data = new_data = get_data_from_pack(obj_entry);
@@ -1379,8 +1380,9 @@ static void fix_unresolved_deltas(struct hashfile *f)
if (objects[d->obj_no].real_type != OBJ_REF_DELTA)
continue;
- base_obj->data = read_object_file(&d->oid, &type,
- &base_obj->size);
+ base_obj->data = repo_read_object_file(the_repository,
+ &d->oid, &type,
+ &base_obj->size);
if (!base_obj->data)
continue;
diff --git a/builtin/log.c b/builtin/log.c
index 717d20e115..d4709ec63d 100644
--- a/builtin/log.c
+++ b/builtin/log.c
@@ -525,7 +525,7 @@ static int show_tag_object(const struct object_id *oid, struct rev_info *rev)
{
unsigned long size;
enum object_type type;
- char *buf = read_object_file(oid, &type, &size);
+ char *buf = repo_read_object_file(the_repository, oid, &type, &size);
int offset = 0;
if (!buf)
@@ -1032,7 +1032,7 @@ static void make_cover_letter(struct rev_info *rev, int use_stdout,
const char *buf = get_commit_buffer(list[i], NULL);
if (has_non_ascii(buf))
need_8bit_cte = 1;
- unuse_commit_buffer(list[i], buf);
+ repo_unuse_commit_buffer(the_repository, list[i], buf);
}
if (!branch_name)
diff --git a/builtin/merge-base.c b/builtin/merge-base.c
index 08d91b1f0c..82567bf79c 100644
--- a/builtin/merge-base.c
+++ b/builtin/merge-base.c
@@ -127,7 +127,7 @@ static void add_one_commit(struct object_id *oid, struct rev_collect *revs)
commit = lookup_commit(the_repository, oid);
if (!commit ||
(commit->object.flags & TMP_MARK) ||
- parse_commit(commit))
+ repo_parse_commit(the_repository, commit))
return;
ALLOC_GROW(revs->commit, revs->nr + 1, revs->alloc);
diff --git a/builtin/merge-tree.c b/builtin/merge-tree.c
index f32941fdab..99403f70f4 100644
--- a/builtin/merge-tree.c
+++ b/builtin/merge-tree.c
@@ -62,7 +62,9 @@ static void *result(struct merge_list *entry, unsigned long *size)
const char *path = entry->path;
if (!entry->stage)
- return read_object_file(&entry->blob->object.oid, &type, size);
+ return repo_read_object_file(the_repository,
+ &entry->blob->object.oid, &type,
+ size);
base = NULL;
if (entry->stage == 1) {
base = entry->blob;
@@ -84,8 +86,9 @@ static void *origin(struct merge_list *entry, unsigned long *size)
enum object_type type;
while (entry) {
if (entry->stage == 2)
- return read_object_file(&entry->blob->object.oid,
- &type, size);
+ return repo_read_object_file(the_repository,
+ &entry->blob->object.oid,
+ &type, size);
entry = entry->link;
}
return NULL;
diff --git a/builtin/mktag.c b/builtin/mktag.c
index 6fb7dc8578..be4366b03f 100644
--- a/builtin/mktag.c
+++ b/builtin/mktag.c
@@ -25,7 +25,8 @@ static int verify_object(const struct object_id *oid, const char *expected_type)
int ret = -1;
enum object_type type;
unsigned long size;
- void *buffer = read_object_file(oid, &type, &size);
+ void *buffer = repo_read_object_file(the_repository, oid, &type,
+ &size);
const struct object_id *repl = lookup_replace_object(the_repository, oid);
if (buffer) {
diff --git a/builtin/name-rev.c b/builtin/name-rev.c
index f1cb45c227..41f2de37e4 100644
--- a/builtin/name-rev.c
+++ b/builtin/name-rev.c
@@ -87,7 +87,7 @@ static void name_rev(struct commit *commit,
int parent_number = 1;
char *to_free = NULL;
- parse_commit(commit);
+ repo_parse_commit(the_repository, commit);
if (commit->date < cutoff)
return;
diff --git a/builtin/notes.c b/builtin/notes.c
index c05cd004ab..7ce3e50b45 100644
--- a/builtin/notes.c
+++ b/builtin/notes.c
@@ -124,7 +124,7 @@ static void copy_obj_to_fd(int fd, const struct object_id *oid)
{
unsigned long size;
enum object_type type;
- char *buf = read_object_file(oid, &type, &size);
+ char *buf = repo_read_object_file(the_repository, oid, &type, &size);
if (buf) {
if (size)
write_or_die(fd, buf, size);
@@ -255,7 +255,7 @@ static int parse_reuse_arg(const struct option *opt, const char *arg, int unset)
if (get_oid(arg, &object))
die(_("failed to resolve '%s' as a valid ref."), arg);
- if (!(buf = read_object_file(&object, &type, &len))) {
+ if (!(buf = repo_read_object_file(the_repository, &object, &type, &len))) {
free(buf);
die(_("failed to read object '%s'."), arg);
}
@@ -610,7 +610,8 @@ static int append_edit(int argc, const char **argv, const char *prefix)
/* Append buf to previous note contents */
unsigned long size;
enum object_type type;
- char *prev_buf = read_object_file(note, &type, &size);
+ char *prev_buf = repo_read_object_file(the_repository, note,
+ &type, &size);
strbuf_grow(&d.buf, size + 1);
if (d.buf.len && prev_buf && size)
@@ -714,7 +715,7 @@ static int merge_commit(struct notes_merge_options *o)
die(_("failed to read ref NOTES_MERGE_PARTIAL"));
else if (!(partial = lookup_commit_reference(the_repository, &oid)))
die(_("could not find commit from NOTES_MERGE_PARTIAL."));
- else if (parse_commit(partial))
+ else if (repo_parse_commit(the_repository, partial))
die(_("could not parse commit from NOTES_MERGE_PARTIAL."));
if (partial->parents)
@@ -735,7 +736,8 @@ static int merge_commit(struct notes_merge_options *o)
/* Reuse existing commit message in reflog message */
memset(&pretty_ctx, 0, sizeof(pretty_ctx));
- format_commit_message(partial, "%s", &msg, &pretty_ctx);
+ repo_format_commit_message(the_repository, partial, "%s", &msg,
+ &pretty_ctx);
strbuf_trim(&msg);
strbuf_insert(&msg, 0, "notes: ", 7);
update_ref(msg.buf, o->local_ref, &oid,
diff --git a/builtin/pack-objects.c b/builtin/pack-objects.c
index 3383ba43d0..2d55b64433 100644
--- a/builtin/pack-objects.c
+++ b/builtin/pack-objects.c
@@ -138,11 +138,13 @@ static void *get_delta(struct object_entry *entry)
void *buf, *base_buf, *delta_buf;
enum object_type type;
- buf = read_object_file(&entry->idx.oid, &type, &size);
+ buf = repo_read_object_file(the_repository, &entry->idx.oid, &type,
+ &size);
if (!buf)
die(_("unable to read %s"), oid_to_hex(&entry->idx.oid));
- base_buf = read_object_file(&DELTA(entry)->idx.oid, &type,
- &base_size);
+ base_buf = repo_read_object_file(the_repository,
+ &DELTA(entry)->idx.oid, &type,
+ &base_size);
if (!base_buf)
die("unable to read %s",
oid_to_hex(&DELTA(entry)->idx.oid));
@@ -292,7 +294,9 @@ static unsigned long write_no_reuse_object(struct hashfile *f, struct object_ent
(st = open_istream(&entry->idx.oid, &type, &size, NULL)) != NULL)
buf = NULL;
else {
- buf = read_object_file(&entry->idx.oid, &type, &size);
+ buf = repo_read_object_file(the_repository,
+ &entry->idx.oid, &type,
+ &size);
if (!buf)
die(_("unable to read %s"),
oid_to_hex(&entry->idx.oid));
@@ -1218,7 +1222,7 @@ static struct pbase_tree_cache *pbase_tree_get(const struct object_id *oid)
/* Did not find one. Either we got a bogus request or
* we need to read and perhaps cache.
*/
- data = read_object_file(oid, &type, &size);
+ data = repo_read_object_file(the_repository, oid, &type, &size);
if (!data)
return NULL;
if (type != OBJ_TREE) {
@@ -1989,7 +1993,9 @@ static int try_delta(struct unpacked *trg, struct unpacked *src,
/* Load data if not already done */
if (!trg->data) {
read_lock();
- trg->data = read_object_file(&trg_entry->idx.oid, &type, &sz);
+ trg->data = repo_read_object_file(the_repository,
+ &trg_entry->idx.oid, &type,
+ &sz);
read_unlock();
if (!trg->data)
die(_("object %s cannot be read"),
@@ -2002,7 +2008,9 @@ static int try_delta(struct unpacked *trg, struct unpacked *src,
}
if (!src->data) {
read_lock();
- src->data = read_object_file(&src_entry->idx.oid, &type, &sz);
+ src->data = repo_read_object_file(the_repository,
+ &src_entry->idx.oid, &type,
+ &sz);
read_unlock();
if (!src->data) {
if (src_entry->preferred_base) {
diff --git a/builtin/reflog.c b/builtin/reflog.c
index b5941c1ff3..53a7f0c8e5 100644
--- a/builtin/reflog.c
+++ b/builtin/reflog.c
@@ -77,7 +77,8 @@ static int tree_is_complete(const struct object_id *oid)
if (!tree->buffer) {
enum object_type type;
unsigned long size;
- void *data = read_object_file(oid, &type, &size);
+ void *data = repo_read_object_file(the_repository, oid, &type,
+ &size);
if (!data) {
tree->object.flags |= INCOMPLETE;
return 0;
@@ -235,7 +236,7 @@ static void mark_reachable(struct expire_reflog_policy_cb *cb)
struct commit *commit = pop_commit(&pending);
if (commit->object.flags & REACHABLE)
continue;
- if (parse_commit(commit))
+ if (repo_parse_commit(the_repository, commit))
continue;
commit->object.flags |= REACHABLE;
if (commit->date < expire_limit) {
diff --git a/builtin/replace.c b/builtin/replace.c
index e0b16ad44b..48b44dba3e 100644
--- a/builtin/replace.c
+++ b/builtin/replace.c
@@ -449,7 +449,7 @@ static int create_graft(int argc, const char **argv, int force, int gentle)
buffer = get_commit_buffer(commit, &size);
strbuf_add(&buf, buffer, size);
- unuse_commit_buffer(commit, buffer);
+ repo_unuse_commit_buffer(the_repository, commit, buffer);
if (replace_parents(&buf, argc - 1, &argv[1]) < 0) {
strbuf_release(&buf);
diff --git a/builtin/shortlog.c b/builtin/shortlog.c
index 88f88e97b2..d46df5af34 100644
--- a/builtin/shortlog.c
+++ b/builtin/shortlog.c
@@ -170,12 +170,13 @@ void shortlog_add_commit(struct shortlog *log, struct commit *commit)
(log->email ? "%cN <%cE>" : "%cN") :
(log->email ? "%aN <%aE>" : "%aN");
- format_commit_message(commit, fmt, &author, &ctx);
+ repo_format_commit_message(the_repository, commit, fmt, &author, &ctx);
if (!log->summary) {
if (log->user_format)
pretty_print_commit(&ctx, commit, &oneline);
else
- format_commit_message(commit, "%s", &oneline, &ctx);
+ repo_format_commit_message(the_repository, commit,
+ "%s", &oneline, &ctx);
}
insert_one_record(log, author.buf, oneline.len ? oneline.buf : "<none>");
diff --git a/builtin/show-branch.c b/builtin/show-branch.c
index 363cf8509a..d7e8d6f15c 100644
--- a/builtin/show-branch.c
+++ b/builtin/show-branch.c
@@ -238,7 +238,7 @@ static void join_revs(struct commit_list **list_p,
parents = parents->next;
if ((this_flag & flags) == flags)
continue;
- parse_commit(p);
+ repo_parse_commit(the_repository, p);
if (mark_seen(p, seen_p) && !still_interesting)
extra--;
p->object.flags |= flags;
@@ -835,7 +835,7 @@ int cmd_show_branch(int ac, const char **av, const char *prefix)
if (!commit)
die(_("cannot find commit %s (%s)"),
ref_name[num_rev], oid_to_hex(&revkey));
- parse_commit(commit);
+ repo_parse_commit(the_repository, commit);
mark_seen(commit, &seen);
/* rev#0 uses bit REV_SHIFT, rev#1 uses bit REV_SHIFT+1,
diff --git a/builtin/tag.c b/builtin/tag.c
index 9a19ffb49f..3aa1ea7f15 100644
--- a/builtin/tag.c
+++ b/builtin/tag.c
@@ -169,7 +169,7 @@ static void write_tag_body(int fd, const struct object_id *oid)
enum object_type type;
char *buf, *sp;
- buf = read_object_file(oid, &type, &size);
+ buf = repo_read_object_file(the_repository, oid, &type, &size);
if (!buf)
return;
/* skip header */
@@ -305,7 +305,7 @@ static void create_reflog_msg(const struct object_id *oid, struct strbuf *sb)
strbuf_addstr(sb, "object of unknown type");
break;
case OBJ_COMMIT:
- if ((buf = read_object_file(oid, &type, &size)) != NULL) {
+ if ((buf = repo_read_object_file(the_repository, oid, &type, &size)) != NULL) {
subject_len = find_commit_subject(buf, &subject_start);
strbuf_insert(sb, sb->len, subject_start, subject_len);
} else {
diff --git a/builtin/unpack-file.c b/builtin/unpack-file.c
index 58652229f2..df4a47bae7 100644
--- a/builtin/unpack-file.c
+++ b/builtin/unpack-file.c
@@ -10,7 +10,7 @@ static char *create_temp_file(struct object_id *oid)
unsigned long size;
int fd;
- buf = read_object_file(oid, &type, &size);
+ buf = repo_read_object_file(the_repository, oid, &type, &size);
if (!buf || type != OBJ_BLOB)
die("unable to read blob object %s", oid_to_hex(oid));
diff --git a/builtin/unpack-objects.c b/builtin/unpack-objects.c
index 30d9413b4b..1e2bc1be3a 100644
--- a/builtin/unpack-objects.c
+++ b/builtin/unpack-objects.c
@@ -425,7 +425,8 @@ static void unpack_delta_entry(enum object_type type, unsigned long delta_size,
if (resolve_against_held(nr, &base_oid, delta_data, delta_size))
return;
- base = read_object_file(&base_oid, &type, &base_size);
+ base = repo_read_object_file(the_repository, &base_oid, &type,
+ &base_size);
if (!base) {
error("failed to read delta-pack base object %s",
oid_to_hex(&base_oid));
diff --git a/builtin/verify-commit.c b/builtin/verify-commit.c
index 7772c07ed7..3caa6cf8e9 100644
--- a/builtin/verify-commit.c
+++ b/builtin/verify-commit.c
@@ -47,7 +47,7 @@ static int verify_commit(const char *name, unsigned flags)
if (get_oid(name, &oid))
return error("commit '%s' not found.", name);
- buf = read_object_file(&oid, &type, &size);
+ buf = repo_read_object_file(the_repository, &oid, &type, &size);
if (!buf)
return error("%s: unable to read file.", name);
if (type != OBJ_COMMIT)
diff --git a/bundle.c b/bundle.c
index a5a71d059e..8080cf1ed1 100644
--- a/bundle.c
+++ b/bundle.c
@@ -224,7 +224,7 @@ static int is_tag_in_date_range(struct object *tag, struct rev_info *revs)
if (revs->max_age == -1 && revs->min_age == -1)
goto out;
- buf = read_object_file(&tag->oid, &type, &size);
+ buf = repo_read_object_file(the_repository, &tag->oid, &type, &size);
if (!buf)
goto out;
line = memmem(buf, size, "\ntagger ", 8);
diff --git a/combine-diff.c b/combine-diff.c
index 41ab5b01de..5593052ae9 100644
--- a/combine-diff.c
+++ b/combine-diff.c
@@ -308,7 +308,7 @@ static char *grab_blob(struct repository *r,
*size = fill_textconv(r, textconv, df, &blob);
free_filespec(df);
} else {
- blob = read_object_file(oid, &type, size);
+ blob = repo_read_object_file(the_repository, oid, &type, size);
if (type != OBJ_BLOB)
die("object '%s' is not a blob!", oid_to_hex(oid));
}
diff --git a/commit-graph.c b/commit-graph.c
index 8a1bec7b8a..fd2f5380e1 100644
--- a/commit-graph.c
+++ b/commit-graph.c
@@ -435,7 +435,7 @@ static void write_graph_chunk_data(struct hashfile *f, int hash_len,
int edge_value;
uint32_t packedDate[2];
- parse_commit(*list);
+ repo_parse_commit(the_repository, *list);
hashwrite(f, get_commit_tree_oid(*list)->hash, hash_len);
parent = (*list)->parents;
@@ -606,7 +606,7 @@ static void close_reachable(struct packed_oid_list *oids)
for (i = 0; i < oids->nr; i++) {
commit = lookup_commit(the_repository, &oids->list[i]);
- if (commit && !parse_commit(commit))
+ if (commit && !repo_parse_commit(the_repository, commit))
add_missing_parents(oids, commit);
}
@@ -783,7 +783,7 @@ void write_commit_graph(const char *obj_dir,
continue;
commits.list[commits.nr] = lookup_commit(the_repository, &oids.list[i]);
- parse_commit(commits.list[commits.nr]);
+ repo_parse_commit(the_repository, commits.list[commits.nr]);
for (parent = commits.list[commits.nr]->parents;
parent; parent = parent->next)
@@ -954,7 +954,7 @@ int verify_commit_graph(struct repository *r, struct commit_graph *g)
graph_commit = lookup_commit(r, &cur_oid);
odb_commit = (struct commit *)create_object(r, cur_oid.hash, alloc_commit_node(r));
- if (parse_commit_internal(odb_commit, 0, 0)) {
+ if (repo_parse_commit_internal(the_repository, odb_commit, 0, 0)) {
graph_report("failed to parse %s from object database",
oid_to_hex(&cur_oid));
continue;
diff --git a/commit.c b/commit.c
index 526b33758d..e86241b802 100644
--- a/commit.c
+++ b/commit.c
@@ -70,7 +70,7 @@ struct commit *lookup_commit_reference_by_name(const char *name)
if (get_oid_committish(name, &oid))
return NULL;
commit = lookup_commit_reference(the_repository, &oid);
- if (parse_commit(commit))
+ if (repo_parse_commit(the_repository, commit))
return NULL;
return commit;
}
@@ -491,7 +491,7 @@ int repo_parse_commit_gently(struct repository *r,
void parse_commit_or_die(struct commit *item)
{
- if (parse_commit(item))
+ if (repo_parse_commit(the_repository, item))
die("unable to parse commit %s",
item ? oid_to_hex(&item->object.oid) : "(null)");
}
@@ -596,7 +596,7 @@ struct commit *pop_most_recent_commit(struct commit_list **list,
while (parents) {
struct commit *commit = parents->item;
- if (!parse_commit(commit) && !(commit->object.flags & mark)) {
+ if (!repo_parse_commit(the_repository, commit) && !(commit->object.flags & mark)) {
commit->object.flags |= mark;
commit_list_insert_by_date(commit, list);
}
@@ -689,7 +689,7 @@ static void record_author_date(struct author_date_slab *author_date,
*(author_date_slab_at(author_date, commit)) = date;
fail_exit:
- unuse_commit_buffer(commit, buffer);
+ repo_unuse_commit_buffer(the_repository, commit, buffer);
}
static int compare_commits_by_author_date(const void *a_, const void *b_,
@@ -1302,7 +1302,7 @@ int parse_signed_commit(const struct commit *commit,
}
line = next;
}
- unuse_commit_buffer(commit, buffer);
+ repo_unuse_commit_buffer(the_repository, commit, buffer);
return saw_signature;
}
@@ -1351,7 +1351,8 @@ static void handle_signed_tag(struct commit *parent, struct commit_extra_header
desc = merge_remote_util(parent);
if (!desc || !desc->obj)
return;
- buf = read_object_file(&desc->obj->oid, &type, &size);
+ buf = repo_read_object_file(the_repository, &desc->obj->oid, &type,
+ &size);
if (!buf || type != OBJ_TAG)
goto free_return;
len = parse_signature(buf, size);
@@ -1429,7 +1430,7 @@ struct commit_extra_header *read_commit_extra_headers(struct commit *commit,
unsigned long size;
const char *buffer = get_commit_buffer(commit, &size);
extra = read_commit_extra_header_lines(buffer, size, exclude);
- unuse_commit_buffer(commit, buffer);
+ repo_unuse_commit_buffer(the_repository, commit, buffer);
return extra;
}
diff --git a/config.c b/config.c
index 3461993f0a..e2b72403a6 100644
--- a/config.c
+++ b/config.c
@@ -1605,7 +1605,7 @@ int git_config_from_blob_oid(config_fn_t fn,
unsigned long size;
int ret;
- buf = read_object_file(oid, &type, &size);
+ buf = repo_read_object_file(the_repository, oid, &type, &size);
if (!buf)
return error(_("unable to load config blob object '%s'"), name);
if (type != OBJ_BLOB) {
diff --git a/diff.c b/diff.c
index c5b5e7ac41..ab16d6dcad 100644
--- a/diff.c
+++ b/diff.c
@@ -3939,7 +3939,8 @@ int diff_populate_filespec(struct repository *r,
return 0;
}
}
- s->data = read_object_file(&s->oid, &type, &s->size);
+ s->data = repo_read_object_file(the_repository, &s->oid,
+ &type, &s->size);
if (!s->data)
die("unable to read %s", oid_to_hex(&s->oid));
s->should_free = 1;
diff --git a/dir.c b/dir.c
index aceb0d4869..8722b1e5d5 100644
--- a/dir.c
+++ b/dir.c
@@ -245,7 +245,7 @@ static int do_read_blob(const struct object_id *oid, struct oid_stat *oid_stat,
*size_out = 0;
*data_out = NULL;
- data = read_object_file(oid, &type, &sz);
+ data = repo_read_object_file(the_repository, oid, &type, &sz);
if (!data || type != OBJ_BLOB) {
free(data);
return -1;
diff --git a/entry.c b/entry.c
index 2a2ab6c839..45723fc958 100644
--- a/entry.c
+++ b/entry.c
@@ -86,7 +86,8 @@ static int create_file(const char *path, unsigned int mode)
static void *read_blob_entry(const struct cache_entry *ce, unsigned long *size)
{
enum object_type type;
- void *blob_data = read_object_file(&ce->oid, &type, size);
+ void *blob_data = repo_read_object_file(the_repository, &ce->oid,
+ &type, size);
if (blob_data) {
if (type == OBJ_BLOB)
diff --git a/fast-import.c b/fast-import.c
index 89bb0c9db3..879b1fbf56 100644
--- a/fast-import.c
+++ b/fast-import.c
@@ -1371,7 +1371,7 @@ static void load_tree(struct tree_entry *root)
die("Can't load tree %s", oid_to_hex(oid));
} else {
enum object_type type;
- buf = read_object_file(oid, &type, &size);
+ buf = repo_read_object_file(the_repository, oid, &type, &size);
if (!buf || type != OBJ_TREE)
die("Can't load tree %s", oid_to_hex(oid));
}
@@ -2931,7 +2931,7 @@ static void cat_blob(struct object_entry *oe, struct object_id *oid)
char *buf;
if (!oe || oe->pack_id == MAX_PACK_ID) {
- buf = read_object_file(oid, &type, &size);
+ buf = repo_read_object_file(the_repository, oid, &type, &size);
} else {
type = oe->type;
buf = gfi_unpack_entry(oe, &size);
@@ -3037,7 +3037,8 @@ static struct object_entry *dereference(struct object_entry *oe,
buf = gfi_unpack_entry(oe, &size);
} else {
enum object_type unused;
- buf = read_object_file(oid, &unused, &size);
+ buf = repo_read_object_file(the_repository, oid, &unused,
+ &size);
}
if (!buf)
die("Can't load object %s", oid_to_hex(oid));
diff --git a/fsck.c b/fsck.c
index a0cee0be59..0d3e8886e2 100644
--- a/fsck.c
+++ b/fsck.c
@@ -447,7 +447,7 @@ static int fsck_walk_commit(struct commit *commit, void *data, struct fsck_optio
int result;
const char *name;
- if (parse_commit(commit))
+ if (repo_parse_commit(the_repository, commit))
return -1;
name = get_object_name(options, &commit->object);
@@ -859,7 +859,7 @@ static int fsck_commit(struct commit *commit, const char *data,
const char *buffer = data ? data : get_commit_buffer(commit, &size);
int ret = fsck_commit_buffer(commit, buffer, size, options);
if (!data)
- unuse_commit_buffer(commit, buffer);
+ repo_unuse_commit_buffer(the_repository, commit, buffer);
return ret;
}
@@ -879,7 +879,8 @@ static int fsck_tag_buffer(struct tag *tag, const char *data,
enum object_type type;
buffer = to_free =
- read_object_file(&tag->object.oid, &type, &size);
+ repo_read_object_file(the_repository,
+ &tag->object.oid, &type, &size);
if (!buffer)
return report(options, &tag->object,
FSCK_MSG_MISSING_TAG_OBJECT,
@@ -1091,7 +1092,7 @@ int fsck_finish(struct fsck_options *options)
continue;
}
- buf = read_object_file(oid, &type, &size);
+ buf = repo_read_object_file(the_repository, oid, &type, &size);
if (!buf) {
if (is_promisor_object(&blob->object.oid))
continue;
diff --git a/grep.c b/grep.c
index f6bd89e40b..216c38a0d5 100644
--- a/grep.c
+++ b/grep.c
@@ -2112,7 +2112,8 @@ static int grep_source_load_oid(struct grep_source *gs)
enum object_type type;
grep_read_lock();
- gs->buf = read_object_file(gs->identifier, &type, &gs->size);
+ gs->buf = repo_read_object_file(the_repository, gs->identifier, &type,
+ &gs->size);
grep_read_unlock();
if (!gs->buf)
diff --git a/http-push.c b/http-push.c
index df504ab6a3..c6206838ed 100644
--- a/http-push.c
+++ b/http-push.c
@@ -364,7 +364,8 @@ static void start_put(struct transfer_request *request)
ssize_t size;
git_zstream stream;
- unpacked = read_object_file(&request->obj->oid, &type, &len);
+ unpacked = repo_read_object_file(the_repository, &request->obj->oid,
+ &type, &len);
hdrlen = xsnprintf(hdr, sizeof(hdr), "%s %lu", type_name(type), len) + 1;
/* Set it up */
diff --git a/log-tree.c b/log-tree.c
index 7443e5fcc7..360dd0b4e6 100644
--- a/log-tree.c
+++ b/log-tree.c
@@ -332,7 +332,8 @@ void fmt_output_commit(struct strbuf *filename,
struct pretty_print_context ctx = {0};
struct strbuf subject = STRBUF_INIT;
- format_commit_message(commit, "%f", &subject, &ctx);
+ repo_format_commit_message(the_repository, commit, "%f", &subject,
+ &ctx);
fmt_output_subject(filename, subject.buf, info);
strbuf_release(&subject);
}
diff --git a/mailmap.c b/mailmap.c
index 962fd86d6d..f3ffea0e17 100644
--- a/mailmap.c
+++ b/mailmap.c
@@ -225,7 +225,7 @@ static int read_mailmap_blob(struct string_list *map,
if (get_oid(name, &oid) < 0)
return 0;
- buf = read_object_file(&oid, &type, &size);
+ buf = repo_read_object_file(the_repository, &oid, &type, &size);
if (!buf)
return error("unable to read mailmap object at %s", name);
if (type != OBJ_BLOB)
diff --git a/match-trees.c b/match-trees.c
index 37653308d3..3fb649ad08 100644
--- a/match-trees.c
+++ b/match-trees.c
@@ -55,7 +55,7 @@ static void *fill_tree_desc_strict(struct tree_desc *desc,
enum object_type type;
unsigned long size;
- buffer = read_object_file(hash, &type, &size);
+ buffer = repo_read_object_file(the_repository, hash, &type, &size);
if (!buffer)
die("unable to read tree (%s)", oid_to_hex(hash));
if (type != OBJ_TREE)
@@ -190,7 +190,7 @@ static int splice_tree(const struct object_id *oid1, const char *prefix,
if (*subpath)
subpath++;
- buf = read_object_file(oid1, &type, &sz);
+ buf = repo_read_object_file(the_repository, oid1, &type, &sz);
if (!buf)
die("cannot read tree %s", oid_to_hex(oid1));
init_tree_desc(&desc, buf, sz);
diff --git a/merge-blobs.c b/merge-blobs.c
index ee0a0e90c9..757e338715 100644
--- a/merge-blobs.c
+++ b/merge-blobs.c
@@ -12,7 +12,8 @@ static int fill_mmfile_blob(mmfile_t *f, struct blob *obj)
unsigned long size;
enum object_type type;
- buf = read_object_file(&obj->object.oid, &type, &size);
+ buf = repo_read_object_file(the_repository, &obj->object.oid, &type,
+ &size);
if (!buf)
return -1;
if (type != OBJ_BLOB) {
@@ -75,7 +76,8 @@ void *merge_blobs(struct index_state *istate, const char *path,
return NULL;
if (!our)
our = their;
- return read_object_file(&our->object.oid, &type, size);
+ return repo_read_object_file(the_repository, &our->object.oid,
+ &type, size);
}
if (fill_mmfile_blob(&f1, our) < 0)
diff --git a/merge-recursive.c b/merge-recursive.c
index 82e9f1d24a..646e265625 100644
--- a/merge-recursive.c
+++ b/merge-recursive.c
@@ -299,7 +299,7 @@ static void output_commit_title(struct merge_options *o, struct commit *commit)
strbuf_add_unique_abbrev(&o->obuf, &commit->object.oid,
DEFAULT_ABBREV);
strbuf_addch(&o->obuf, ' ');
- if (parse_commit(commit) != 0)
+ if (repo_parse_commit(the_repository, commit) != 0)
strbuf_addstr(&o->obuf, _("(bad commit)\n"));
else {
const char *title;
@@ -307,7 +307,7 @@ static void output_commit_title(struct merge_options *o, struct commit *commit)
int len = find_commit_subject(msg, &title);
if (len)
strbuf_addf(&o->obuf, "%.*s\n", len, title);
- unuse_commit_buffer(commit, msg);
+ repo_unuse_commit_buffer(the_repository, commit, msg);
}
}
flush_output(o);
@@ -957,7 +957,7 @@ static int update_file_flags(struct merge_options *o,
goto update_index;
}
- buf = read_object_file(oid, &type, &size);
+ buf = repo_read_object_file(the_repository, oid, &type, &size);
if (!buf)
return err(o, _("cannot read object %s '%s'"), oid_to_hex(oid), path);
if (type != OBJ_BLOB) {
@@ -1160,7 +1160,8 @@ static void print_commit(struct commit *commit)
struct strbuf sb = STRBUF_INIT;
struct pretty_print_context ctx = {0};
ctx.date_mode.type = DATE_NORMAL;
- format_commit_message(commit, " %h: %m %s", &sb, &ctx);
+ repo_format_commit_message(the_repository, commit, " %h: %m %s", &sb,
+ &ctx);
fprintf(stderr, "%s\n", sb.buf);
strbuf_release(&sb);
}
@@ -2944,7 +2945,7 @@ static int read_oid_strbuf(struct merge_options *o,
void *buf;
enum object_type type;
unsigned long size;
- buf = read_object_file(oid, &type, &size);
+ buf = repo_read_object_file(the_repository, oid, &type, &size);
if (!buf)
return err(o, _("cannot read object %s"), oid_to_hex(oid));
if (type != OBJ_BLOB) {
@@ -3522,7 +3523,7 @@ static struct commit *get_ref(const struct object_id *oid, const char *name)
return make_virtual_commit((struct tree*)object, name);
if (object->type != OBJ_COMMIT)
return NULL;
- if (parse_commit((struct commit *)object))
+ if (repo_parse_commit(the_repository, (struct commit *)object))
return NULL;
return (struct commit *)object;
}
diff --git a/negotiator/default.c b/negotiator/default.c
index 4b78f6bf36..19071a97df 100644
--- a/negotiator/default.c
+++ b/negotiator/default.c
@@ -25,7 +25,7 @@ static void rev_list_push(struct negotiation_state *ns,
if (!(commit->object.flags & mark)) {
commit->object.flags |= mark;
- if (parse_commit(commit))
+ if (repo_parse_commit(the_repository, commit))
return;
prio_queue_put(&ns->rev_list, commit);
@@ -68,7 +68,7 @@ static void mark_common(struct negotiation_state *ns, struct commit *commit,
if (!ancestors_only && !(o->flags & POPPED))
ns->non_common_revs--;
if (!o->parsed && !dont_parse)
- if (parse_commit(commit))
+ if (repo_parse_commit(the_repository, commit))
return;
for (parents = commit->parents;
@@ -95,7 +95,7 @@ static const struct object_id *get_rev(struct negotiation_state *ns)
return NULL;
commit = prio_queue_get(&ns->rev_list);
- parse_commit(commit);
+ repo_parse_commit(the_repository, commit);
parents = commit->parents;
commit->object.flags |= POPPED;
diff --git a/negotiator/skipping.c b/negotiator/skipping.c
index dffbc76c49..9b76f95654 100644
--- a/negotiator/skipping.c
+++ b/negotiator/skipping.c
@@ -177,7 +177,7 @@ static const struct object_id *get_rev(struct data *data)
if (!(commit->object.flags & COMMON) && !entry->ttl)
to_send = commit;
- parse_commit(commit);
+ repo_parse_commit(the_repository, commit);
for (p = commit->parents; p; p = p->next)
parent_pushed |= push_parent(data, entry, p->item);
diff --git a/notes-cache.c b/notes-cache.c
index d87e7ca91c..94196ef0f6 100644
--- a/notes-cache.c
+++ b/notes-cache.c
@@ -21,7 +21,8 @@ static int notes_cache_match_validity(const char *ref, const char *validity)
return 0;
memset(&pretty_ctx, 0, sizeof(pretty_ctx));
- format_commit_message(commit, "%s", &msg, &pretty_ctx);
+ repo_format_commit_message(the_repository, commit, "%s", &msg,
+ &pretty_ctx);
strbuf_trim(&msg);
ret = !strcmp(msg.buf, validity);
@@ -79,7 +80,7 @@ char *notes_cache_get(struct notes_cache *c, struct object_id *key_oid,
value_oid = get_note(&c->tree, key_oid);
if (!value_oid)
return NULL;
- value = read_object_file(value_oid, &type, &size);
+ value = repo_read_object_file(the_repository, value_oid, &type, &size);
*outsize = size;
return value;
diff --git a/notes-merge.c b/notes-merge.c
index 13dd9ba158..af1f5306dd 100644
--- a/notes-merge.c
+++ b/notes-merge.c
@@ -324,7 +324,7 @@ static void write_note_to_worktree(const struct object_id *obj,
{
enum object_type type;
unsigned long size;
- void *buf = read_object_file(note, &type, &size);
+ void *buf = repo_read_object_file(the_repository, note, &type, &size);
if (!buf)
die("cannot read note %s for object %s",
@@ -723,7 +723,7 @@ int notes_merge_commit(struct notes_merge_options *o,
create_notes_commit(partial_tree, partial_commit->parents, msg,
strlen(msg), result_oid);
- unuse_commit_buffer(partial_commit, buffer);
+ repo_unuse_commit_buffer(the_repository, partial_commit, buffer);
if (o->verbosity >= 4)
printf("Finalized notes merge commit: %s\n",
oid_to_hex(result_oid));
diff --git a/notes-utils.c b/notes-utils.c
index 14ea03178e..f25ef54a77 100644
--- a/notes-utils.c
+++ b/notes-utils.c
@@ -22,7 +22,7 @@ void create_notes_commit(struct notes_tree *t, struct commit_list *parents,
if (!read_ref(t->ref, &parent_oid)) {
struct commit *parent = lookup_commit(the_repository,
&parent_oid);
- if (parse_commit(parent))
+ if (repo_parse_commit(the_repository, parent))
die("Failed to find/parse commit %s", t->ref);
commit_list_insert(parent, &parents);
}
diff --git a/notes.c b/notes.c
index 32d3dbcc1e..766d540dd1 100644
--- a/notes.c
+++ b/notes.c
@@ -797,13 +797,15 @@ int combine_notes_concatenate(struct object_id *cur_oid,
/* read in both note blob objects */
if (!is_null_oid(new_oid))
- new_msg = read_object_file(new_oid, &new_type, &new_len);
+ new_msg = repo_read_object_file(the_repository, new_oid,
+ &new_type, &new_len);
if (!new_msg || !new_len || new_type != OBJ_BLOB) {
free(new_msg);
return 0;
}
if (!is_null_oid(cur_oid))
- cur_msg = read_object_file(cur_oid, &cur_type, &cur_len);
+ cur_msg = repo_read_object_file(the_repository, cur_oid,
+ &cur_type, &cur_len);
if (!cur_msg || !cur_len || cur_type != OBJ_BLOB) {
free(cur_msg);
free(new_msg);
@@ -859,7 +861,7 @@ static int string_list_add_note_lines(struct string_list *list,
return 0;
/* read_sha1_file NUL-terminates */
- data = read_object_file(oid, &t, &len);
+ data = repo_read_object_file(the_repository, oid, &t, &len);
if (t != OBJ_BLOB || !data || !len) {
free(data);
return t != OBJ_BLOB || !data;
@@ -1218,7 +1220,7 @@ static void format_note(struct notes_tree *t, const struct object_id *object_oid
if (!oid)
return;
- if (!(msg = read_object_file(oid, &type, &msglen)) || type != OBJ_BLOB) {
+ if (!(msg = repo_read_object_file(the_repository, oid, &type, &msglen)) || type != OBJ_BLOB) {
free(msg);
return;
}
diff --git a/pretty.c b/pretty.c
index 948f5346cf..4ab5b44da0 100644
--- a/pretty.c
+++ b/pretty.c
@@ -1799,7 +1799,8 @@ void pretty_print_commit(struct pretty_print_context *pp,
int need_8bit_cte = pp->need_8bit_cte;
if (pp->fmt == CMIT_FMT_USERFORMAT) {
- format_commit_message(commit, user_format, sb, pp);
+ repo_format_commit_message(the_repository, commit,
+ user_format, sb, pp);
return;
}
@@ -1861,7 +1862,7 @@ void pretty_print_commit(struct pretty_print_context *pp,
if (cmit_fmt_is_mail(pp->fmt) && sb->len <= beginning_of_body)
strbuf_addch(sb, '\n');
- unuse_commit_buffer(commit, reencoded);
+ repo_unuse_commit_buffer(the_repository, commit, reencoded);
}
void pp_commit_easy(enum cmit_fmt fmt, const struct commit *commit,
diff --git a/read-cache.c b/read-cache.c
index b707edd044..e0669b174e 100644
--- a/read-cache.c
+++ b/read-cache.c
@@ -232,7 +232,7 @@ static int ce_compare_link(const struct cache_entry *ce, size_t expected_size)
if (strbuf_readlink(&sb, ce->name, expected_size))
return -1;
- buffer = read_object_file(&ce->oid, &type, &size);
+ buffer = repo_read_object_file(the_repository, &ce->oid, &type, &size);
if (buffer) {
if (size == sb.len)
match = memcmp(buffer, sb.buf, size);
@@ -2901,7 +2901,8 @@ void *read_blob_data_from_index(const struct index_state *istate,
}
if (pos < 0)
return NULL;
- data = read_object_file(&istate->cache[pos]->oid, &type, &sz);
+ data = repo_read_object_file(the_repository, &istate->cache[pos]->oid,
+ &type, &sz);
if (!data || type != OBJ_BLOB) {
free(data);
return NULL;
diff --git a/remote-testsvn.c b/remote-testsvn.c
index 3af708c5b6..c846c9ee3b 100644
--- a/remote-testsvn.c
+++ b/remote-testsvn.c
@@ -62,7 +62,7 @@ static char *read_ref_note(const struct object_id *oid)
init_notes(NULL, notes_ref, NULL, 0);
if (!(note_oid = get_note(NULL, oid)))
return NULL; /* note tree not found */
- if (!(msg = read_object_file(note_oid, &type, &msglen)))
+ if (!(msg = repo_read_object_file(the_repository, note_oid, &type, &msglen)))
error("Empty notes tree. %s", notes_ref);
else if (!msglen || type != OBJ_BLOB) {
error("Note contains unusable content. "
@@ -109,7 +109,7 @@ static int note2mark_cb(const struct object_id *object_oid,
enum object_type type;
struct rev_note note;
- if (!(msg = read_object_file(note_oid, &type, &msglen)) ||
+ if (!(msg = repo_read_object_file(the_repository, note_oid, &type, &msglen)) ||
!msglen || type != OBJ_BLOB) {
free(msg);
return 1;
diff --git a/remote.c b/remote.c
index 348417f0a7..e16e8e002f 100644
--- a/remote.c
+++ b/remote.c
@@ -1822,7 +1822,7 @@ int ref_newer(const struct object_id *new_oid, const struct object_id *old_oid)
return 0;
new_commit = (struct commit *) o;
- if (parse_commit(new_commit) < 0)
+ if (repo_parse_commit(the_repository, new_commit) < 0)
return 0;
used = list = NULL;
diff --git a/rerere.c b/rerere.c
index 8d4ac8426b..084bedfe55 100644
--- a/rerere.c
+++ b/rerere.c
@@ -988,8 +988,9 @@ static int handle_cache(struct index_state *istate, const char *path,
break;
i = ce_stage(ce) - 1;
if (!mmfile[i].ptr) {
- mmfile[i].ptr = read_object_file(&ce->oid, &type,
- &size);
+ mmfile[i].ptr = repo_read_object_file(the_repository,
+ &ce->oid, &type,
+ &size);
mmfile[i].size = size;
}
}
diff --git a/revision.c b/revision.c
index 28366eaccf..01c11107b8 100644
--- a/revision.c
+++ b/revision.c
@@ -274,7 +274,7 @@ static struct commit *handle_commit(struct rev_info *revs,
if (object->type == OBJ_COMMIT) {
struct commit *commit = (struct commit *)object;
- if (parse_commit(commit) < 0)
+ if (repo_parse_commit(the_repository, commit) < 0)
die("unable to parse commit %s", name);
if (flags & UNINTERESTING) {
mark_parents_uninteresting(commit);
@@ -688,7 +688,7 @@ static void try_to_simplify_commit(struct rev_info *revs, struct commit *commit)
ts->treesame[0] = 1;
}
}
- if (parse_commit(p) < 0)
+ if (repo_parse_commit(the_repository, p) < 0)
die("cannot simplify commit %s (because of %s)",
oid_to_hex(&commit->object.oid),
oid_to_hex(&p->object.oid));
@@ -721,7 +721,7 @@ static void try_to_simplify_commit(struct rev_info *revs, struct commit *commit)
* IOW, we pretend this parent is a
* "root" commit.
*/
- if (parse_commit(p) < 0)
+ if (repo_parse_commit(the_repository, p) < 0)
die("cannot simplify commit %s (invalid %s)",
oid_to_hex(&commit->object.oid),
oid_to_hex(&p->object.oid));
@@ -800,7 +800,7 @@ static int add_parents_to_list(struct rev_info *revs, struct commit *commit,
parent = parent->next;
if (p)
p->object.flags |= UNINTERESTING;
- if (parse_commit_gently(p, 1) < 0)
+ if (repo_parse_commit_gently(the_repository, p, 1) < 0)
continue;
if (p->parents)
mark_parents_uninteresting(p);
@@ -828,7 +828,7 @@ static int add_parents_to_list(struct rev_info *revs, struct commit *commit,
struct commit *p = parent->item;
int gently = revs->ignore_missing_links ||
revs->exclude_promisor_objects;
- if (parse_commit_gently(p, gently) < 0) {
+ if (repo_parse_commit_gently(the_repository, p, gently) < 0) {
if (revs->exclude_promisor_objects &&
is_promisor_object(&p->object.oid)) {
if (revs->first_parent_only)
@@ -3095,7 +3095,7 @@ static int commit_match(struct commit *commit, struct rev_info *opt)
retval = grep_buffer(&opt->grep_filter,
(char *)message, strlen(message));
strbuf_release(&buf);
- unuse_commit_buffer(commit, message);
+ repo_unuse_commit_buffer(the_repository, commit, message);
return opt->invert_grep ? !retval : retval;
}
diff --git a/sequencer.c b/sequencer.c
index da4e727cc3..5d612831e9 100644
--- a/sequencer.c
+++ b/sequencer.c
@@ -344,7 +344,7 @@ static void free_message(struct commit *commit, struct commit_message *msg)
free(msg->parent_label);
free(msg->label);
free(msg->subject);
- unuse_commit_buffer(commit, msg->message);
+ repo_unuse_commit_buffer(the_repository, commit, msg->message);
}
static void print_advice(int show_hint, struct replay_opts *opts)
@@ -604,7 +604,7 @@ static int is_index_unchanged(void)
* the commit is invalid, parse_commit() will complain. So
* there is nothing for us to say here. Just return failure.
*/
- if (parse_commit(head_commit))
+ if (repo_parse_commit(the_repository, head_commit))
return -1;
if (!(cache_tree_oid = get_cache_tree_oid()))
@@ -1142,13 +1142,15 @@ void print_commit_summary(const char *prefix, const struct object_id *oid,
commit = lookup_commit(the_repository, oid);
if (!commit)
die(_("couldn't look up newly created commit"));
- if (parse_commit(commit))
+ if (repo_parse_commit(the_repository, commit))
die(_("could not parse newly created commit"));
strbuf_addstr(&format, "format:%h] %s");
- format_commit_message(commit, "%an <%ae>", &author_ident, &pctx);
- format_commit_message(commit, "%cn <%ce>", &committer_ident, &pctx);
+ repo_format_commit_message(the_repository, commit, "%an <%ae>",
+ &author_ident, &pctx);
+ repo_format_commit_message(the_repository, commit, "%cn <%ce>",
+ &committer_ident, &pctx);
if (strbuf_cmp(&author_ident, &committer_ident)) {
strbuf_addstr(&format, "\n Author: ");
strbuf_addbuf_percentquote(&format, &author_ident);
@@ -1156,7 +1158,8 @@ void print_commit_summary(const char *prefix, const struct object_id *oid,
if (flags & SUMMARY_SHOW_AUTHOR_DATE) {
struct strbuf date = STRBUF_INIT;
- format_commit_message(commit, "%ad", &date, &pctx);
+ repo_format_commit_message(the_repository, commit, "%ad",
+ &date, &pctx);
strbuf_addstr(&format, "\n Date: ");
strbuf_addbuf_percentquote(&format, &date);
strbuf_release(&date);
@@ -1221,7 +1224,7 @@ static int parse_head(struct commit **head)
warning(_("HEAD %s is not a commit!"),
oid_to_hex(&oid));
}
- if (parse_commit(current_head))
+ if (repo_parse_commit(the_repository, current_head))
return error(_("could not parse HEAD commit"));
}
*head = current_head;
@@ -1271,7 +1274,8 @@ static int try_to_commit(struct strbuf *msg, const char *author,
hook_commit = "HEAD";
}
author = amend_author = get_author(message);
- unuse_commit_buffer(current_head, message);
+ repo_unuse_commit_buffer(the_repository, current_head,
+ message);
if (!author) {
res = error(_("unable to parse commit author"));
goto out;
@@ -1381,12 +1385,12 @@ static int is_original_commit_empty(struct commit *commit)
{
const struct object_id *ptree_oid;
- if (parse_commit(commit))
+ if (repo_parse_commit(the_repository, commit))
return error(_("could not parse commit %s"),
oid_to_hex(&commit->object.oid));
if (commit->parents) {
struct commit *parent = commit->parents->item;
- if (parse_commit(parent))
+ if (repo_parse_commit(the_repository, parent))
return error(_("could not parse parent commit %s"),
oid_to_hex(&parent->object.oid));
ptree_oid = get_commit_tree_oid(parent);
@@ -1557,7 +1561,8 @@ static int update_squash_messages(enum todo_command command,
find_commit_subject(head_message, &body);
if (write_message(body, strlen(body),
rebase_path_fixup_msg(), 0)) {
- unuse_commit_buffer(head_commit, head_message);
+ repo_unuse_commit_buffer(the_repository, head_commit,
+ head_message);
return error(_("cannot write '%s'"),
rebase_path_fixup_msg());
}
@@ -1569,7 +1574,8 @@ static int update_squash_messages(enum todo_command command,
strbuf_addstr(&buf, "\n\n");
strbuf_addstr(&buf, body);
- unuse_commit_buffer(head_commit, head_message);
+ repo_unuse_commit_buffer(the_repository, head_commit,
+ head_message);
}
if (!(message = get_commit_buffer(commit, NULL)))
@@ -1592,7 +1598,7 @@ static int update_squash_messages(enum todo_command command,
strbuf_add_commented_lines(&buf, body, strlen(body));
} else
return error(_("unknown command: %d"), command);
- unuse_commit_buffer(commit, message);
+ repo_unuse_commit_buffer(the_repository, commit, message);
res = write_message(buf.buf, buf.len, rebase_path_squash_msg(), 0);
strbuf_release(&buf);
@@ -1729,7 +1735,7 @@ static int do_pick_commit(enum todo_command command, struct commit *commit,
msg_file = NULL;
goto fast_forward_edit;
}
- if (parent && parse_commit(parent) < 0)
+ if (parent && repo_parse_commit(the_repository, parent) < 0)
/* TRANSLATORS: The first %s will be a "todo" command like
"revert" or "pick", the second %s a SHA1. */
return error(_("%s: cannot parse parent commit %s"),
@@ -2357,7 +2363,8 @@ static int walk_revs_populate_todo(struct todo_list *todo_list,
subject_len = find_commit_subject(commit_buffer, &subject);
strbuf_addf(&todo_list->buf, "%s %s %.*s\n", command_string,
short_commit_name(commit), subject_len, subject);
- unuse_commit_buffer(commit, commit_buffer);
+ repo_unuse_commit_buffer(the_repository, commit,
+ commit_buffer);
}
if (!todo_list->nr)
@@ -2616,7 +2623,8 @@ static int make_patch(struct commit *commit, struct replay_opts *opts)
const char *commit_buffer = get_commit_buffer(commit, NULL);
find_commit_subject(commit_buffer, &subject);
res |= write_message(subject, strlen(subject), buf.buf, 1);
- unuse_commit_buffer(commit, commit_buffer);
+ repo_unuse_commit_buffer(the_repository, commit,
+ commit_buffer);
}
strbuf_release(&buf);
@@ -3013,7 +3021,7 @@ static int do_merge(struct commit *commit, const char *arg, int arg_len,
find_commit_subject(message, &body);
len = strlen(body);
ret = write_message(body, len, git_path_merge_msg(the_repository), 0);
- unuse_commit_buffer(commit, message);
+ repo_unuse_commit_buffer(the_repository, commit, message);
if (ret) {
error_errno(_("could not write '%s'"),
git_path_merge_msg(the_repository));
@@ -3658,11 +3666,13 @@ static int commit_staged_changes(struct replay_opts *opts,
if (parse_head(&commit) ||
!(p = get_commit_buffer(commit, NULL)) ||
write_message(p, strlen(p), path, 0)) {
- unuse_commit_buffer(commit, p);
+ repo_unuse_commit_buffer(the_repository,
+ commit, p);
return error(_("could not write file: "
"'%s'"), path);
}
- unuse_commit_buffer(commit, p);
+ repo_unuse_commit_buffer(the_repository,
+ commit, p);
}
}
@@ -4564,7 +4574,7 @@ int skip_unnecessary_picks(void)
continue;
if (item->command != TODO_PICK)
break;
- if (parse_commit(item->commit)) {
+ if (repo_parse_commit(the_repository, item->commit)) {
todo_list_release(&todo_list);
return error(_("could not parse commit '%s'"),
oid_to_hex(&item->commit->object.oid));
@@ -4691,12 +4701,13 @@ int rearrange_squash(void)
*commit_todo_item_at(&commit_todo, item->commit) = item;
- parse_commit(item->commit);
+ repo_parse_commit(the_repository, item->commit);
commit_buffer = get_commit_buffer(item->commit, NULL);
find_commit_subject(commit_buffer, &subject);
format_subject(&buf, subject, " ");
subject = subjects[i] = strbuf_detach(&buf, &subject_len);
- unuse_commit_buffer(item->commit, commit_buffer);
+ repo_unuse_commit_buffer(the_repository, item->commit,
+ commit_buffer);
if ((skip_prefix(subject, "fixup! ", &p) ||
skip_prefix(subject, "squash! ", &p))) {
struct commit *commit2;
diff --git a/sha1-file.c b/sha1-file.c
index ce47524679..4f86034d25 100644
--- a/sha1-file.c
+++ b/sha1-file.c
@@ -1456,7 +1456,8 @@ void *read_object_with_reference(const struct object_id *oid,
int ref_length = -1;
const char *ref_type = NULL;
- buffer = read_object_file(&actual_oid, &type, &isize);
+ buffer = repo_read_object_file(the_repository, &actual_oid,
+ &type, &isize);
if (!buffer)
return NULL;
if (type == required_type) {
diff --git a/sha1-name.c b/sha1-name.c
index c9cc1318b7..71e6bb7a9a 100644
--- a/sha1-name.c
+++ b/sha1-name.c
@@ -357,7 +357,8 @@ static int show_ambiguous_object(const struct object_id *oid, void *data)
if (commit) {
struct pretty_print_context pp = {0};
pp.date_mode.type = DATE_SHORT;
- format_commit_message(commit, " %ad - %s", &desc, &pp);
+ repo_format_commit_message(the_repository, commit,
+ " %ad - %s", &desc, &pp);
}
} else if (type == OBJ_TAG) {
struct tag *tag = lookup_tag(the_repository, oid);
@@ -849,7 +850,7 @@ static int get_parent(const char *name, int len,
if (ret)
return ret;
commit = lookup_commit_reference(the_repository, &oid);
- if (parse_commit(commit))
+ if (repo_parse_commit(the_repository, commit))
return -1;
if (!idx) {
oidcpy(result, &commit->object.oid);
@@ -881,7 +882,7 @@ static int get_nth_ancestor(const char *name, int len,
return -1;
while (generation--) {
- if (parse_commit(commit) || !commit->parents)
+ if (repo_parse_commit(the_repository, commit) || !commit->parents)
return -1;
commit = commit->parents->item;
}
@@ -1152,7 +1153,7 @@ static int get_oid_oneline(const char *prefix, struct object_id *oid,
buf = get_commit_buffer(commit, NULL);
p = strstr(buf, "\n\n");
matches = negative ^ (p && !regexec(®ex, p + 2, 0, NULL, 0));
- unuse_commit_buffer(commit, buf);
+ repo_unuse_commit_buffer(the_repository, commit, buf);
if (matches) {
oidcpy(oid, &commit->object.oid);
diff --git a/shallow.c b/shallow.c
index e656ce8b9c..124c03f003 100644
--- a/shallow.c
+++ b/shallow.c
@@ -210,7 +210,7 @@ struct commit_list *get_shallow_commits_by_rev_list(int ac, const char **av,
struct commit *c = p->item;
struct commit_list *parent;
- if (parse_commit(c))
+ if (repo_parse_commit(the_repository, c))
die("unable to parse commit %s",
oid_to_hex(&c->object.oid));
@@ -533,7 +533,7 @@ static void paint_down(struct paint_info *info, const struct object_id *oid,
if (c->object.flags & BOTTOM)
continue;
- if (parse_commit(c))
+ if (repo_parse_commit(the_repository, c))
die("unable to parse commit %s",
oid_to_hex(&c->object.oid));
diff --git a/submodule-config.c b/submodule-config.c
index fc2c41b947..4b2104979b 100644
--- a/submodule-config.c
+++ b/submodule-config.c
@@ -552,7 +552,8 @@ static const struct submodule *config_from(struct submodule_cache *cache,
if (submodule)
goto out;
- config = read_object_file(&oid, &type, &config_size);
+ config = repo_read_object_file(the_repository, &oid, &type,
+ &config_size);
if (!config || type != OBJ_BLOB)
goto out;
diff --git a/t/helper/test-revision-walking.c b/t/helper/test-revision-walking.c
index 625b2dbf82..a03cfcdae1 100644
--- a/t/helper/test-revision-walking.c
+++ b/t/helper/test-revision-walking.c
@@ -19,7 +19,8 @@ static void print_commit(struct commit *commit)
struct strbuf sb = STRBUF_INIT;
struct pretty_print_context ctx = {0};
ctx.date_mode.type = DATE_NORMAL;
- format_commit_message(commit, " %m %s", &sb, &ctx);
+ repo_format_commit_message(the_repository, commit, " %m %s", &sb,
+ &ctx);
printf("%s\n", sb.buf);
strbuf_release(&sb);
}
diff --git a/tag.c b/tag.c
index 1db663d716..b5a5a4c152 100644
--- a/tag.c
+++ b/tag.c
@@ -52,7 +52,7 @@ int gpg_verify_tag(const struct object_id *oid, const char *name_to_report,
find_unique_abbrev(oid, DEFAULT_ABBREV),
type_name(type));
- buf = read_object_file(oid, &type, &size);
+ buf = repo_read_object_file(the_repository, oid, &type, &size);
if (!buf)
return error("%s: unable to read file.",
name_to_report ?
@@ -200,7 +200,8 @@ int parse_tag(struct tag *item)
if (item->object.parsed)
return 0;
- data = read_object_file(&item->object.oid, &type, &size);
+ data = repo_read_object_file(the_repository, &item->object.oid, &type,
+ &size);
if (!data)
return error("Could not read %s",
oid_to_hex(&item->object.oid));
diff --git a/tree-walk.c b/tree-walk.c
index 77b37f36fa..ec6045828b 100644
--- a/tree-walk.c
+++ b/tree-walk.c
@@ -715,8 +715,10 @@ enum follow_symlinks_result get_tree_entry_follow_symlinks(struct object_id *tre
*/
retval = DANGLING_SYMLINK;
- contents = read_object_file(¤t_tree_oid, &type,
- &link_len);
+ contents = repo_read_object_file(the_repository,
+ ¤t_tree_oid,
+ &type,
+ &link_len);
if (!contents)
goto done;
diff --git a/tree.c b/tree.c
index 215d3fdc7c..e741180482 100644
--- a/tree.c
+++ b/tree.c
@@ -105,7 +105,7 @@ static int read_tree_1(struct tree *tree, struct strbuf *base,
oid_to_hex(entry.oid),
base->buf, entry.path);
- if (parse_commit(commit))
+ if (repo_parse_commit(the_repository, commit))
die("Invalid commit %s in submodule path %s%s",
oid_to_hex(entry.oid),
base->buf, entry.path);
@@ -221,7 +221,8 @@ int parse_tree_gently(struct tree *item, int quiet_on_missing)
if (item->object.parsed)
return 0;
- buffer = read_object_file(&item->object.oid, &type, &size);
+ buffer = repo_read_object_file(the_repository, &item->object.oid,
+ &type, &size);
if (!buffer)
return quiet_on_missing ? -1 :
error("Could not read %s",
diff --git a/walker.c b/walker.c
index 96990d84da..3e398cd485 100644
--- a/walker.c
+++ b/walker.c
@@ -78,7 +78,7 @@ static int process_commit(struct walker *walker, struct commit *commit)
{
struct commit_list *parents;
- if (parse_commit(commit))
+ if (repo_parse_commit(the_repository, commit))
return -1;
while (complete && complete->item->date >= commit->date) {
diff --git a/xdiff-interface.c b/xdiff-interface.c
index ec6e574e4a..742553de49 100644
--- a/xdiff-interface.c
+++ b/xdiff-interface.c
@@ -192,7 +192,7 @@ void read_mmblob(mmfile_t *ptr, const struct object_id *oid)
return;
}
- ptr->ptr = read_object_file(oid, &type, &size);
+ ptr->ptr = repo_read_object_file(the_repository, oid, &type, &size);
if (!ptr->ptr || type != OBJ_BLOB)
die("unable to read blob object %s", oid_to_hex(oid));
ptr->size = size;
--
2.19.0
^ permalink raw reply related [relevance 2%]
* [PATCH 18/19] submodule: don't add submodule as odb for push
2018-10-11 21:17 6% [RFC PATCH 00/19] Bring more repository handles into our code base Stefan Beller
2018-10-11 21:17 18% ` [PATCH 17/19] submodule: use submodule repos for object lookup Stefan Beller
@ 2018-10-11 21:17 14% ` Stefan Beller
2018-10-11 21:17 2% ` [PATCH 19/19] Apply semantic patches from previous patches Stefan Beller
2018-10-11 23:31 2% ` [RFC PATCH 00/19] Bring more repository handles into our code base Junio C Hamano
3 siblings, 1 reply; 200+ results
From: Stefan Beller @ 2018-10-11 21:17 UTC (permalink / raw)
To: git; +Cc: jonathantanmy, Stefan Beller
The submodule was added as an alternative in eb21c732d6 (push: teach
--recurse-submodules the on-demand option, 2012-03-29), but was
not explained, why.
In similar code, submodule_has_commits, the submodule is added as an
alternative to perform a quick check if we need to dive into the submodule.
However in push_submodule
(a) for_each_remote_ref_submodule will also provide the quick check and
(b) after that we don't need to have submodule objects around, as all
further code is to spawn a separate process.
Signed-off-by: Stefan Beller <sbeller@google.com>
---
submodule.c | 3 ---
1 file changed, 3 deletions(-)
diff --git a/submodule.c b/submodule.c
index 5e1a6c0b7c..f70d75ef45 100644
--- a/submodule.c
+++ b/submodule.c
@@ -1006,9 +1006,6 @@ static int push_submodule(const char *path,
const struct string_list *push_options,
int dry_run)
{
- if (add_submodule_odb(path))
- return 1;
-
if (for_each_remote_ref_submodule(path, has_remote, NULL) > 0) {
struct child_process cp = CHILD_PROCESS_INIT;
argv_array_push(&cp.args, "push");
--
2.19.0
^ permalink raw reply related [relevance 14%]
* [PATCH 17/19] submodule: use submodule repos for object lookup
2018-10-11 21:17 6% [RFC PATCH 00/19] Bring more repository handles into our code base Stefan Beller
@ 2018-10-11 21:17 18% ` Stefan Beller
2018-10-11 21:17 14% ` [PATCH 18/19] submodule: don't add submodule as odb for push Stefan Beller
` (2 subsequent siblings)
3 siblings, 1 reply; 200+ results
From: Stefan Beller @ 2018-10-11 21:17 UTC (permalink / raw)
To: git; +Cc: jonathantanmy, Stefan Beller
This converts the 'show_submodule_header' function to use
the repository API properly, such that the submodule objects
are not added to the main object store.
Signed-off-by: Stefan Beller <sbeller@google.com>
---
submodule.c | 48 ++++++++++++++++++++++++++++++++++++++----------
1 file changed, 38 insertions(+), 10 deletions(-)
diff --git a/submodule.c b/submodule.c
index 442229bb49..5e1a6c0b7c 100644
--- a/submodule.c
+++ b/submodule.c
@@ -443,7 +443,7 @@ static int prepare_submodule_summary(struct rev_info *rev, const char *path,
return prepare_revision_walk(rev);
}
-static void print_submodule_summary(struct rev_info *rev, struct diff_options *o)
+static void print_submodule_summary(struct repository *r, struct rev_info *rev, struct diff_options *o)
{
static const char format[] = " %m %s";
struct strbuf sb = STRBUF_INIT;
@@ -454,7 +454,8 @@ static void print_submodule_summary(struct rev_info *rev, struct diff_options *o
ctx.date_mode = rev->date_mode;
ctx.output_encoding = get_log_output_encoding();
strbuf_setlen(&sb, 0);
- format_commit_message(commit, format, &sb, &ctx);
+ repo_format_commit_message(r, commit, format, &sb,
+ &ctx);
strbuf_addch(&sb, '\n');
if (commit->object.flags & SYMMETRIC_LEFT)
diff_emit_submodule_del(o, sb.buf);
@@ -481,12 +482,37 @@ void prepare_submodule_repo_env(struct argv_array *out)
DEFAULT_GIT_DIR_ENVIRONMENT);
}
+/*
+ * Initialize 'out' based on the provided submodule path.
+ *
+ * Unlike repo_submodule_init, this tolerates submodules not present
+ * in .gitmodules. NEEDSWORK: The repo_submodule_init behavior is
+ * preferrable. This function exists only to preserve historical behavior.
+ *
+ * Returns 0 on success, -1 when the submodule is not present.
+ */
+static int open_submodule(struct repository *out, const char *path)
+{
+ struct strbuf sb = STRBUF_INIT;
+
+ if (submodule_to_gitdir(&sb, path) || repo_init(out, sb.buf, NULL)) {
+ strbuf_release(&sb);
+ return -1;
+ }
+
+ out->submodule_prefix = xstrdup(path);
+
+ strbuf_release(&sb);
+ return 0;
+}
+
/* Helper function to display the submodule header line prior to the full
* summary output. If it can locate the submodule objects directory it will
* attempt to lookup both the left and right commits and put them into the
* left and right pointers.
*/
-static void show_submodule_header(struct diff_options *o, const char *path,
+static void show_submodule_header(struct diff_options *o, struct repository *sub,
+ const char *path,
struct object_id *one, struct object_id *two,
unsigned dirty_submodule,
struct commit **left, struct commit **right,
@@ -507,7 +533,7 @@ static void show_submodule_header(struct diff_options *o, const char *path,
else if (is_null_oid(two))
message = "(submodule deleted)";
- if (add_submodule_odb(path)) {
+ if (open_submodule(sub, path) < 0) {
if (!message)
message = "(commits not present)";
goto output_header;
@@ -517,8 +543,8 @@ static void show_submodule_header(struct diff_options *o, const char *path,
* Attempt to lookup the commit references, and determine if this is
* a fast forward or fast backwards update.
*/
- *left = lookup_commit_reference(the_repository, one);
- *right = lookup_commit_reference(the_repository, two);
+ *left = lookup_commit_reference(sub, one);
+ *right = lookup_commit_reference(sub, two);
/*
* Warn about missing commits in the submodule project, but only if
@@ -528,7 +554,7 @@ static void show_submodule_header(struct diff_options *o, const char *path,
(!is_null_oid(two) && !*right))
message = "(commits not present)";
- *merge_bases = get_merge_bases(*left, *right);
+ *merge_bases = repo_get_merge_bases(sub, *left, *right);
if (*merge_bases) {
if ((*merge_bases)->item == *left)
fast_forward = 1;
@@ -562,8 +588,9 @@ void show_submodule_summary(struct diff_options *o, const char *path,
struct rev_info rev;
struct commit *left = NULL, *right = NULL;
struct commit_list *merge_bases = NULL;
+ struct repository sub;
- show_submodule_header(o, path, one, two, dirty_submodule,
+ show_submodule_header(o, &sub, path, one, two, dirty_submodule,
&left, &right, &merge_bases);
/*
@@ -580,7 +607,7 @@ void show_submodule_summary(struct diff_options *o, const char *path,
goto out;
}
- print_submodule_summary(&rev, o);
+ print_submodule_summary(&sub, &rev, o);
out:
if (merge_bases)
@@ -598,8 +625,9 @@ void show_submodule_inline_diff(struct diff_options *o, const char *path,
struct commit_list *merge_bases = NULL;
struct child_process cp = CHILD_PROCESS_INIT;
struct strbuf sb = STRBUF_INIT;
+ struct repository sub;
- show_submodule_header(o, path, one, two, dirty_submodule,
+ show_submodule_header(o, &sub, path, one, two, dirty_submodule,
&left, &right, &merge_bases);
/* We need a valid left and right commit to display a difference */
--
2.19.0
^ permalink raw reply related [relevance 18%]
* [RFC PATCH 00/19] Bring more repository handles into our code base
@ 2018-10-11 21:17 6% Stefan Beller
2018-10-11 21:17 18% ` [PATCH 17/19] submodule: use submodule repos for object lookup Stefan Beller
` (3 more replies)
0 siblings, 4 replies; 200+ results
From: Stefan Beller @ 2018-10-11 21:17 UTC (permalink / raw)
To: git; +Cc: jonathantanmy, Stefan Beller
This applies on nd/the-index (b3c7eef9b05) and is the logical continuation of
the object store series, which I sent over the last year.
The previous series did take a very slow and pedantic approach,
using a #define trick, see cfc62fc98c for details, but it turns out,
that it doesn't work:
When changing the signature of widely used functions, it burdens the
maintainer in resolving the semantic conflicts.
In the orginal approach this was called a feature, as then we can ensure
that not bugs creep into the code base during the merge window (while such
a refactoring series wanders from pu to master). It turns out this
was not well received and was just burdensome.
The #define trick doesn't buy us much to begin with when dealing with
non-merge-conflicts. For example, see deref_tag at tag.c:68, which got
the repository argument in 286d258d4f (tag.c: allow deref_tag to handle
arbitrary repositories, 2018-06-28) but lost its property of working on any
repository while 8c4cc32689 (tag: don't warn if target is missing but
promised, 2018-07-12) was in flight simultaneously.
Another example of failure of this approach is seen in patch 5, which
shows that the pedantry was missed.
This series takes another approach as it doesn't change the signature of
functions, but introduces new functions that can deal with arbitrary
repositories, keeping the old function signature around using a shallow wrapper.
Additionally each patch adds a semantic patch, that would port from the old to
the new function. These semantic patches are all applied in the very last patch,
but we could omit applying the last patch if it causes too many merge conflicts
and trickl in the semantic patches over time when there are no merge conflicts.
The original goal of all these refactoring series was to remove add_submodule_odb
in submodule.c, which was partially reached with this series. I'll investigate the
remaining calls in another series, but it shows we're close to be done with these
large refactorings as far as I am concerned.
Thanks,
Stefan
Stefan Beller (19):
sha1_file: allow read_object to read objects in arbitrary repositories
packfile: allow has_packed_and_bad to handle arbitrary repositories
object-store: allow read_object_file_extended to read from arbitrary
repositories
object-store: prepare read_object_file to deal with arbitrary
repositories
object: parse_object to honor its repository argument
commit: allow parse_commit* to handle arbitrary repositories
commit.c: allow paint_down_to_common to handle arbitrary repositories
commit.c: allow merge_bases_many to handle arbitrary repositories
commit.c: allow remove_redundant to handle arbitrary repositories
commit: allow get_merge_bases_many_0 to handle arbitrary repositories
commit: prepare get_merge_bases to handle arbitrary repositories
commit: prepare get_commit_buffer to handle arbitrary repositories
commit: prepare in_merge_bases[_many] to handle arbitrary repositories
commit: prepare repo_unuse_commit_buffer to handle arbitrary
repositories
commit: prepare logmsg_reencode to handle arbitrary repositories
pretty: prepare format_commit_message to handle arbitrary repositories
submodule: use submodule repos for object lookup
submodule: don't add submodule as odb for push
Apply semantic patches from previous patches
apply.c | 6 +-
archive.c | 5 +-
bisect.c | 5 +-
blame.c | 15 +--
builtin/am.c | 2 +-
builtin/blame.c | 4 +-
builtin/cat-file.c | 21 +++--
builtin/checkout.c | 4 +-
builtin/commit.c | 13 ++-
builtin/describe.c | 4 +-
builtin/difftool.c | 3 +-
builtin/fast-export.c | 7 +-
builtin/fmt-merge-msg.c | 8 +-
builtin/grep.c | 2 +-
builtin/index-pack.c | 8 +-
builtin/log.c | 4 +-
builtin/merge-base.c | 2 +-
builtin/merge-tree.c | 9 +-
builtin/mktag.c | 3 +-
builtin/name-rev.c | 2 +-
builtin/notes.c | 12 ++-
builtin/pack-objects.c | 22 +++--
builtin/reflog.c | 5 +-
builtin/replace.c | 2 +-
builtin/shortlog.c | 5 +-
builtin/show-branch.c | 4 +-
builtin/tag.c | 4 +-
builtin/unpack-file.c | 2 +-
builtin/unpack-objects.c | 3 +-
builtin/verify-commit.c | 2 +-
bundle.c | 2 +-
combine-diff.c | 2 +-
commit-graph.c | 8 +-
commit.c | 120 ++++++++++++++----------
commit.h | 67 ++++++++++---
config.c | 2 +-
contrib/coccinelle/the_repository.cocci | 114 ++++++++++++++++++++++
diff.c | 3 +-
dir.c | 2 +-
entry.c | 3 +-
fast-import.c | 7 +-
fsck.c | 9 +-
grep.c | 3 +-
http-push.c | 3 +-
log-tree.c | 3 +-
mailmap.c | 2 +-
match-trees.c | 4 +-
merge-blobs.c | 6 +-
merge-recursive.c | 13 +--
negotiator/default.c | 6 +-
negotiator/skipping.c | 2 +-
notes-cache.c | 5 +-
notes-merge.c | 4 +-
notes-utils.c | 2 +-
notes.c | 10 +-
object-store.h | 13 ++-
object.c | 2 +-
packfile.c | 5 +-
packfile.h | 2 +-
pretty.c | 33 ++++---
pretty.h | 7 +-
read-cache.c | 5 +-
remote-testsvn.c | 4 +-
remote.c | 2 +-
rerere.c | 5 +-
revision.c | 12 +--
sequencer.c | 55 ++++++-----
sha1-file.c | 22 +++--
sha1-name.c | 9 +-
shallow.c | 4 +-
streaming.c | 2 +-
submodule-config.c | 3 +-
submodule.c | 51 +++++++---
t/helper/test-revision-walking.c | 3 +-
tag.c | 5 +-
tree-walk.c | 6 +-
tree.c | 5 +-
walker.c | 2 +-
xdiff-interface.c | 2 +-
79 files changed, 571 insertions(+), 278 deletions(-)
create mode 100644 contrib/coccinelle/the_repository.cocci
--
2.19.0
^ permalink raw reply [relevance 6%]
* Re: What's cooking in git.git (Oct 2018, #01; Wed, 10)
2018-10-10 18:55 4% ` Stefan Beller
@ 2018-10-11 2:00 2% ` Junio C Hamano
0 siblings, 0 replies; 200+ results
From: Junio C Hamano @ 2018-10-11 2:00 UTC (permalink / raw)
To: Stefan Beller; +Cc: git
Stefan Beller <sbeller@google.com> writes:
>> * pw/diff-color-moved-ws-fix (2018-10-04) 5 commits
> I would suggest merging to 'next'.
OK.
>> * sb/strbuf-h-update (2018-09-29) 1 commit
> The patch as-is just adds names everywhere.
> I'd be happy to resend with either
> (a) not enforcing names everywhere, but only as needed or
> (b) having names everywhere, capitalizing them NAMES in
> the doc comment.
>
> I am tempted to ask for
> (c) take as-is, defer the rewording of doc strings for a follow up patch.
As long as the planned update eventually comes before all of us
forget, (c) is fine by me. I'll mark it to be merged to 'next' for
now, and follow through that plan, unless somebody else stops me
before it happens.
>> * sb/submodule-recursive-fetch-gets-the-tip (2018-09-12) 9 commits
> Will resend after a local review.
OK.
Thanks for helping me in updating the status for various topics.
^ permalink raw reply [relevance 2%]
* Re: [PATCH v6 09/10] submodule: support reading .gitmodules when it's not in the working tree
2018-10-10 18:56 7% ` Antonio Ospite
@ 2018-10-10 22:55 7% ` Stefan Beller
0 siblings, 0 replies; 200+ results
From: Stefan Beller @ 2018-10-10 22:55 UTC (permalink / raw)
To: Antonio Ospite
Cc: Junio C Hamano, git, Jonathan Nieder, Jeff King,
SZEDER Gábor
On Wed, Oct 10, 2018 at 11:56 AM Antonio Ospite <ao2@ao2.it> wrote:
>
> On Mon, 8 Oct 2018 15:19:00 -0700
> Stefan Beller <sbeller@google.com> wrote:
>
> > > +test_expect_success 'not writing gitmodules config file when it is not checked out' '
> > > + test_must_fail git -C super submodule--helper config submodule.submodule.url newurl
> >
> > This only checks the exit code, do we also want to check for
> >
> > test_path_is_missing .gitmodules ?
> >
>
> OK, I agree, let's re-check also *after* we tried and failed to set
> a config value, just to be sure that the code does not get accidentally
> changed in the future to create the file. I'll add the check.
>
> > > +test_expect_success 'initialising submodule when the gitmodules config is not checked out' '
> > > + git -C super submodule init
> > > +'
> > > +
> > > +test_expect_success 'showing submodule summary when the gitmodules config is not checked out' '
> > > + git -C super submodule summary
> > > +'
> >
> > Same for these, is the exit code enough, or do we want to look at
> > specific things?
> >
>
> Except for the "summary" test which was not even exercising the
> config_from_gitmodule path, checking exist status should be sufficient
> to verify that "submodule--helper config" does not fail, but we can
> surely do better.
>
> I will add checks to confirm that not only the commands exited without
> errors but they also achieved the desired effect, to validate the actual
> high-level use case advertised by the test file. This should be more
> future-proof.
>
> And I think I'll merge the summary and the update tests.
>
> > > +
> > > +test_expect_success 'updating submodule when the gitmodules config is not checked out' '
> > > + (cd submodule &&
> > > + echo file2 >file2 &&
> > > + git add file2 &&
> > > + git commit -m "add file2 to submodule"
> > > + ) &&
> > > + git -C super submodule update
> >
> > git status would want to be clean afterwards?
>
> Mmh, this should have been "submodule update --remote" in the first
> place to have any effect, I'll take the chance and rewrite this test in
> a different way and also check the effect of the update operation, and
> the repository status.
>
> I'll be something like this:
>
> ORIG_SUBMODULE=$(git -C submodule rev-parse HEAD)
> ORIG_UPSTREAM=$(git -C upstream rev-parse HEAD)
> ORIG_SUPER=$(git -C super rev-parse HEAD)
>
> test_expect_success 're-updating submodule when the gitmodules config is not checked out' '
> test_when_finished "git -C submodule reset --hard $ORIG_SUBMODULE;
> git -C upstream reset --hard $ORIG_UPSTREAM;
> git -C super reset --hard $ORIG_SUPER;
> git -C upstream submodule update --remote;
> git -C super pull;
> git -C super submodule update --remote" &&
> (cd submodule &&
> echo file2 >file2 &&
> git add file2 &&
> test_tick &&
> git commit -m "add file2 to submodule"
> ) &&
> (cd upstream &&
> git submodule update --remote &&
> git add submodule &&
> test_tick &&
> git commit -m "Update submodule"
> ) &&
> git -C super pull &&
> # The --for-status options reads the gitmdoules config
gitmodules
> git -C super submodule summary --for-status >actual &&
> cat >expect <<-\EOF &&
> * submodule 951c301...a939200 (1):
hardcoding hash values burdens the plan to migrate to another
hash function,
rev1=$(git -C submodule rev-parse --short HEAD^)
rev2=$(git -C submodule rev-parse --short HEAD)
and then use ${rev1}..${rev2} ?
> < add file2 to submodule
>
> EOF
> test_cmp expect actual &&
> # Test that the update actually succeeds
> test_path_is_missing super/submodule/file2 &&
> git -C super submodule update &&
> test_cmp submodule/file2 super/submodule/file2 &&
> git -C super status --short >output &&
> test_must_be_empty output
> '
>
> Maybe a little overkill?
Wow, very thorough! You might call it overkill, but now that you have it...
> The "upstream" repo will be added in test 1 to better clarify the roles
> of the involved repositories.
>
> The commit ids should be stable because of test_tick, shouldn't they?
Yes, but see
Documentation/technical/hash-function-transition.txt
that a couple people are working on. Let's be nice to them. :-)
Stefan
^ permalink raw reply [relevance 7%]
* Re: [PATCH] builtin/grep.c: remote superflous submodule code
@ 2018-10-10 22:49 7% ` Stefan Beller
0 siblings, 0 replies; 200+ results
From: Stefan Beller @ 2018-10-10 22:49 UTC (permalink / raw)
To: Jonathan Tan; +Cc: Junio C Hamano, Antonio Ospite, git
On Tue, Oct 9, 2018 at 5:10 PM Jonathan Tan <jonathantanmy@google.com> wrote:
>
> > It claimed that grep would still need some explicit handling, but that is
> > not the call to repo_read_gitmodules (applying this patch on top of
> > ff6f1f564c4 still keep the test suite happy, specifically
> > t7814-grep-recurse-submodules, which contains a test
> > "grep history with moved submoules")
>
> Firstly, spelling of "remove" and "superfluous" in the commit title.
>
> I don't think the "grep history with moved submodules" test exercises
> much. That test only tests the superproject > submodule case, but we
> need a superproject > submodule > sub-submodule case, because what is
> being removed is a call to repo_read_gitmodules() on a repository
> ("struct repository submodule") that has a superproject ("struct
> repository *superproject"). In other words, we need a submodule that has
> its own gitmodules.
Right; we do have a test 'grep and nested submodules', which still passes.
I added another test, that would grep through nested submodules in
the history (not checked out), but that would not work on nested submodules
with or without this patch applied. (As the nested submodule is not checked
out, is_submodule_active(repo, path) would return false and we'd not dive
into the nested submodule.
I looked into ao/submodule-wo-gitmodules-checked-out, as that touches
this area of code as well and promises to allow working with submodules
when .gitmodules is not checked out, it doesn't help this use case, either.
That is (as Antonio diagnosed), due to get_oid not working with a repository
handle, yet.
> > The special handling is the call to gitmodules_config_oid which was added
> > already in 74ed43711f (grep: enable recurse-submodules to work on
> > <tree> objects, 2016-12-16), but then was still named
> > gitmodules_config_sha1.
>
> If you're stating that gitmodules_config_oid() is where the .gitmodules
> file is lazily loaded, it doesn't seem to be that way, because that
> function works only on the_repository (at least on 'master' and 'next').
yes, that is why nested submodules do not work currently when they
are not in the working tree.
>
> > This is a resend of origin/sb/grep-submodule-cleanup,
> > and I think picking ff6f1f564c4 as the base for the series would
> > also be appropriate.
>
> Any particular reason why you suggest that commit (which is more than a
> year old)? It seems that basing this on 'master' is fine.
After more analysis, I think we'd want to wait for Antonios series to land
and then build on top of that, while also getting get_oid converted.
Regarding this patch, let's retract it for now and revisit it once we have
more submodule infrastructure working.
Thanks,
Stefan
^ permalink raw reply [relevance 7%]
* [PATCH v5 9/9] builtin/fetch: check for submodule updates for non branch fetches
2018-10-10 21:49 10% [PATCH v5 0/9] fetch: make sure submodule oids are fetched Stefan Beller
` (6 preceding siblings ...)
2018-10-10 21:50 23% ` [PATCH v5 8/9] fetch: retry fetching submodules if needed objects were not fetched Stefan Beller
@ 2018-10-10 21:50 21% ` Stefan Beller
7 siblings, 0 replies; 200+ results
From: Stefan Beller @ 2018-10-10 21:50 UTC (permalink / raw)
To: git; +Cc: Stefan Beller
Gerrit, the code review tool, has a different workflow than our mailing
list based approach. Usually users upload changes to a Gerrit server and
continuous integration and testing happens by bots. Sometimes however a
user wants to checkout a change locally and look at it locally. For this
use case, Gerrit offers a command line snippet to copy and paste to your
terminal, which looks like
git fetch https://<host>/gerrit refs/changes/<id> &&
git checkout FETCH_HEAD
For Gerrit changes that contain changing submodule gitlinks, it would be
easy to extend both the fetch and checkout with the '--recurse-submodules'
flag, such that this command line snippet would produce the state of a
change locally.
However the functionality added in the previous patch, which would
ensure that we fetch the objects in the submodule that the gitlink pointed
at, only works for remote tracking branches so far, not for FETCH_HEAD.
Make sure that fetching a superproject to its FETCH_HEAD, also respects
the existence checks for objects in the submodule recursion.
Signed-off-by: Stefan Beller <sbeller@google.com>
---
builtin/fetch.c | 5 ++++-
t/t5526-fetch-submodules.sh | 2 +-
2 files changed, 5 insertions(+), 2 deletions(-)
diff --git a/builtin/fetch.c b/builtin/fetch.c
index e3b03ad3bd..f2d9e548bf 100644
--- a/builtin/fetch.c
+++ b/builtin/fetch.c
@@ -894,11 +894,14 @@ static int store_updated_refs(const char *raw_url, const char *remote_name,
rc |= update_local_ref(ref, what, rm, ¬e,
summary_width);
free(ref);
- } else
+ } else {
+ check_for_new_submodule_commits(&rm->old_oid);
format_display(¬e, '*',
*kind ? kind : "branch", NULL,
*what ? what : "HEAD",
"FETCH_HEAD", summary_width);
+ }
+
if (note.len) {
if (verbosity >= 0 && !shown_url) {
fprintf(stderr, _("From %.*s\n"),
diff --git a/t/t5526-fetch-submodules.sh b/t/t5526-fetch-submodules.sh
index af12c50e7d..a509eabb04 100755
--- a/t/t5526-fetch-submodules.sh
+++ b/t/t5526-fetch-submodules.sh
@@ -615,7 +615,7 @@ test_expect_success "fetch new commits on-demand when they are not reachable" '
git update-ref refs/changes/2 $D &&
(
cd downstream &&
- git fetch --recurse-submodules --recurse-submodules-default on-demand origin refs/changes/2:refs/heads/my_branch &&
+ git fetch --recurse-submodules origin refs/changes/2 &&
git -C submodule cat-file -t $C &&
git checkout --recurse-submodules FETCH_HEAD
)
--
2.19.0
^ permalink raw reply related [relevance 21%]
* [PATCH v5 8/9] fetch: retry fetching submodules if needed objects were not fetched
2018-10-10 21:49 10% [PATCH v5 0/9] fetch: make sure submodule oids are fetched Stefan Beller
` (5 preceding siblings ...)
2018-10-10 21:50 26% ` [PATCH v5 7/9] submodule: fetch in submodules git directory instead of in worktree Stefan Beller
@ 2018-10-10 21:50 23% ` Stefan Beller
2018-10-10 21:50 21% ` [PATCH v5 9/9] builtin/fetch: check for submodule updates for non branch fetches Stefan Beller
7 siblings, 0 replies; 200+ results
From: Stefan Beller @ 2018-10-10 21:50 UTC (permalink / raw)
To: git; +Cc: Stefan Beller
Currently when git-fetch is asked to recurse into submodules, it dispatches
a plain "git-fetch -C <submodule-dir>" (with some submodule related options
such as prefix and recusing strategy, but) without any information of the
remote or the tip that should be fetched.
This works surprisingly well in some workflows (such as using submodules
as a third party library), while not so well in other scenarios, such
as in a Gerrit topic-based workflow, that can tie together changes
(potentially across repositories) on the server side. One of the parts
of such a Gerrit workflow is to download a change when wanting to examine
it, and you'd want to have its submodule changes that are in the same
topic downloaded as well. However these submodule changes reside in their
own repository in their own ref (refs/changes/<int>).
Retry fetching a submodule by object name if the object id that the
superproject points to, cannot be found.
This retrying does not happen when the "git fetch" done at the
superproject is not storing the fetched results in remote
tracking branches (i.e. instead just recording them to
FETCH_HEAD) in this step. A later patch will fix this.
builtin/fetch used to only inspect submodules when they were fetched
"on-demand", as in either on/off case it was clear whether the submodule
needs to be fetched. However to know whether we need to try fetching the
object ids, we need to identify the object names, which is done in this
function check_for_new_submodule_commits(), so we'll also run that code
in case the submodule recursion is set to "on".
Signed-off-by: Stefan Beller <sbeller@google.com>
---
builtin/fetch.c | 9 +-
submodule.c | 185 ++++++++++++++++++++++++++++++------
t/t5526-fetch-submodules.sh | 16 ++++
3 files changed, 177 insertions(+), 33 deletions(-)
diff --git a/builtin/fetch.c b/builtin/fetch.c
index 0696abfc2a..e3b03ad3bd 100644
--- a/builtin/fetch.c
+++ b/builtin/fetch.c
@@ -707,8 +707,7 @@ static int update_local_ref(struct ref *ref,
what = _("[new ref]");
}
- if ((recurse_submodules != RECURSE_SUBMODULES_OFF) &&
- (recurse_submodules != RECURSE_SUBMODULES_ON))
+ if (recurse_submodules != RECURSE_SUBMODULES_OFF)
check_for_new_submodule_commits(&ref->new_oid);
r = s_update_ref(msg, ref, 0);
format_display(display, r ? '!' : '*', what,
@@ -723,8 +722,7 @@ static int update_local_ref(struct ref *ref,
strbuf_add_unique_abbrev(&quickref, ¤t->object.oid, DEFAULT_ABBREV);
strbuf_addstr(&quickref, "..");
strbuf_add_unique_abbrev(&quickref, &ref->new_oid, DEFAULT_ABBREV);
- if ((recurse_submodules != RECURSE_SUBMODULES_OFF) &&
- (recurse_submodules != RECURSE_SUBMODULES_ON))
+ if (recurse_submodules != RECURSE_SUBMODULES_OFF)
check_for_new_submodule_commits(&ref->new_oid);
r = s_update_ref("fast-forward", ref, 1);
format_display(display, r ? '!' : ' ', quickref.buf,
@@ -738,8 +736,7 @@ static int update_local_ref(struct ref *ref,
strbuf_add_unique_abbrev(&quickref, ¤t->object.oid, DEFAULT_ABBREV);
strbuf_addstr(&quickref, "...");
strbuf_add_unique_abbrev(&quickref, &ref->new_oid, DEFAULT_ABBREV);
- if ((recurse_submodules != RECURSE_SUBMODULES_OFF) &&
- (recurse_submodules != RECURSE_SUBMODULES_ON))
+ if (recurse_submodules != RECURSE_SUBMODULES_OFF)
check_for_new_submodule_commits(&ref->new_oid);
r = s_update_ref("forced-update", ref, 1);
format_display(display, r ? '!' : '+', quickref.buf,
diff --git a/submodule.c b/submodule.c
index 3f791f2277..05799362e0 100644
--- a/submodule.c
+++ b/submodule.c
@@ -1127,8 +1127,12 @@ struct submodule_parallel_fetch {
int result;
struct string_list changed_submodule_names;
+ struct get_next_submodule_task **retry;
+ int retry_nr, retry_alloc;
};
-#define SPF_INIT {0, ARGV_ARRAY_INIT, NULL, NULL, 0, 0, 0, 0, STRING_LIST_INIT_DUP }
+#define SPF_INIT {0, ARGV_ARRAY_INIT, NULL, NULL, 0, 0, 0, 0, \
+ STRING_LIST_INIT_DUP, \
+ NULL, 0, 0}
static void calculate_changed_submodule_paths(
struct submodule_parallel_fetch *spf)
@@ -1233,6 +1237,56 @@ static int get_fetch_recurse_config(const struct submodule *submodule,
return spf->default_option;
}
+struct get_next_submodule_task {
+ struct repository *repo;
+ const struct submodule *sub;
+ unsigned free_sub : 1; /* Do we need to free the submodule? */
+ struct oid_array *commits;
+};
+
+static const struct submodule *get_default_submodule(const char *path)
+{
+ struct submodule *ret = NULL;
+ const char *name = default_name_or_path(path);
+
+ if (!name)
+ return NULL;
+
+ ret = xmalloc(sizeof(*ret));
+ memset(ret, 0, sizeof(*ret));
+ ret->path = name;
+ ret->name = name;
+
+ return (const struct submodule *) ret;
+}
+
+static struct get_next_submodule_task *get_next_submodule_task_create(
+ struct repository *r, const char *path)
+{
+ struct get_next_submodule_task *task = xmalloc(sizeof(*task));
+ memset(task, 0, sizeof(*task));
+
+ task->sub = submodule_from_path(r, &null_oid, path);
+ if (!task->sub) {
+ task->sub = get_default_submodule(path);
+ task->free_sub = 1;
+ }
+
+ return task;
+}
+
+static void get_next_submodule_task_release(struct get_next_submodule_task *p)
+{
+ if (p->free_sub)
+ free((void*)p->sub);
+ p->free_sub = 0;
+ p->sub = NULL;
+
+ if (p->repo)
+ repo_clear(p->repo);
+ FREE_AND_NULL(p->repo);
+}
+
static struct repository *get_submodule_repo_for(struct repository *r,
const struct submodule *sub)
{
@@ -1259,39 +1313,35 @@ static struct repository *get_submodule_repo_for(struct repository *r,
static int get_next_submodule(struct child_process *cp,
struct strbuf *err, void *data, void **task_cb)
{
- int ret = 0;
struct submodule_parallel_fetch *spf = data;
for (; spf->count < spf->r->index->cache_nr; spf->count++) {
- struct strbuf submodule_prefix = STRBUF_INIT;
+ int recurse_config;
const struct cache_entry *ce = spf->r->index->cache[spf->count];
const char *default_argv;
- const struct submodule *submodule;
- struct repository *repo;
- struct submodule default_submodule = SUBMODULE_INIT;
+ struct get_next_submodule_task *task;
if (!S_ISGITLINK(ce->ce_mode))
continue;
- submodule = submodule_from_path(spf->r, &null_oid, ce->name);
- if (!submodule) {
- const char *name = default_name_or_path(ce->name);
- if (name) {
- default_submodule.path = name;
- default_submodule.name = name;
- submodule = &default_submodule;
- }
+ task = get_next_submodule_task_create(spf->r, ce->name);
+
+ if (!task->sub) {
+ free(task);
+ continue;
}
- switch (get_fetch_recurse_config(submodule, spf))
+ recurse_config = get_fetch_recurse_config(task->sub, spf);
+
+ switch (recurse_config)
{
default:
case RECURSE_SUBMODULES_DEFAULT:
case RECURSE_SUBMODULES_ON_DEMAND:
- if (!submodule ||
+ if (!task->sub ||
!string_list_lookup(
&spf->changed_submodule_names,
- submodule->name))
+ task->sub->name))
continue;
default_argv = "on-demand";
break;
@@ -1302,12 +1352,12 @@ static int get_next_submodule(struct child_process *cp,
continue;
}
- strbuf_addf(&submodule_prefix, "%s%s/", spf->prefix, ce->name);
- repo = get_submodule_repo_for(spf->r, submodule);
- if (repo) {
+ task->repo = get_submodule_repo_for(spf->r, task->sub);
+ if (task->repo) {
+ struct strbuf submodule_prefix = STRBUF_INIT;
child_process_init(cp);
prepare_submodule_repo_env_in_gitdir(&cp->env_array);
- cp->dir = xstrdup(repo->gitdir);
+ cp->dir = task->repo->gitdir;
cp->git_cmd = 1;
if (!spf->quiet)
strbuf_addf(err, "Fetching submodule %s%s\n",
@@ -1316,18 +1366,51 @@ static int get_next_submodule(struct child_process *cp,
argv_array_pushv(&cp->args, spf->args.argv);
argv_array_push(&cp->args, default_argv);
argv_array_push(&cp->args, "--submodule-prefix");
+
+ strbuf_addf(&submodule_prefix, "%s%s/",
+ spf->prefix,
+ task->sub->path);
argv_array_push(&cp->args, submodule_prefix.buf);
- repo_clear(repo);
- free(repo);
- ret = 1;
- }
- strbuf_release(&submodule_prefix);
- if (ret) {
spf->count++;
+ *task_cb = task;
+
+ strbuf_release(&submodule_prefix);
return 1;
+ } else {
+ get_next_submodule_task_release(task);
+ free(task);
}
}
+
+ if (spf->retry_nr) {
+ struct get_next_submodule_task *task = spf->retry[spf->retry_nr - 1];
+ struct strbuf submodule_prefix = STRBUF_INIT;
+ spf->retry_nr--;
+
+ strbuf_addf(&submodule_prefix, "%s%s/", spf->prefix, task->sub->path);
+
+ child_process_init(cp);
+ prepare_submodule_repo_env_in_gitdir(&cp->env_array);
+ cp->git_cmd = 1;
+ cp->dir = task->repo->gitdir;
+
+ argv_array_init(&cp->args);
+ argv_array_pushv(&cp->args, spf->args.argv);
+ argv_array_push(&cp->args, "on-demand");
+ argv_array_push(&cp->args, "--submodule-prefix");
+ argv_array_push(&cp->args, submodule_prefix.buf);
+
+ /* NEEDSWORK: have get_default_remote from s--h */
+ argv_array_push(&cp->args, "origin");
+ oid_array_for_each_unique(task->commits,
+ append_oid_to_argv, &cp->args);
+
+ *task_cb = task;
+ strbuf_release(&submodule_prefix);
+ return 1;
+ }
+
return 0;
}
@@ -1335,20 +1418,68 @@ static int fetch_start_failure(struct strbuf *err,
void *cb, void *task_cb)
{
struct submodule_parallel_fetch *spf = cb;
+ struct get_next_submodule_task *task = task_cb;
spf->result = 1;
+ get_next_submodule_task_release(task);
return 0;
}
+static int commit_exists_in_sub(const struct object_id *oid, void *data)
+{
+ struct repository *subrepo = data;
+
+ enum object_type type = oid_object_info(subrepo, oid, NULL);
+
+ return type != OBJ_COMMIT;
+}
+
static int fetch_finish(int retvalue, struct strbuf *err,
void *cb, void *task_cb)
{
struct submodule_parallel_fetch *spf = cb;
+ struct get_next_submodule_task *task = task_cb;
+ const struct submodule *sub;
+
+ struct string_list_item *it;
+ struct oid_array *commits;
if (retvalue)
spf->result = 1;
+ if (!task)
+ return 0;
+
+ sub = task->sub;
+ if (!sub)
+ goto out;
+
+ it = string_list_lookup(&spf->changed_submodule_names, sub->name);
+ if (!it)
+ goto out;
+
+ commits = it->util;
+ oid_array_filter(commits,
+ commit_exists_in_sub,
+ task->repo);
+
+ /* Are there commits that do not exist? */
+ if (commits->nr) {
+ /* We already tried fetching them, do not try again. */
+ if (task->commits)
+ return 0;
+
+ task->commits = commits;
+ ALLOC_GROW(spf->retry, spf->retry_nr + 1, spf->retry_alloc);
+ spf->retry[spf->retry_nr] = task;
+ spf->retry_nr++;
+ return 0;
+ }
+
+out:
+ get_next_submodule_task_release(task);
+
return 0;
}
diff --git a/t/t5526-fetch-submodules.sh b/t/t5526-fetch-submodules.sh
index 42692219a1..af12c50e7d 100755
--- a/t/t5526-fetch-submodules.sh
+++ b/t/t5526-fetch-submodules.sh
@@ -605,4 +605,20 @@ test_expect_success "fetch new commits when submodule got renamed" '
test_cmp expect actual
'
+test_expect_success "fetch new commits on-demand when they are not reachable" '
+ git checkout --detach &&
+ C=$(git -C submodule commit-tree -m "new change outside refs/heads" HEAD^{tree}) &&
+ git -C submodule update-ref refs/changes/1 $C &&
+ git update-index --cacheinfo 160000 $C submodule &&
+ git commit -m "updated submodule outside of refs/heads" &&
+ D=$(git rev-parse HEAD) &&
+ git update-ref refs/changes/2 $D &&
+ (
+ cd downstream &&
+ git fetch --recurse-submodules --recurse-submodules-default on-demand origin refs/changes/2:refs/heads/my_branch &&
+ git -C submodule cat-file -t $C &&
+ git checkout --recurse-submodules FETCH_HEAD
+ )
+'
+
test_done
--
2.19.0
^ permalink raw reply related [relevance 23%]
* [PATCH v5 7/9] submodule: fetch in submodules git directory instead of in worktree
2018-10-10 21:49 10% [PATCH v5 0/9] fetch: make sure submodule oids are fetched Stefan Beller
` (4 preceding siblings ...)
2018-10-10 21:50 24% ` [PATCH v5 6/9] repository: repo_submodule_init to take a submodule struct Stefan Beller
@ 2018-10-10 21:50 26% ` Stefan Beller
2018-10-10 21:50 23% ` [PATCH v5 8/9] fetch: retry fetching submodules if needed objects were not fetched Stefan Beller
2018-10-10 21:50 21% ` [PATCH v5 9/9] builtin/fetch: check for submodule updates for non branch fetches Stefan Beller
7 siblings, 0 replies; 200+ results
From: Stefan Beller @ 2018-10-10 21:50 UTC (permalink / raw)
To: git; +Cc: Stefan Beller
This patch started as a refactoring to make 'get_next_submodule' more
readable, but upon doing so, I realized that "git fetch" of the submodule
actually doesn't need to be run in the submodules worktree. So let's run
it in its git dir instead.
That should pave the way towards fetching submodules that are currently
not checked out.
This patch leaks the cp->dir in get_next_submodule, as any further
callback in run_processes_parallel doesn't have access to the child
process any more. In an early iteration of this patch, the function
get_submodule_repo_for directly returned the string containing the
git directory, which would be a better design choice for this patch.
However the next patch both fixes the memory leak of cp->dir and also has
a use case for using the full repository handle of the submodule, so
it makes sense to introduce the get_submodule_repo_for here already.
Signed-off-by: Stefan Beller <sbeller@google.com>
---
submodule.c | 51 +++++++++++++++++++++++++++----------
t/t5526-fetch-submodules.sh | 7 ++++-
2 files changed, 44 insertions(+), 14 deletions(-)
diff --git a/submodule.c b/submodule.c
index dd478ed70b..3f791f2277 100644
--- a/submodule.c
+++ b/submodule.c
@@ -481,6 +481,12 @@ void prepare_submodule_repo_env(struct argv_array *out)
DEFAULT_GIT_DIR_ENVIRONMENT);
}
+static void prepare_submodule_repo_env_in_gitdir(struct argv_array *out)
+{
+ prepare_submodule_repo_env_no_git_dir(out);
+ argv_array_pushf(out, "%s=.", GIT_DIR_ENVIRONMENT);
+}
+
/* Helper function to display the submodule header line prior to the full
* summary output. If it can locate the submodule objects directory it will
* attempt to lookup both the left and right commits and put them into the
@@ -1227,6 +1233,29 @@ static int get_fetch_recurse_config(const struct submodule *submodule,
return spf->default_option;
}
+static struct repository *get_submodule_repo_for(struct repository *r,
+ const struct submodule *sub)
+{
+ struct repository *ret = xmalloc(sizeof(*ret));
+
+ if (repo_submodule_init(ret, r, sub)) {
+ /*
+ * No entry in .gitmodules? Technically not a submodule,
+ * but historically we supported repositories that happen to be
+ * in-place where a gitlink is. Keep supporting them.
+ */
+ struct strbuf gitdir = STRBUF_INIT;
+ strbuf_repo_worktree_path(&gitdir, r, "%s/.git", sub->path);
+ if (repo_init(ret, gitdir.buf, NULL)) {
+ strbuf_release(&gitdir);
+ return NULL;
+ }
+ strbuf_release(&gitdir);
+ }
+
+ return ret;
+}
+
static int get_next_submodule(struct child_process *cp,
struct strbuf *err, void *data, void **task_cb)
{
@@ -1234,12 +1263,11 @@ static int get_next_submodule(struct child_process *cp,
struct submodule_parallel_fetch *spf = data;
for (; spf->count < spf->r->index->cache_nr; spf->count++) {
- struct strbuf submodule_path = STRBUF_INIT;
- struct strbuf submodule_git_dir = STRBUF_INIT;
struct strbuf submodule_prefix = STRBUF_INIT;
const struct cache_entry *ce = spf->r->index->cache[spf->count];
- const char *git_dir, *default_argv;
+ const char *default_argv;
const struct submodule *submodule;
+ struct repository *repo;
struct submodule default_submodule = SUBMODULE_INIT;
if (!S_ISGITLINK(ce->ce_mode))
@@ -1274,16 +1302,12 @@ static int get_next_submodule(struct child_process *cp,
continue;
}
- strbuf_repo_worktree_path(&submodule_path, spf->r, "%s", ce->name);
- strbuf_addf(&submodule_git_dir, "%s/.git", submodule_path.buf);
strbuf_addf(&submodule_prefix, "%s%s/", spf->prefix, ce->name);
- git_dir = read_gitfile(submodule_git_dir.buf);
- if (!git_dir)
- git_dir = submodule_git_dir.buf;
- if (is_directory(git_dir)) {
+ repo = get_submodule_repo_for(spf->r, submodule);
+ if (repo) {
child_process_init(cp);
- cp->dir = strbuf_detach(&submodule_path, NULL);
- prepare_submodule_repo_env(&cp->env_array);
+ prepare_submodule_repo_env_in_gitdir(&cp->env_array);
+ cp->dir = xstrdup(repo->gitdir);
cp->git_cmd = 1;
if (!spf->quiet)
strbuf_addf(err, "Fetching submodule %s%s\n",
@@ -1293,10 +1317,11 @@ static int get_next_submodule(struct child_process *cp,
argv_array_push(&cp->args, default_argv);
argv_array_push(&cp->args, "--submodule-prefix");
argv_array_push(&cp->args, submodule_prefix.buf);
+
+ repo_clear(repo);
+ free(repo);
ret = 1;
}
- strbuf_release(&submodule_path);
- strbuf_release(&submodule_git_dir);
strbuf_release(&submodule_prefix);
if (ret) {
spf->count++;
diff --git a/t/t5526-fetch-submodules.sh b/t/t5526-fetch-submodules.sh
index 6c2f9b2ba2..42692219a1 100755
--- a/t/t5526-fetch-submodules.sh
+++ b/t/t5526-fetch-submodules.sh
@@ -566,7 +566,12 @@ test_expect_success 'fetching submodule into a broken repository' '
test_must_fail git -C dst status &&
test_must_fail git -C dst diff &&
- test_must_fail git -C dst fetch --recurse-submodules
+
+ # git-fetch cannot find the git directory of the submodule,
+ # so it will do nothing, successfully, as it cannot distinguish between
+ # this broken submodule and a submodule that was just set active but
+ # not cloned yet
+ git -C dst fetch --recurse-submodules
'
test_expect_success "fetch new commits when submodule got renamed" '
--
2.19.0
^ permalink raw reply related [relevance 26%]
* [PATCH v5 6/9] repository: repo_submodule_init to take a submodule struct
2018-10-10 21:49 10% [PATCH v5 0/9] fetch: make sure submodule oids are fetched Stefan Beller
` (3 preceding siblings ...)
2018-10-10 21:50 17% ` [PATCH v5 5/9] submodule.c: do not copy around submodule list Stefan Beller
@ 2018-10-10 21:50 24% ` Stefan Beller
2018-10-10 21:50 26% ` [PATCH v5 7/9] submodule: fetch in submodules git directory instead of in worktree Stefan Beller
` (2 subsequent siblings)
7 siblings, 0 replies; 200+ results
From: Stefan Beller @ 2018-10-10 21:50 UTC (permalink / raw)
To: git; +Cc: Stefan Beller
When constructing a struct repository for a submodule for some revision
of the superproject where the submodule is not contained in the index,
it may not be present in the working tree currently either. In that
siutation giving a 'path' argument is not useful. Upgrade the
repo_submodule_init function to take a struct submodule instead.
While we are at it, overhaul the repo_submodule_init function by renaming
the submodule repository struct, which is to be initialized, to a name
that is not confused with the struct submodule as easily.
Also move its documentation into the header file.
Signed-off-by: Stefan Beller <sbeller@google.com>
---
builtin/grep.c | 17 ++++++++++-------
builtin/ls-files.c | 12 +++++++-----
builtin/submodule--helper.c | 2 +-
repository.c | 27 ++++++++++-----------------
repository.h | 11 +++++++++--
5 files changed, 37 insertions(+), 32 deletions(-)
diff --git a/builtin/grep.c b/builtin/grep.c
index 601f801158..81c53c862b 100644
--- a/builtin/grep.c
+++ b/builtin/grep.c
@@ -418,16 +418,19 @@ static int grep_submodule(struct grep_opt *opt, struct repository *superproject,
const struct object_id *oid,
const char *filename, const char *path)
{
- struct repository submodule;
+ struct repository subrepo;
+ const struct submodule *sub = submodule_from_path(superproject,
+ &null_oid, path);
+
int hit;
if (!is_submodule_active(superproject, path))
return 0;
- if (repo_submodule_init(&submodule, superproject, path))
+ if (repo_submodule_init(&subrepo, superproject, sub))
return 0;
- repo_read_gitmodules(&submodule);
+ repo_read_gitmodules(&subrepo);
/*
* NEEDSWORK: This adds the submodule's object directory to the list of
@@ -440,7 +443,7 @@ static int grep_submodule(struct grep_opt *opt, struct repository *superproject,
* object.
*/
grep_read_lock();
- add_to_alternates_memory(submodule.objects->objectdir);
+ add_to_alternates_memory(subrepo.objects->objectdir);
grep_read_unlock();
if (oid) {
@@ -465,14 +468,14 @@ static int grep_submodule(struct grep_opt *opt, struct repository *superproject,
init_tree_desc(&tree, data, size);
hit = grep_tree(opt, pathspec, &tree, &base, base.len,
- object->type == OBJ_COMMIT, &submodule);
+ object->type == OBJ_COMMIT, &subrepo);
strbuf_release(&base);
free(data);
} else {
- hit = grep_cache(opt, &submodule, pathspec, 1);
+ hit = grep_cache(opt, &subrepo, pathspec, 1);
}
- repo_clear(&submodule);
+ repo_clear(&subrepo);
return hit;
}
diff --git a/builtin/ls-files.c b/builtin/ls-files.c
index 7f9919a362..4d1649c1b3 100644
--- a/builtin/ls-files.c
+++ b/builtin/ls-files.c
@@ -206,17 +206,19 @@ static void show_files(struct repository *repo, struct dir_struct *dir);
static void show_submodule(struct repository *superproject,
struct dir_struct *dir, const char *path)
{
- struct repository submodule;
+ struct repository subrepo;
+ const struct submodule *sub = submodule_from_path(superproject,
+ &null_oid, path);
- if (repo_submodule_init(&submodule, superproject, path))
+ if (repo_submodule_init(&subrepo, superproject, sub))
return;
- if (repo_read_index(&submodule) < 0)
+ if (repo_read_index(&subrepo) < 0)
die("index file corrupt");
- show_files(&submodule, dir);
+ show_files(&subrepo, dir);
- repo_clear(&submodule);
+ repo_clear(&subrepo);
}
static void show_ce(struct repository *repo, struct dir_struct *dir,
diff --git a/builtin/submodule--helper.c b/builtin/submodule--helper.c
index 247881189f..8214e77688 100644
--- a/builtin/submodule--helper.c
+++ b/builtin/submodule--helper.c
@@ -2038,7 +2038,7 @@ static int ensure_core_worktree(int argc, const char **argv, const char *prefix)
if (!sub)
BUG("We could get the submodule handle before?");
- if (repo_submodule_init(&subrepo, the_repository, path))
+ if (repo_submodule_init(&subrepo, the_repository, sub))
die(_("could not get a repository handle for submodule '%s'"), path);
if (!repo_config_get_string(&subrepo, "core.worktree", &cw)) {
diff --git a/repository.c b/repository.c
index 5dd1486718..aabe64ee5d 100644
--- a/repository.c
+++ b/repository.c
@@ -166,30 +166,23 @@ int repo_init(struct repository *repo,
return -1;
}
-/*
- * Initialize 'submodule' as the submodule given by 'path' in parent repository
- * 'superproject'.
- * Return 0 upon success and a non-zero value upon failure.
- */
-int repo_submodule_init(struct repository *submodule,
+int repo_submodule_init(struct repository *subrepo,
struct repository *superproject,
- const char *path)
+ const struct submodule *sub)
{
- const struct submodule *sub;
struct strbuf gitdir = STRBUF_INIT;
struct strbuf worktree = STRBUF_INIT;
int ret = 0;
- sub = submodule_from_path(superproject, &null_oid, path);
if (!sub) {
ret = -1;
goto out;
}
- strbuf_repo_worktree_path(&gitdir, superproject, "%s/.git", path);
- strbuf_repo_worktree_path(&worktree, superproject, "%s", path);
+ strbuf_repo_worktree_path(&gitdir, superproject, "%s/.git", sub->path);
+ strbuf_repo_worktree_path(&worktree, superproject, "%s", sub->path);
- if (repo_init(submodule, gitdir.buf, worktree.buf)) {
+ if (repo_init(subrepo, gitdir.buf, worktree.buf)) {
/*
* If initilization fails then it may be due to the submodule
* not being populated in the superproject's worktree. Instead
@@ -201,16 +194,16 @@ int repo_submodule_init(struct repository *submodule,
strbuf_repo_git_path(&gitdir, superproject,
"modules/%s", sub->name);
- if (repo_init(submodule, gitdir.buf, NULL)) {
+ if (repo_init(subrepo, gitdir.buf, NULL)) {
ret = -1;
goto out;
}
}
- submodule->submodule_prefix = xstrfmt("%s%s/",
- superproject->submodule_prefix ?
- superproject->submodule_prefix :
- "", path);
+ subrepo->submodule_prefix = xstrfmt("%s%s/",
+ superproject->submodule_prefix ?
+ superproject->submodule_prefix :
+ "", sub->path);
out:
strbuf_release(&gitdir);
diff --git a/repository.h b/repository.h
index 9f16c42c1e..a9c7a5baa5 100644
--- a/repository.h
+++ b/repository.h
@@ -116,9 +116,16 @@ void repo_set_worktree(struct repository *repo, const char *path);
void repo_set_hash_algo(struct repository *repo, int algo);
void initialize_the_repository(void);
int repo_init(struct repository *r, const char *gitdir, const char *worktree);
-int repo_submodule_init(struct repository *submodule,
+
+/*
+ * Initialize the repository 'subrepo' as the submodule given by the
+ * struct submodule 'sub' in parent repository 'superproject'.
+ * Return 0 upon success and a non-zero value upon failure.
+ */
+struct submodule;
+int repo_submodule_init(struct repository *subrepo,
struct repository *superproject,
- const char *path);
+ const struct submodule *sub);
void repo_clear(struct repository *repo);
/*
--
2.19.0
^ permalink raw reply related [relevance 24%]
* [PATCH v5 5/9] submodule.c: do not copy around submodule list
2018-10-10 21:49 10% [PATCH v5 0/9] fetch: make sure submodule oids are fetched Stefan Beller
` (2 preceding siblings ...)
2018-10-10 21:50 16% ` [PATCH v5 4/9] submodule.c: move global changed_submodule_names into fetch submodule struct Stefan Beller
@ 2018-10-10 21:50 17% ` Stefan Beller
2018-10-10 21:50 24% ` [PATCH v5 6/9] repository: repo_submodule_init to take a submodule struct Stefan Beller
` (3 subsequent siblings)
7 siblings, 0 replies; 200+ results
From: Stefan Beller @ 2018-10-10 21:50 UTC (permalink / raw)
To: git; +Cc: Stefan Beller
'calculate_changed_submodule_paths' uses a local list to compute the
changed submodules, and then produces the result by copying appropriate
items into the result list.
Instead use the result list directly and prune items afterwards
using string_list_remove_empty_items.
By doing so we'll have access to the util pointer for longer that
contains the commits that we need to fetch, which will be
useful in a later patch.
Signed-off-by: Stefan Beller <sbeller@google.com>
---
submodule.c | 19 ++++++++++---------
1 file changed, 10 insertions(+), 9 deletions(-)
diff --git a/submodule.c b/submodule.c
index 17103379ba..dd478ed70b 100644
--- a/submodule.c
+++ b/submodule.c
@@ -1128,8 +1128,7 @@ static void calculate_changed_submodule_paths(
struct submodule_parallel_fetch *spf)
{
struct argv_array argv = ARGV_ARRAY_INIT;
- struct string_list changed_submodules = STRING_LIST_INIT_DUP;
- const struct string_list_item *name;
+ struct string_list_item *name;
/* No need to check if there are no submodules configured */
if (!submodule_from_path(the_repository, NULL, NULL))
@@ -1146,9 +1145,9 @@ static void calculate_changed_submodule_paths(
* Collect all submodules (whether checked out or not) for which new
* commits have been recorded upstream in "changed_submodule_names".
*/
- collect_changed_submodules(&changed_submodules, &argv);
+ collect_changed_submodules(&spf->changed_submodule_names, &argv);
- for_each_string_list_item(name, &changed_submodules) {
+ for_each_string_list_item(name, &spf->changed_submodule_names) {
struct oid_array *commits = name->util;
const struct submodule *submodule;
const char *path = NULL;
@@ -1162,12 +1161,14 @@ static void calculate_changed_submodule_paths(
if (!path)
continue;
- if (!submodule_has_commits(path, commits))
- string_list_append(&spf->changed_submodule_names,
- name->string);
+ if (submodule_has_commits(path, commits)) {
+ oid_array_clear(commits);
+ *name->string = '\0';
+ }
}
- free_submodules_oids(&changed_submodules);
+ string_list_remove_empty_items(&spf->changed_submodule_names, 1);
+
argv_array_clear(&argv);
oid_array_clear(&ref_tips_before_fetch);
oid_array_clear(&ref_tips_after_fetch);
@@ -1363,7 +1364,7 @@ int fetch_populated_submodules(struct repository *r,
argv_array_clear(&spf.args);
out:
- string_list_clear(&spf.changed_submodule_names, 1);
+ free_submodules_oids(&spf.changed_submodule_names);
return spf.result;
}
--
2.19.0
^ permalink raw reply related [relevance 17%]
* [PATCH v5 4/9] submodule.c: move global changed_submodule_names into fetch submodule struct
2018-10-10 21:49 10% [PATCH v5 0/9] fetch: make sure submodule oids are fetched Stefan Beller
2018-10-10 21:50 24% ` [PATCH v5 2/9] submodule.c: fix indentation Stefan Beller
2018-10-10 21:50 17% ` [PATCH v5 3/9] submodule.c: sort changed_submodule_names before searching it Stefan Beller
@ 2018-10-10 21:50 16% ` Stefan Beller
2018-10-10 21:50 17% ` [PATCH v5 5/9] submodule.c: do not copy around submodule list Stefan Beller
` (4 subsequent siblings)
7 siblings, 0 replies; 200+ results
From: Stefan Beller @ 2018-10-10 21:50 UTC (permalink / raw)
To: git; +Cc: Stefan Beller
The `changed_submodule_names` are only used for fetching, so let's make it
part of the struct that is passed around for fetching submodules.
Signed-off-by: Stefan Beller <sbeller@google.com>
---
submodule.c | 42 +++++++++++++++++++++++-------------------
1 file changed, 23 insertions(+), 19 deletions(-)
diff --git a/submodule.c b/submodule.c
index 22c64bd855..17103379ba 100644
--- a/submodule.c
+++ b/submodule.c
@@ -25,7 +25,7 @@
#include "commit-reach.h"
static int config_update_recurse_submodules = RECURSE_SUBMODULES_OFF;
-static struct string_list changed_submodule_names = STRING_LIST_INIT_DUP;
+
static int initialized_fetch_ref_tips;
static struct oid_array ref_tips_before_fetch;
static struct oid_array ref_tips_after_fetch;
@@ -1110,7 +1110,22 @@ void check_for_new_submodule_commits(struct object_id *oid)
oid_array_append(&ref_tips_after_fetch, oid);
}
-static void calculate_changed_submodule_paths(void)
+struct submodule_parallel_fetch {
+ int count;
+ struct argv_array args;
+ struct repository *r;
+ const char *prefix;
+ int command_line_option;
+ int default_option;
+ int quiet;
+ int result;
+
+ struct string_list changed_submodule_names;
+};
+#define SPF_INIT {0, ARGV_ARRAY_INIT, NULL, NULL, 0, 0, 0, 0, STRING_LIST_INIT_DUP }
+
+static void calculate_changed_submodule_paths(
+ struct submodule_parallel_fetch *spf)
{
struct argv_array argv = ARGV_ARRAY_INIT;
struct string_list changed_submodules = STRING_LIST_INIT_DUP;
@@ -1148,7 +1163,8 @@ static void calculate_changed_submodule_paths(void)
continue;
if (!submodule_has_commits(path, commits))
- string_list_append(&changed_submodule_names, name->string);
+ string_list_append(&spf->changed_submodule_names,
+ name->string);
}
free_submodules_oids(&changed_submodules);
@@ -1185,18 +1201,6 @@ int submodule_touches_in_range(struct object_id *excl_oid,
return ret;
}
-struct submodule_parallel_fetch {
- int count;
- struct argv_array args;
- struct repository *r;
- const char *prefix;
- int command_line_option;
- int default_option;
- int quiet;
- int result;
-};
-#define SPF_INIT {0, ARGV_ARRAY_INIT, NULL, NULL, 0, 0, 0, 0}
-
static int get_fetch_recurse_config(const struct submodule *submodule,
struct submodule_parallel_fetch *spf)
{
@@ -1257,7 +1261,7 @@ static int get_next_submodule(struct child_process *cp,
case RECURSE_SUBMODULES_ON_DEMAND:
if (!submodule ||
!string_list_lookup(
- &changed_submodule_names,
+ &spf->changed_submodule_names,
submodule->name))
continue;
default_argv = "on-demand";
@@ -1349,8 +1353,8 @@ int fetch_populated_submodules(struct repository *r,
argv_array_push(&spf.args, "--recurse-submodules-default");
/* default value, "--submodule-prefix" and its value are added later */
- calculate_changed_submodule_paths();
- string_list_sort(&changed_submodule_names);
+ calculate_changed_submodule_paths(&spf);
+ string_list_sort(&spf.changed_submodule_names);
run_processes_parallel(max_parallel_jobs,
get_next_submodule,
fetch_start_failure,
@@ -1359,7 +1363,7 @@ int fetch_populated_submodules(struct repository *r,
argv_array_clear(&spf.args);
out:
- string_list_clear(&changed_submodule_names, 1);
+ string_list_clear(&spf.changed_submodule_names, 1);
return spf.result;
}
--
2.19.0
^ permalink raw reply related [relevance 16%]
* [PATCH v5 3/9] submodule.c: sort changed_submodule_names before searching it
2018-10-10 21:49 10% [PATCH v5 0/9] fetch: make sure submodule oids are fetched Stefan Beller
2018-10-10 21:50 24% ` [PATCH v5 2/9] submodule.c: fix indentation Stefan Beller
@ 2018-10-10 21:50 17% ` Stefan Beller
2018-10-10 21:50 16% ` [PATCH v5 4/9] submodule.c: move global changed_submodule_names into fetch submodule struct Stefan Beller
` (5 subsequent siblings)
7 siblings, 0 replies; 200+ results
From: Stefan Beller @ 2018-10-10 21:50 UTC (permalink / raw)
To: git; +Cc: Stefan Beller
We can string_list_insert() to maintain sorted-ness of the
list as we find new items, or we can string_list_append() to
build an unsorted list and sort it at the end just once.
As we do not rely on the sortedness while building the
list, we pick the "append and sort at the end" as it
has better worst case execution times.
Signed-off-by: Stefan Beller <sbeller@google.com>
---
submodule.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/submodule.c b/submodule.c
index 0de9e2800a..22c64bd855 100644
--- a/submodule.c
+++ b/submodule.c
@@ -1256,7 +1256,7 @@ static int get_next_submodule(struct child_process *cp,
case RECURSE_SUBMODULES_DEFAULT:
case RECURSE_SUBMODULES_ON_DEMAND:
if (!submodule ||
- !unsorted_string_list_lookup(
+ !string_list_lookup(
&changed_submodule_names,
submodule->name))
continue;
@@ -1350,6 +1350,7 @@ int fetch_populated_submodules(struct repository *r,
/* default value, "--submodule-prefix" and its value are added later */
calculate_changed_submodule_paths();
+ string_list_sort(&changed_submodule_names);
run_processes_parallel(max_parallel_jobs,
get_next_submodule,
fetch_start_failure,
--
2.19.0
^ permalink raw reply related [relevance 17%]
* [PATCH v5 2/9] submodule.c: fix indentation
2018-10-10 21:49 10% [PATCH v5 0/9] fetch: make sure submodule oids are fetched Stefan Beller
@ 2018-10-10 21:50 24% ` Stefan Beller
2018-10-10 21:50 17% ` [PATCH v5 3/9] submodule.c: sort changed_submodule_names before searching it Stefan Beller
` (6 subsequent siblings)
7 siblings, 0 replies; 200+ results
From: Stefan Beller @ 2018-10-10 21:50 UTC (permalink / raw)
To: git; +Cc: Stefan Beller
The submodule subsystem is really bad at staying within 80 characters.
Fix it while we are here.
Signed-off-by: Stefan Beller <sbeller@google.com>
---
submodule.c | 9 ++++++---
1 file changed, 6 insertions(+), 3 deletions(-)
diff --git a/submodule.c b/submodule.c
index b53cb6e9c4..0de9e2800a 100644
--- a/submodule.c
+++ b/submodule.c
@@ -1244,7 +1244,8 @@ static int get_next_submodule(struct child_process *cp,
if (!submodule) {
const char *name = default_name_or_path(ce->name);
if (name) {
- default_submodule.path = default_submodule.name = name;
+ default_submodule.path = name;
+ default_submodule.name = name;
submodule = &default_submodule;
}
}
@@ -1254,8 +1255,10 @@ static int get_next_submodule(struct child_process *cp,
default:
case RECURSE_SUBMODULES_DEFAULT:
case RECURSE_SUBMODULES_ON_DEMAND:
- if (!submodule || !unsorted_string_list_lookup(&changed_submodule_names,
- submodule->name))
+ if (!submodule ||
+ !unsorted_string_list_lookup(
+ &changed_submodule_names,
+ submodule->name))
continue;
default_argv = "on-demand";
break;
--
2.19.0
^ permalink raw reply related [relevance 24%]
* [PATCH v5 0/9] fetch: make sure submodule oids are fetched
@ 2018-10-10 21:49 10% Stefan Beller
2018-10-10 21:50 24% ` [PATCH v5 2/9] submodule.c: fix indentation Stefan Beller
` (7 more replies)
0 siblings, 8 replies; 200+ results
From: Stefan Beller @ 2018-10-10 21:49 UTC (permalink / raw)
To: git; +Cc: Stefan Beller
This is nearly the same as sent in [1], with one commit message fixed.
[1] https://public-inbox.org/git/20180925194755.105578-1-sbeller@google.com/
and replaces sb/submodule-recursive-fetch-gets-the-tip.
Thanks,
Stefan
Stefan Beller (9):
sha1-array: provide oid_array_filter
submodule.c: fix indentation
submodule.c: sort changed_submodule_names before searching it
submodule.c: move global changed_submodule_names into fetch submodule
struct
submodule.c: do not copy around submodule list
repository: repo_submodule_init to take a submodule struct
submodule: fetch in submodules git directory instead of in worktree
fetch: retry fetching submodules if needed objects were not fetched
builtin/fetch: check for submodule updates for non branch fetches
Documentation/technical/api-oid-array.txt | 5 +
builtin/fetch.c | 14 +-
builtin/grep.c | 17 +-
builtin/ls-files.c | 12 +-
builtin/submodule--helper.c | 2 +-
repository.c | 27 +--
repository.h | 11 +-
sha1-array.c | 17 ++
sha1-array.h | 3 +
submodule.c | 275 +++++++++++++++++-----
t/t5526-fetch-submodules.sh | 23 +-
11 files changed, 311 insertions(+), 95 deletions(-)
--
2.19.0
^ permalink raw reply [relevance 10%]
* Re: [PATCH v6 09/10] submodule: support reading .gitmodules when it's not in the working tree
2018-10-08 22:19 4% ` Stefan Beller
@ 2018-10-10 18:56 7% ` Antonio Ospite
2018-10-10 22:55 7% ` Stefan Beller
0 siblings, 1 reply; 200+ results
From: Antonio Ospite @ 2018-10-10 18:56 UTC (permalink / raw)
To: Stefan Beller
Cc: Junio C Hamano, git, Jonathan Nieder, Jeff King,
SZEDER Gábor
On Mon, 8 Oct 2018 15:19:00 -0700
Stefan Beller <sbeller@google.com> wrote:
> > +test_expect_success 'not writing gitmodules config file when it is not checked out' '
> > + test_must_fail git -C super submodule--helper config submodule.submodule.url newurl
>
> This only checks the exit code, do we also want to check for
>
> test_path_is_missing .gitmodules ?
>
OK, I agree, let's re-check also *after* we tried and failed to set
a config value, just to be sure that the code does not get accidentally
changed in the future to create the file. I'll add the check.
> > +test_expect_success 'initialising submodule when the gitmodules config is not checked out' '
> > + git -C super submodule init
> > +'
> > +
> > +test_expect_success 'showing submodule summary when the gitmodules config is not checked out' '
> > + git -C super submodule summary
> > +'
>
> Same for these, is the exit code enough, or do we want to look at
> specific things?
>
Except for the "summary" test which was not even exercising the
config_from_gitmodule path, checking exist status should be sufficient
to verify that "submodule--helper config" does not fail, but we can
surely do better.
I will add checks to confirm that not only the commands exited without
errors but they also achieved the desired effect, to validate the actual
high-level use case advertised by the test file. This should be more
future-proof.
And I think I'll merge the summary and the update tests.
> > +
> > +test_expect_success 'updating submodule when the gitmodules config is not checked out' '
> > + (cd submodule &&
> > + echo file2 >file2 &&
> > + git add file2 &&
> > + git commit -m "add file2 to submodule"
> > + ) &&
> > + git -C super submodule update
>
> git status would want to be clean afterwards?
Mmh, this should have been "submodule update --remote" in the first
place to have any effect, I'll take the chance and rewrite this test in
a different way and also check the effect of the update operation, and
the repository status.
I'll be something like this:
ORIG_SUBMODULE=$(git -C submodule rev-parse HEAD)
ORIG_UPSTREAM=$(git -C upstream rev-parse HEAD)
ORIG_SUPER=$(git -C super rev-parse HEAD)
test_expect_success 're-updating submodule when the gitmodules config is not checked out' '
test_when_finished "git -C submodule reset --hard $ORIG_SUBMODULE;
git -C upstream reset --hard $ORIG_UPSTREAM;
git -C super reset --hard $ORIG_SUPER;
git -C upstream submodule update --remote;
git -C super pull;
git -C super submodule update --remote" &&
(cd submodule &&
echo file2 >file2 &&
git add file2 &&
test_tick &&
git commit -m "add file2 to submodule"
) &&
(cd upstream &&
git submodule update --remote &&
git add submodule &&
test_tick &&
git commit -m "Update submodule"
) &&
git -C super pull &&
# The --for-status options reads the gitmdoules config
git -C super submodule summary --for-status >actual &&
cat >expect <<-\EOF &&
* submodule 951c301...a939200 (1):
< add file2 to submodule
EOF
test_cmp expect actual &&
# Test that the update actually succeeds
test_path_is_missing super/submodule/file2 &&
git -C super submodule update &&
test_cmp submodule/file2 super/submodule/file2 &&
git -C super status --short >output &&
test_must_be_empty output
'
Maybe a little overkill?
The "upstream" repo will be added in test 1 to better clarify the roles
of the involved repositories.
The commit ids should be stable because of test_tick, shouldn't they?
Thanks for the comments, they helped improving the quality of the tests
once again.
I'll wait a few days before sending a v7, hopefully someone will find
time to take another look at patch 9 and comment also on patch 10, and
give an opinion on the "mergeability" status of the whole patchset.
Ciao ciao,
Antonio
--
Antonio Ospite
https://ao2.it
https://twitter.com/ao2it
A: Because it messes up the order in which people normally read text.
See http://en.wikipedia.org/wiki/Posting_style
Q: Why is top-posting such a bad thing?
^ permalink raw reply [relevance 7%]
* Re: What's cooking in git.git (Oct 2018, #01; Wed, 10)
@ 2018-10-10 18:55 4% ` Stefan Beller
2018-10-11 2:00 2% ` Junio C Hamano
0 siblings, 1 reply; 200+ results
From: Stefan Beller @ 2018-10-10 18:55 UTC (permalink / raw)
To: Junio C Hamano; +Cc: git
> * pw/diff-color-moved-ws-fix (2018-10-04) 5 commits
> - diff --color-moved: fix a memory leak
> - diff --color-moved-ws: fix another memory leak
> - diff --color-moved-ws: fix a memory leak
> - diff --color-moved-ws: fix out of bounds string access
> - diff --color-moved-ws: fix double free crash
>
> Various fixes to "diff --color-moved-ws".
>
> What's the status of this topic?
Per [1] ("The whole series is
Reviewed-by: Stefan Beller <sbeller@google.com>"),
I would suggest merging to 'next'.
[1] https://public-inbox.org/git/CAGZ79kbamUK=d+-ejy9vopDiVZF7OVOngz1Zx9y04VR3HnmoXg@mail.gmail.com/
> * sb/strbuf-h-update (2018-09-29) 1 commit
> - strbuf.h: format according to coding guidelines
>
> Code clean-up to serve as a BCP example.
>
> What's the status of this one after the discussion thread stopped here?
> cf. <CAGZ79kbV6QjsFKcD2uG_P9j1AvzSNQSi-_jXGQ9w0YU9fjhEGg@mail.gmail.com>
I was waiting for more discussion and stricter guidelines,
which never happened.
The only controversial issue about this patch is whether we want
to name all parameters or only when we feel like it.
Peff did not seem to care about this particular detail
https://public-inbox.org/git/20180929073827.GD2174@sigill.intra.peff.net/
You suggested to embrace it further and use caps for the parameter
names in the docs comment.
https://public-inbox.org/git/xmqq8t3lb8uu.fsf@gitster-ct.c.googlers.com/
The patch as-is just adds names everywhere.
I'd be happy to resend with either
(a) not enforcing names everywhere, but only as needed or
(b) having names everywhere, capitalizing them NAMES in
the doc comment.
I am tempted to ask for
(c) take as-is, defer the rewording of doc strings for a follow up patch.
> * sb/grep-submodule-cleanup (2018-10-10) 1 commit
> - builtin/grep.c: remove superfluous submodule code
>
> Code clean-up.
>
> cf. <20181010001037.74709-1-jonathantanmy@google.com>
Will resend.
> * bw/submodule-name-to-dir (2018-08-10) 2 commits
> - submodule: munge paths to submodule git directories
> - submodule: create helper to build paths to submodule gitdirs
>
> In modern repository layout, the real body of a cloned submodule
> repository is held in .git/modules/ of the superproject, indexed by
> the submodule name. URLencode the submodule name before computing
> the name of the directory to make sure they form a flat namespace.
>
> Kicked back to 'pu', expecting further work on the topic.
> cf. <CAGZ79kYnbjaPoWdda0SM_-_X77mVyYC7JO61OV8nm2yj3Q1OvQ@mail.gmail.com>
Thanks.
>
> * sb/submodule-move-head-with-corruption (2018-08-28) 2 commits
> - submodule.c: warn about missing submodule git directories
> - t2013: add test for missing but active submodule
>
> Will discard and wait for a cleaned-up rewrite.
> cf. <20180907195349.GA103699@aiede.svl.corp.google.com>
Yeah I think discarding this is the right move.
> * sb/submodule-recursive-fetch-gets-the-tip (2018-09-12) 9 commits
> - builtin/fetch: check for submodule updates for non branch fetches
> - fetch: retry fetching submodules if sha1 were not fetched
> - submodule: fetch in submodules git directory instead of in worktree
> - submodule.c: do not copy around submodule list
> - submodule: move global changed_submodule_names into fetch submodule struct
> - submodule.c: sort changed_submodule_names before searching it
> - submodule.c: fix indentation
> - sha1-array: provide oid_array_filter
> - string-list: add string_list_{pop, last} functions
>
> "git fetch --recurse-submodules" may not fetch the necessary commit
> that is bound to the superproject, which is getting corrected.
>
> Expecting a reroll.
> cf. <b16af8c0-0435-0de4-ed6c-53888d6190af@ramsayjones.plus.com>
is fixed in
https://public-inbox.org/git/20180917213559.126404-7-sbeller@google.com/
> cf. <CAGZ79kbavjVbTqXsmtjW6=jhkq47_p3mc6=92xOp4_mfhqDtvw@mail.gmail.com>
That is fixed locally
> cf. <CAGZ79kZKKf9N8yx9EuCRZhrZS_mA2218PouEG7aHDhK2bJGEdA@mail.gmail.com>
That has been addressed via
https://public-inbox.org/git/20180925194755.105578-1-sbeller@google.com/
Will resend after a local review.
> * pk/rebase-in-c-6-final (2018-10-09) 1 commit
> - rebase: default to using the builtin rebase
> (this branch uses ag/rebase-i-in-c, js/rebase-in-c-5.5-work-with-rebase-i-in-c, pk/rebase-in-c, pk/rebase-in-c-2-basic, pk/rebase-in-c-3-acts, pk/rebase-in-c-4-opts and pk/rebase-in-c-5-test; is tangled with ag/sequencer-reduce-rewriting-todo, jc/rebase-in-c-5-test-typofix and js/rebase-i-break.)
>
> The final step of rewriting "rebase -i" in C.
>
> Undecided.
> I've been using this (i.e. the whole "rebase -i" and "rebase"
> rewritten in C) in my personal build, and I also know users on
> Windows port have been using it with the last feature release. I
> am tempted to merge the whole thing to 'next' soonish.
>
> Opinions? It's the last chance to remove any existing and avoid
> any future "oops, that was wrong, and here is a fix-up"
> embarrassment in these topics.
Yes, please merge to next.
Stefan
^ permalink raw reply [relevance 4%]
* [PATCH] builtin/grep.c: remote superflous submodule code
@ 2018-10-09 18:35 16% Stefan Beller
0 siblings, 1 reply; 200+ results
From: Stefan Beller @ 2018-10-09 18:35 UTC (permalink / raw)
To: gitster, ao2; +Cc: git, Stefan Beller
In f9ee2fcdfa (grep: recurse in-process using 'struct repository',
2017-08-02), we introduced a call to repo_read_gitmodules in builtin/grep
to simplify the submodule handling.
After ff6f1f564c4 (submodule-config: lazy-load a repository's .gitmodules
file, 2017-08-03) this is no longer necessary, but that commit did not
cleanup the whole tree, but just show cased the new way how to deal with
submodules in ls-files.
It claimed that grep would still need some explicit handling, but that is
not the call to repo_read_gitmodules (applying this patch on top of
ff6f1f564c4 still keep the test suite happy, specifically
t7814-grep-recurse-submodules, which contains a test
"grep history with moved submoules")
The special handling is the call to gitmodules_config_oid which was added
already in 74ed43711f (grep: enable recurse-submodules to work on
<tree> objects, 2016-12-16), but then was still named
gitmodules_config_sha1.
Signed-off-by: Stefan Beller <sbeller@google.com>
Acked-by: Antonio Ospite <ao2@ao2.it>
---
This is a resend of origin/sb/grep-submodule-cleanup,
and I think picking ff6f1f564c4 as the base for the series would
also be appropriate.
Stefan
builtin/grep.c | 2 --
1 file changed, 2 deletions(-)
diff --git a/builtin/grep.c b/builtin/grep.c
index 601f801158..a6272b9c2f 100644
--- a/builtin/grep.c
+++ b/builtin/grep.c
@@ -427,8 +427,6 @@ static int grep_submodule(struct grep_opt *opt, struct repository *superproject,
if (repo_submodule_init(&submodule, superproject, path))
return 0;
- repo_read_gitmodules(&submodule);
-
/*
* NEEDSWORK: This adds the submodule's object directory to the list of
* alternates for the single in-memory object store. This has some bad
--
2.19.0
^ permalink raw reply related [relevance 16%]
* Re: [PATCH 2/3] midx: close multi-pack-index on repack
@ 2018-10-09 18:15 4% ` Stefan Beller
0 siblings, 0 replies; 200+ results
From: Stefan Beller @ 2018-10-09 18:15 UTC (permalink / raw)
To: Derrick Stolee; +Cc: Junio C Hamano, gitgitgadget, git, Derrick Stolee
On Tue, Oct 9, 2018 at 7:11 AM Derrick Stolee <stolee@gmail.com> wrote:
>
> CC Stefan: Is there a plan to make get_object_directory() take a
> repository parameter?
Not an immediate plan. Regarding the large refactorings I am mostly
waiting for nd/the-index to stabilize (which may be stable enough by
now) before proceeding. Specifically 2abf350385 (revision.c: remove
implicit dependency on the_index, 2018-09-21) is a missing piece that
I want to build on.
My next step is to look into making use of the refactorings that we did
over the last year, so I'd rather look into more submodule code again.
To come back to your question:
git grep get_object_directory |wc -l
30
git grep "objects->objectdir" |wc -l
11
2 out of the 11 matches are from the implementation of get_object_directory
So either we'd actually teach get_object_directory to take a repository
and inspect the 11 occurrences of direct access to objects->objectdir
or we'd view get_object_directory() as a fancy wrapper, that we might want
to drop eventually.
^ permalink raw reply [relevance 4%]
* Re: [PATCH] builtin/grep.c: remote superflous submodule code
2018-10-07 0:33 4% ` Junio C Hamano
@ 2018-10-09 0:14 6% ` Stefan Beller
0 siblings, 0 replies; 200+ results
From: Stefan Beller @ 2018-10-09 0:14 UTC (permalink / raw)
To: Junio C Hamano; +Cc: git, Antonio Ospite
> Well, submodule-config.c has its implementation and another caller,
> which technically is outside submodule.c ;-)
i.e. there is a typo in my commit message.
I meant to say submodule-config.c
> repo_read_gitmodules
> has two more callers in unpack-trees.c these days, so perhaps we can
> do without this last paragraph.
Gah, looking at that code, did we have any reason to rush that series?
c.f. https://public-inbox.org/git/20170811171811.GC1472@book.hvoigt.net/
On Sat, Oct 6, 2018 at 5:33 PM Junio C Hamano <gitster@pobox.com> wrote:
>
> Stefan Beller <sbeller@google.com> writes:
>
> > After ff6f1f564c4 (submodule-config: lazy-load a repository's .gitmodules
> > file, 2017-08-03) this is no longer necessary, but that commit did not
> > cleanup the whole tree, but just show cased the new way how to deal with
> > submodules in ls-files.
>
> The log message of the above one singles out "grep" as a special
> case and explalins why it did not touch, by the way. You probably
> need to explain the reason why "this is no longer necessary" a bit
> better than the above---as it stands, it is "ff6f1f564c4 said it
> still is necessary, I say it is not".
That is true.
For grep, the reason seems to be, that we check is_submodule_active
based off the index, i.e. using
module = submodule_from_path(repo, &null_oid, path);
as the deciding factor, which falls in line with lazyloading.
However the use of the specialized gitmodules_config_oid
in grep is also guarded by the same commit ff6f1f564c4.
Going back to the use case of unpack-trees.c,
I think that we need to keep it there as alternatives
seem to be more complicated.
So I guess I'll just resend with a better commit message.
Thanks,
Stefan
^ permalink raw reply [relevance 6%]
* Re: [PATCH 13/14] submodule: make zero-oid comparison hash function agnostic
@ 2018-10-08 23:10 4% ` Stefan Beller
0 siblings, 0 replies; 200+ results
From: Stefan Beller @ 2018-10-08 23:10 UTC (permalink / raw)
To: brian m. carlson; +Cc: git, Jeff King, Eric Sunshine, Duy Nguyen
On Mon, Oct 8, 2018 at 2:57 PM brian m. carlson
<sandals@crustytoothpaste.net> wrote:
>
> With SHA-256, the length of the all-zeros object ID is longer. Add a
> function to git-submodule.sh to check if a full hex object ID is the
> all-zeros value, and use it to check the output we're parsing from git
> diff-files or diff-index.
>
> Signed-off-by: brian m. carlson <sandals@crustytoothpaste.net>
> ---
Nice!
^ permalink raw reply [relevance 4%]
* Re: [PATCH v6 09/10] submodule: support reading .gitmodules when it's not in the working tree
@ 2018-10-08 22:19 4% ` Stefan Beller
2018-10-10 18:56 7% ` Antonio Ospite
0 siblings, 1 reply; 200+ results
From: Stefan Beller @ 2018-10-08 22:19 UTC (permalink / raw)
To: Antonio Ospite
Cc: Junio C Hamano, git, Brandon Williams, Jonathan Nieder, Jeff King,
SZEDER Gábor
> +test_expect_success 'not writing gitmodules config file when it is not checked out' '
> + test_must_fail git -C super submodule--helper config submodule.submodule.url newurl
This only checks the exit code, do we also want to check for
test_path_is_missing .gitmodules ?
> +test_expect_success 'initialising submodule when the gitmodules config is not checked out' '
> + git -C super submodule init
> +'
> +
> +test_expect_success 'showing submodule summary when the gitmodules config is not checked out' '
> + git -C super submodule summary
> +'
Same for these, is the exit code enough, or do we want to look at
specific things?
> +
> +test_expect_success 'updating submodule when the gitmodules config is not checked out' '
> + (cd submodule &&
> + echo file2 >file2 &&
> + git add file2 &&
> + git commit -m "add file2 to submodule"
> + ) &&
> + git -C super submodule update
git status would want to be clean afterwards?
^ permalink raw reply [relevance 4%]
* Re: [PATCH v6 08/10] submodule: add a helper to check if it is safe to write to .gitmodules
2018-10-06 23:44 4% ` Junio C Hamano
@ 2018-10-08 12:37 2% ` Antonio Ospite
0 siblings, 0 replies; 200+ results
From: Antonio Ospite @ 2018-10-08 12:37 UTC (permalink / raw)
To: Junio C Hamano
Cc: Stefan Beller, git, Brandon Williams, Jonathan Nieder, Jeff King,
SZEDER Gábor
On Sun, 07 Oct 2018 08:44:20 +0900
Junio C Hamano <gitster@pobox.com> wrote:
> Stefan Beller <sbeller@google.com> writes:
>
> >> static int module_config(int argc, const char **argv, const char *prefix)
> >> {
> >> + enum {
> >> + CHECK_WRITEABLE = 1
> >> + } command = 0;
> >
> > Can we have the default named? Then we would only use states
> > from within the enum?
>
> Why? Do we use a half-intelligent "switch () { case ...: ... }"
> checker that would otherwise complain if we handled "case 0" in such
> a switch statement, or something like that?
>
> Are we going to gain a lot more enum members, by the way? At this
> point, this looks more like a
>
> unsigned check_writable = 0; /* default is not to check */
>
> to me.
Hi,
the CHECK_WRITEABLE operation is alternative to the get/set ones, not
an addition, so I can see the rationale behind Stefan's suggestion:
either have named enums members for all command "modes" or for none of
them; however other users of enum+OPT_CMDMODE seems to think like the
enum is for commands passed as *options* and the unnamed default is for
actions derived from *arguments*. I don't have a strong opinion on this
matter, tho, so just tell me what you prefer and I'll do it for v7.
Using an enum was to have a more explicit syntax in case other commands
were going to be added in the future (I imagine "--stage" or
"--list-all" as possible additions), and does not affect the generated
code, so I though it was worth it.
Anyways, these are really details, let's concentrate on patches 9 and
10 which deserve much more attention. :)
Thanks you,
Antonio
--
Antonio Ospite
https://ao2.it
https://twitter.com/ao2it
A: Because it messes up the order in which people normally read text.
See http://en.wikipedia.org/wiki/Posting_style
Q: Why is top-posting such a bad thing?
^ permalink raw reply [relevance 2%]
* Re: [PATCH] builtin/grep.c: remote superflous submodule code
2018-10-05 22:45 16% [PATCH] builtin/grep.c: remote superflous submodule code Stefan Beller
2018-10-06 8:59 4% ` Antonio Ospite
2018-10-07 0:29 7% ` Junio C Hamano
@ 2018-10-07 0:33 4% ` Junio C Hamano
2018-10-09 0:14 6% ` Stefan Beller
2 siblings, 1 reply; 200+ results
From: Junio C Hamano @ 2018-10-07 0:33 UTC (permalink / raw)
To: Stefan Beller; +Cc: git, ao2
Stefan Beller <sbeller@google.com> writes:
> After ff6f1f564c4 (submodule-config: lazy-load a repository's .gitmodules
> file, 2017-08-03) this is no longer necessary, but that commit did not
> cleanup the whole tree, but just show cased the new way how to deal with
> submodules in ls-files.
The log message of the above one singles out "grep" as a special
case and explalins why it did not touch, by the way. You probably
need to explain the reason why "this is no longer necessary" a bit
better than the above---as it stands, it is "ff6f1f564c4 said it
still is necessary, I say it is not".
^ permalink raw reply [relevance 4%]
* Re: [PATCH] builtin/grep.c: remote superflous submodule code
2018-10-05 22:45 16% [PATCH] builtin/grep.c: remote superflous submodule code Stefan Beller
2018-10-06 8:59 4% ` Antonio Ospite
@ 2018-10-07 0:29 7% ` Junio C Hamano
2018-10-07 0:33 4% ` Junio C Hamano
2 siblings, 0 replies; 200+ results
From: Junio C Hamano @ 2018-10-07 0:29 UTC (permalink / raw)
To: Stefan Beller; +Cc: git, ao2
Stefan Beller <sbeller@google.com> writes:
> In f9ee2fcdfa (grep: recurse in-process using 'struct repository',
> 2017-08-02), we introduced a call to repo_read_gitmodules in builtin/grep
> to simplify the submodule handling.
>
> After ff6f1f564c4 (submodule-config: lazy-load a repository's .gitmodules
> file, 2017-08-03) this is no longer necessary, but that commit did not
> cleanup the whole tree, but just show cased the new way how to deal with
> submodules in ls-files.
>
> Cleanup the only remaining caller to repo_read_gitmodules outside of
> submodule.c
Well, submodule-config.c has its implementation and another caller,
which technically is outside submodule.c ;-) repo_read_gitmodules
has two more callers in unpack-trees.c these days, so perhaps we can
do without this last paragraph.
^ permalink raw reply [relevance 7%]
* Re: [PATCH v6 08/10] submodule: add a helper to check if it is safe to write to .gitmodules
2018-10-05 23:50 4% ` Stefan Beller
2018-10-06 9:19 4% ` Antonio Ospite
@ 2018-10-06 23:44 4% ` Junio C Hamano
2018-10-08 12:37 2% ` Antonio Ospite
1 sibling, 1 reply; 200+ results
From: Junio C Hamano @ 2018-10-06 23:44 UTC (permalink / raw)
To: Stefan Beller
Cc: Antonio Ospite, git, Brandon Williams, Jonathan Nieder, Jeff King,
SZEDER Gábor
Stefan Beller <sbeller@google.com> writes:
>> static int module_config(int argc, const char **argv, const char *prefix)
>> {
>> + enum {
>> + CHECK_WRITEABLE = 1
>> + } command = 0;
>
> Can we have the default named? Then we would only use states
> from within the enum?
Why? Do we use a half-intelligent "switch () { case ...: ... }"
checker that would otherwise complain if we handled "case 0" in such
a switch statement, or something like that?
Are we going to gain a lot more enum members, by the way? At this
point, this looks more like a
unsigned check_writable = 0; /* default is not to check */
to me.
^ permalink raw reply [relevance 4%]
* Re: [PATCH v6 08/10] submodule: add a helper to check if it is safe to write to .gitmodules
2018-10-05 23:50 4% ` Stefan Beller
@ 2018-10-06 9:19 4% ` Antonio Ospite
2018-10-06 23:44 4% ` Junio C Hamano
1 sibling, 0 replies; 200+ results
From: Antonio Ospite @ 2018-10-06 9:19 UTC (permalink / raw)
To: Stefan Beller
Cc: Junio C Hamano, git, Brandon Williams, Jonathan Nieder, Jeff King,
SZEDER Gábor
On Fri, 5 Oct 2018 16:50:10 -0700
Stefan Beller <sbeller@google.com> wrote:
> > static int module_config(int argc, const char **argv, const char *prefix)
> > {
> > + enum {
> > + CHECK_WRITEABLE = 1
> > + } command = 0;
>
> Can we have the default named? Then we would only use states
> from within the enum?
The default would mean:
"no command passed as a CLI *option*"
I copied this style from builtin/bisect--helper.c::cmd_bisect__helper()
and it's also used in builtin/rebase--helper.c
I can add a name for the default enum value but I am not sure what it
should be: NO_COMMAND_OPTION, COMMAND_DEFAULT, MODE_DEFAULT?
Ciao,
Antonio
--
Antonio Ospite
https://ao2.it
https://twitter.com/ao2it
A: Because it messes up the order in which people normally read text.
See http://en.wikipedia.org/wiki/Posting_style
Q: Why is top-posting such a bad thing?
^ permalink raw reply [relevance 4%]
* Re: [PATCH] builtin/grep.c: remote superflous submodule code
2018-10-05 22:45 16% [PATCH] builtin/grep.c: remote superflous submodule code Stefan Beller
@ 2018-10-06 8:59 4% ` Antonio Ospite
2018-10-07 0:29 7% ` Junio C Hamano
2018-10-07 0:33 4% ` Junio C Hamano
2 siblings, 0 replies; 200+ results
From: Antonio Ospite @ 2018-10-06 8:59 UTC (permalink / raw)
To: Stefan Beller; +Cc: git
On Fri, 5 Oct 2018 15:45:57 -0700
Stefan Beller <sbeller@google.com> wrote:
> In f9ee2fcdfa (grep: recurse in-process using 'struct repository',
> 2017-08-02), we introduced a call to repo_read_gitmodules in builtin/grep
> to simplify the submodule handling.
>
> After ff6f1f564c4 (submodule-config: lazy-load a repository's .gitmodules
> file, 2017-08-03) this is no longer necessary, but that commit did not
> cleanup the whole tree, but just show cased the new way how to deal with
> submodules in ls-files.
>
> Cleanup the only remaining caller to repo_read_gitmodules outside of
> submodule.c
>
> Signed-off-by: Stefan Beller <sbeller@google.com>
Not sure if I am entitled to formally ack it, but:
Acked-by: Antonio Ospite <ao2@ao2.it>
> ---
>
> Antonio Ospite writes:
> > BTW, with Stefan Beller we also identified some unneeded code which
> > could have been removed to alleviate the issue, but that would not have
> > solved it completely; so, I am not removing the unnecessary call to
> > repo_read_gitmodules() builtin/grep.c in this series, possibly this can
> > become a stand-alone change.
>
> Here is the stand-alone change.
>
Thank you for sending it.
Ciao,
Antonio
--
Antonio Ospite
https://ao2.it
https://twitter.com/ao2it
A: Because it messes up the order in which people normally read text.
See http://en.wikipedia.org/wiki/Posting_style
Q: Why is top-posting such a bad thing?
^ permalink raw reply [relevance 4%]
* Re: [PATCH v6 08/10] submodule: add a helper to check if it is safe to write to .gitmodules
@ 2018-10-05 23:50 4% ` Stefan Beller
2018-10-06 9:19 4% ` Antonio Ospite
2018-10-06 23:44 4% ` Junio C Hamano
0 siblings, 2 replies; 200+ results
From: Stefan Beller @ 2018-10-05 23:50 UTC (permalink / raw)
To: Antonio Ospite
Cc: Junio C Hamano, git, Brandon Williams, Jonathan Nieder, Jeff King,
SZEDER Gábor
> static int module_config(int argc, const char **argv, const char *prefix)
> {
> + enum {
> + CHECK_WRITEABLE = 1
> + } command = 0;
Can we have the default named? Then we would only use states
from within the enum?
^ permalink raw reply [relevance 4%]
* [PATCH] builtin/grep.c: remote superflous submodule code
@ 2018-10-05 22:45 16% Stefan Beller
2018-10-06 8:59 4% ` Antonio Ospite
` (2 more replies)
0 siblings, 3 replies; 200+ results
From: Stefan Beller @ 2018-10-05 22:45 UTC (permalink / raw)
To: git; +Cc: ao2, Stefan Beller
In f9ee2fcdfa (grep: recurse in-process using 'struct repository',
2017-08-02), we introduced a call to repo_read_gitmodules in builtin/grep
to simplify the submodule handling.
After ff6f1f564c4 (submodule-config: lazy-load a repository's .gitmodules
file, 2017-08-03) this is no longer necessary, but that commit did not
cleanup the whole tree, but just show cased the new way how to deal with
submodules in ls-files.
Cleanup the only remaining caller to repo_read_gitmodules outside of
submodule.c
Signed-off-by: Stefan Beller <sbeller@google.com>
---
Antonio Ospite writes:
> BTW, with Stefan Beller we also identified some unneeded code which
> could have been removed to alleviate the issue, but that would not have
> solved it completely; so, I am not removing the unnecessary call to
> repo_read_gitmodules() builtin/grep.c in this series, possibly this can
> become a stand-alone change.
Here is the stand-alone change.
The patch [1] contains the lines as deleted below in the context lines
but they would not conflict as there is one empty line between the changes
in this patch in [1].
[1] https://public-inbox.org/git/20181005130601.15879-10-ao2@ao2.it/
builtin/grep.c | 2 --
1 file changed, 2 deletions(-)
diff --git a/builtin/grep.c b/builtin/grep.c
index 601f801158..a6272b9c2f 100644
--- a/builtin/grep.c
+++ b/builtin/grep.c
@@ -427,8 +427,6 @@ static int grep_submodule(struct grep_opt *opt, struct repository *superproject,
if (repo_submodule_init(&submodule, superproject, path))
return 0;
- repo_read_gitmodules(&submodule);
-
/*
* NEEDSWORK: This adds the submodule's object directory to the list of
* alternates for the single in-memory object store. This has some bad
--
2.19.0
^ permalink raw reply related [relevance 16%]
* Re: [PATCH] grep: provide a noop --recursive option
2018-10-05 13:05 2% ` Mischa POSLAWSKY
@ 2018-10-05 19:17 5% ` Stefan Beller
0 siblings, 0 replies; 200+ results
From: Stefan Beller @ 2018-10-05 19:17 UTC (permalink / raw)
To: git; +Cc: Junio C Hamano, Ævar Arnfjörð Bjarmason, git, myon
On Fri, Oct 5, 2018 at 6:05 AM Mischa POSLAWSKY <git@shiar.nl> wrote:
>
> Junio C Hamano wrote 2018-10-05 1:19 (-0700):
> > Stefan Beller <sbeller@google.com> writes:
> >
> > > git-grep is always file/tree recursive, but there is --recurse-submodules
> > > which is off by default. Instead of providing a short alias to a noop,
> > > we could use -r for submodules. (And if you happen to have no
> > > submodules, this is a noop for you)
> >
> > I am not sure if it is an overall win for those who do have and use
> > submodules to easily be able to go recursive with a short-and-sweet
> > 'r', or even they want to work inside one project at a time most of
> > the time. If the latter, then using 'r' for recurse-submodules is
> > going to be a mistake (besides, other commands that have 'recursive'
> > typically use 'r' for its shorthand,and 'r' does not stand for
> > 'recurse-submodules' for them).
>
> Personally I would welcome a shorthand for --recurse-submodules,
> especially if --r^I no longer completes to this.
The new switch differs by one dash, so I'd think the double dashed
version would still autocomplete.
Unrelated to this, but more to submodules:
There is submodule.recurse which you may want to set.
Would you be interested in a more specific config option there?
(i.e. grep.recurseSubmodules to only apply to grep recursing into
submodules, just like fetch.recurseSubmodules only applies to fetch)
> It is also closer to the behaviour provided by grep -r as that recurses
> into submodules as well.
That sort of makes for the grep case, but not for other commands.
See the related discussion at
https://public-inbox.org/git/20180907064026.GB172953@aiede.svl.corp.google.com/
^ permalink raw reply [relevance 5%]
* [PATCH v6 00/10] Make submodules work if .gitmodules is not checked out
@ 2018-10-05 13:05 6% Antonio Ospite
0 siblings, 2 replies; 200+ results
From: Antonio Ospite @ 2018-10-05 13:05 UTC (permalink / raw)
To: gitster
Cc: git, Brandon Williams, Jonathan Nieder, Stefan Beller, Jeff King,
SZEDER Gábor, Antonio Ospite
Hi,
this series teaches git to try and read the .gitmodules file from the
index (:.gitmodules) or from the current branch (HEAD:.gitmodules) when
the file is not readily available in the working tree.
This can be used, together with sparse checkouts, to enable submodule
usage with programs like vcsh[1] which manage multiple repositories with
their working trees sharing the same path.
[1] https://github.com/RichiH/vcsh
Thanks to SZEDER Gábor we found out that the changes in patch 9 could
allow to access the object store in an inconsistent way when using
multi-threading in "git grep --recurse-submodules", this has been dealt
with in this revision.
BTW, with Stefan Beller we also identified some unneeded code which
could have been removed to alleviate the issue, but that would not have
solved it completely; so, I am not removing the unnecessary call to
repo_read_gitmodules() builtin/grep.c in this series, possibly this can
become a stand-alone change.
The problems from above also made me investigate what implications the
current use of a global object store had on my new addition, and now
I know that there is one case which I cannot support just yet: nested
submodules without .gitmodules in their working tree.
This case has been documented with a warning and test_expect_failure
items in tests, and hitting the unsupported case does not alter the
current behavior of git.
Apart form patch 9 and 10 there are no major changes to the previous
iteration.
IMHO we are in a place where the problem has been analyzed with enough
depth, the limitations have been identified and dealt with in a way that
should not affect current users nor diminish the usefulness of the new
code.
v5 of the series is here:
https://public-inbox.org/git/20180917140940.3839-1-ao2@ao2.it/
v4 of the series is here:
https://public-inbox.org/git/20180824132951.8000-1-ao2@ao2.it/
v3 of the series is here:
https://public-inbox.org/git/20180814110525.17801-1-ao2@ao2.it/
v2 of the series is here:
https://public-inbox.org/git/20180802134634.10300-1-ao2@ao2.it/
v1 of the series, with some background info, is here:
https://public-inbox.org/git/20180514105823.8378-1-ao2@ao2.it/
Changes since v5:
* print_config_from_gitmodules() in patch 1 now accepts a struct
repository argument.
* submodule--helper config in patch 5 has been adjusted to use the new
signature of print_config_from_gitmodules().
* In patch 9 the grep read lock in builtin/grep.c now covers all code
paths involving config_from_gitmodules().
FTR git-grep is the only place where config_from_gitmodules() is
called from multi-threaded code.
* Patch 9 also documents the rare case that cannot be supported just
yet, and adds a warning to the user.
* In patch 9, config_from_gitmodules() now does not read any config
when the config source is not specified.(I added a catch-all "else"
block) This match more closely the behavior of the old code using
git_config_from_file.
* Added a new test tool in patch 10 to exercise config_read_config()
in a more direct way, passing an arbitrary repository.
Admittedly, patch 10 performs a similar test to the one added to
t7814 in patch 9, so I'd be OK with dropping patch 10 if it is too
specific.
Thank you,
Antonio
Antonio Ospite (10):
submodule: add a print_config_from_gitmodules() helper
submodule: factor out a config_set_in_gitmodules_file_gently function
t7411: merge tests 5 and 6
t7411: be nicer to future tests and really clean things up
submodule--helper: add a new 'config' subcommand
submodule: use the 'submodule--helper config' command
t7506: clean up .gitmodules properly before setting up new scenario
submodule: add a helper to check if it is safe to write to .gitmodules
submodule: support reading .gitmodules when it's not in the working
tree
t/helper: add test-submodule-nested-repo-config
Makefile | 1 +
builtin/grep.c | 17 ++-
builtin/submodule--helper.c | 40 ++++++
cache.h | 2 +
git-submodule.sh | 13 +-
submodule-config.c | 68 ++++++++-
submodule-config.h | 2 +
submodule.c | 28 +++-
submodule.h | 1 +
t/helper/test-submodule-nested-repo-config.c | 30 ++++
t/helper/test-tool.c | 1 +
t/helper/test-tool.h | 1 +
t/t7411-submodule-config.sh | 141 +++++++++++++++++--
t/t7416-submodule-sparse-gitmodules.sh | 78 ++++++++++
t/t7506-status-submodule.sh | 3 +-
t/t7814-grep-recurse-submodules.sh | 16 +++
16 files changed, 410 insertions(+), 32 deletions(-)
create mode 100644 t/helper/test-submodule-nested-repo-config.c
create mode 100755 t/t7416-submodule-sparse-gitmodules.sh
--
Antonio Ospite
https://ao2.it
https://twitter.com/ao2it
A: Because it messes up the order in which people normally read text.
See http://en.wikipedia.org/wiki/Posting_style
Q: Why is top-posting such a bad thing?
^ permalink raw reply [relevance 6%]
* Re: [PATCH] grep: provide a noop --recursive option
2018-10-05 8:19 5% ` Junio C Hamano
@ 2018-10-05 13:05 2% ` Mischa POSLAWSKY
2018-10-05 19:17 5% ` Stefan Beller
0 siblings, 1 reply; 200+ results
From: Mischa POSLAWSKY @ 2018-10-05 13:05 UTC (permalink / raw)
To: Junio C Hamano
Cc: Stefan Beller, Ævar Arnfjörð Bjarmason, git, myon
Junio C Hamano wrote 2018-10-05 1:19 (-0700):
> Stefan Beller <sbeller@google.com> writes:
>
> > git-grep is always file/tree recursive, but there is --recurse-submodules
> > which is off by default. Instead of providing a short alias to a noop,
> > we could use -r for submodules. (And if you happen to have no
> > submodules, this is a noop for you)
>
> I am not sure if it is an overall win for those who do have and use
> submodules to easily be able to go recursive with a short-and-sweet
> 'r', or even they want to work inside one project at a time most of
> the time. If the latter, then using 'r' for recurse-submodules is
> going to be a mistake (besides, other commands that have 'recursive'
> typically use 'r' for its shorthand,and 'r' does not stand for
> 'recurse-submodules' for them).
Personally I would welcome a shorthand for --recurse-submodules,
especially if --r^I no longer completes to this.
It is also closer to the behaviour provided by grep -r as that recurses
into submodules as well.
Regards,
--
Mischa
^ permalink raw reply [relevance 2%]
* Re: [PATCH] grep: provide a noop --recursive option
2018-10-01 19:23 5% ` Stefan Beller
2018-10-05 8:15 2% ` Christoph Berg
@ 2018-10-05 8:19 5% ` Junio C Hamano
2018-10-05 13:05 2% ` Mischa POSLAWSKY
1 sibling, 1 reply; 200+ results
From: Junio C Hamano @ 2018-10-05 8:19 UTC (permalink / raw)
To: Stefan Beller; +Cc: Ævar Arnfjörð Bjarmason, git, myon
Stefan Beller <sbeller@google.com> writes:
> git-grep is always file/tree recursive, but there is --recurse-submodules
> which is off by default. Instead of providing a short alias to a noop,
> we could use -r for submodules. (And if you happen to have no
> submodules, this is a noop for you)
I am not sure if it is an overall win for those who do have and use
submodules to easily be able to go recursive with a short-and-sweet
'r', or even they want to work inside one project at a time most of
the time. If the latter, then using 'r' for recurse-submodules is
going to be a mistake (besides, other commands that have 'recursive'
typically use 'r' for its shorthand,and 'r' does not stand for
'recurse-submodules' for them).
^ permalink raw reply [relevance 5%]
* Re: [PATCH] grep: provide a noop --recursive option
2018-10-01 19:23 5% ` Stefan Beller
@ 2018-10-05 8:15 2% ` Christoph Berg
2018-10-05 8:19 5% ` Junio C Hamano
1 sibling, 0 replies; 200+ results
From: Christoph Berg @ 2018-10-05 8:15 UTC (permalink / raw)
To: Stefan Beller; +Cc: Ævar Arnfjörð Bjarmason, git, Junio C Hamano
Re: Stefan Beller 2018-10-01 <CAGZ79kbw96x2Dow7d-sUfOHXiVN8j9KgqEObo+TrVd5zWKbaEA@mail.gmail.com>
> git-grep is always file/tree recursive, but there is --recurse-submodules
> which is off by default. Instead of providing a short alias to a noop,
> we could use -r for submodules. (And if you happen to have no
> submodules, this is a noop for you)
That would be fine for me.
Thanks,
Christoph
^ permalink raw reply [relevance 2%]
* Re: [PATCH 1/1] protocol: limit max protocol version per service
@ 2018-10-03 22:47 3% ` Stefan Beller
0 siblings, 0 replies; 200+ results
From: Stefan Beller @ 2018-10-03 22:47 UTC (permalink / raw)
To: Josh Steadmon; +Cc: git, Jonathan Nieder, Junio C Hamano
On Wed, Oct 3, 2018 at 2:34 PM Josh Steadmon <steadmon@google.com> wrote:
>
> I believe that git-upload-archive can still speak version 1 without any
> trouble, but it at least doesn't break anything in the test suite to
> limit this to v0 either.
ok, let me just play around with archive to follow along:
Running the excerpt that I found in one of the tests in t5000
GIT_TRACE_PACKET=1 git archive --remote=. HEAD >b5.tar
(whoooa ... that spews a lot of text into my terminal, which makes
sense as transported tar files unlike packets starting with PACK are
not cut short; so onwards:)
$ git init test && cd test
$ GIT_TRACE_PACKET=1 git archive --remote=. HEAD >b5.tar
...
git< \2fatal: no such ref: HEAD
...
fatal: sent error to the client: git upload-archive: archiver died
with error
remote: git upload-archive: archiver died with error
This sounds similar to a bug that I have on my todo list for
clone recursing into submodules. Maybe we need to talk
about HEAD-less repositories and how to solve handling them
in general. Usually it doesn't happen except for corner cases like
now, so:
$ git commit --allow-empty -m "commit"
[master (root-commit) 24d7967] commit
$ GIT_TRACE_PACKET=1 git archive --remote=. HEAD >b5.tar
15:28:00.762615 pkt-line.c:80 packet: git> argument HEAD
15:28:00.762704 pkt-line.c:80 packet: git> 0000
15:28:00.766336 pkt-line.c:80 packet: git> ACK
15:28:00.766428 pkt-line.c:80 packet: git> 0000
15:28:00.766483 pkt-line.c:80 packet: git< ACK
15:28:00.766508 pkt-line.c:80 packet: git< 0000
15:28:00.767694 pkt-line.c:80 packet: git< \2
15:28:00.767524 pkt-line.c:80 packet: git< argument HEAD
15:28:00.767583 pkt-line.c:80 packet: git< 0000
remote: 15:28:00.767524 pkt-line.c:80 packet: git<
argument HEAD
remote: 15:28:00.767583 pkt-line.c:80 packet: git< 0000
15:28:00.768758 pkt-line.c:80 packet: git> 0000
15:28:00.770475 pkt-line.c:80 packet: git<
\1pax_global_header
... \0\0\0\0\0\0\0\0\0\0\0\0\ ...
# not too bad but the tar file contains a lot of zeros
$
Ah I forgot the crucial part, so
$ GIT_TRACE_PACKET=1 git -c protocol.version=1 archive --remote=.
HEAD >b5.tar
15:33:10.030508 pkt-line.c:80 packet: git> argument HEAD
...
This pretty much looks like v0 as v1 would send a "version 1", c.f.
$ GIT_TRACE_PACKET=1 git -c protocol.version=1 fetch .
15:34:26.111013 pkt-line.c:80 packet: upload-pack> version 1
15:34:26.111140 pkt-line.c:80 packet: fetch< version 1
....
>
> Is there a method or design for advertising multiple acceptable versions
> from the client?
I think the client can send multiple versions, looking through protocol.c
(and not the documentation as I should for this:)
/*
* Determine which protocol version the client has requested. Since
* multiple 'version' keys can be sent by the client, indicating that
* the client is okay to speak any of them, select the greatest version
* that the client has requested. This is due to the assumption that
* the most recent protocol version will be the most state-of-the-art.
*/
...
const char *git_protocol = getenv(GIT_PROTOCOL_ENVIRONMENT);
string_list_split(&list, git_protocol, ':', -1);
...
for_each_string_list_item(item, &list) {
...
if (skip_prefix(item->string, "version=", &value))
...
in determine_protocol_version_server which already had the client
speak to it, so I think at least the server can deal with multiple versions.
But given that we transport the version in env variables, we'd
need to be extra careful if we just did not see the version
in the --remote=. above?
> From my understanding, we can only add a single
> version=X field in the advertisement, but IIUC we can extend this fairly
> easily? Perhaps we can have "version=X" to mean the preferred version,
> and then a repeatable "acceptable_version=Y" field or similar?
Just re-use "version X", separated by colons as above.
> > From a maintenance perspective, do we want to keep
> > this part of the code central, as it ties protocol (as proxied
> > by service name) to the max version number?
> > I would think that we'd rather have the decision local to the
> > code, i.e. builtin/fetch would need to tell protocol.c that it
> > can do (0,1,2) and builtin/push can do (0,1), and then the
> > networking layers of code would figure out by the input
> > from the caller and the input from the user (configured
> > protocol.version) what is the best to go forward from
> > then on.
>
> I like having it centralized, because enforcing this in git_connect()
> and discover_refs() catches all the outgoing version advertisements, but
> there's lots of code paths that lead to those two functions that would
> all have to have the acceptable version numbers plumbed through.
Makes sense.
> I suppose we could also have a registry of services to version numbers,
> but I tend to dislike non-local sources of data. But if the list likes
> that approach better, I'll be happy to implement it.
> > But I guess having the central place here is not to
> > bad either. How will it cope with the desire of protocol v2
> > to have only one end point (c.f. serve.{c,h} via builtin/serve
> > as "git serve") ?
>
> I'm not sure about this. In my series to add a v2 archive command, I
> added support for a new endpoint for proto v2 and I don't recall seeing
> any complaints, but that is still open for review.
Ah I guess new end points would imply that you can speak at least
a given version N.
> I suppose if we are strict about serving from a single endpoint, the
> version registry makes more sense, and individual operations can declare
> acceptable version numbers before calling any network code?
Ah yeah, that makes sense!
Thanks for your explanations and prodding,
Stefan
^ permalink raw reply [relevance 3%]
* Re: [BUG] Segfault in "git submodule"
2018-10-01 21:31 4% ` Raymond Jennings
@ 2018-10-02 4:10 2% ` Raymond Jennings
0 siblings, 0 replies; 200+ results
From: Raymond Jennings @ 2018-10-02 4:10 UTC (permalink / raw)
To: sbeller; +Cc: pclouds, avarab, git
I instructed downstream to update their repository.
On Mon, Oct 1, 2018 at 2:31 PM Raymond Jennings <shentino@gmail.com> wrote:
>
> Yes, git 2.16.4 to be exact.
>
> I upgraded to 2.19 after ~arch keywording the package on gentoo and
> that fixed it.
> On Mon, Oct 1, 2018 at 12:19 PM Stefan Beller <sbeller@google.com> wrote:
> >
> > On Sat, Sep 29, 2018 at 9:43 AM Raymond Jennings <shentino@gmail.com> wrote:
> > >
> > > I have a repo, but it appears to be specific to staging area state.
> > > It only segfaults when I have a certain file deleted.
> > >
> > > Where do you want me to upload it?
> > > On Sat, Sep 29, 2018 at 8:34 AM Duy Nguyen <pclouds@gmail.com> wrote:
> > > >
> > > > On Sat, Sep 29, 2018 at 5:31 PM Ævar Arnfjörð Bjarmason
> > > > <avarab@gmail.com> wrote:
> > > > > > #1 refs_resolve_ref_unsafe (refs=0x0,
> > > > > > refname=refname@entry=0x55e863062253 "HEAD",
> > > > > > resolve_flags=resolve_flags@entry=1, oid=oid@entry=0x7ffdc834b1c0,
> > > > > > flags=flags@entry=0x7ffdc834b1bc) at refs.c:1493
> > > >
> > > > refs is NULL. It looks like somebody fails to get the right submodule
> > > > ref store (or tries to get it but fails to check if it may return
> > > > NULL)
> >
> > This is spot on.
> >
> > Raymond, are you on Git v2.16.0 by any chance?
> > (and if now which version are you on).
> >
> > I suspect 2.16, as that is a version of Git, in which there happens to be
> > a call into the refs subsystem in submodule--helper.c in line 624.
> >
> > Is it possible to upgrade Git (to v2.18.0 or later) or cherry-pick
> > 74b6bda32f (submodule: check for NULL return of get_submodule_ref_store(),
> > 2018-03-28) into your copy of Git?
> >
> > Thanks,
> > Stefan
^ permalink raw reply [relevance 2%]
* [PATCH] submodule update --remote: introduce pinning
2018-09-06 22:54 7% ` Jonathan Nieder
@ 2018-10-02 0:17 26% ` Stefan Beller
0 siblings, 0 replies; 200+ results
From: Stefan Beller @ 2018-10-02 0:17 UTC (permalink / raw)
To: jrnieder; +Cc: git, sbeller
gitmodules(5) sayeth:
submodule.<name>.branch
A remote branch name for tracking updates in the upstream
submodule. If the option is not specified, it defaults to master.
This doesn't allow having a "pinned" submodule that should not be updated
from upstream. We should change this to have no default --- if branch is
not specified, don't update that submodule, just like in Gerrit's
corresponding feature[1].
[1] https://gerrit-review.googlesource.com/Documentation/user-submodules.html#_defining_the_submodule_branch
However changing defaults is difficult as it may surprise the user,
Jonathan came up with a 4 step plan:
Step 0: introduce
# current behavior:
git submodule update --remote --remote-default-to-master
# new behavior:
git submodule update --remote --remote-pinned
and treat plain "git submodule update --remote" as --default-to-master.
Step 1: when neither --default-to-master nor --no-default-to-master
has been passed, warn when encountering a submodule with no branch
and treat it as "master".
Step 2: when neither --default-to-master nor --no-default-to-master
has been passed, error out when encountering a submodule with no
branch.
Step 3: when neither --default-to-master nor --no-default-to-master
has been passed, warn when encountering a submodule with no branch
and treat it as pinned.
Step 4: eliminate the warning.
This implements step 0 and 1.
Signed-off-by: Stefan Beller <sbeller@google.com>
---
Sorry that this took so long, this is a rough patch for steps 0 & 1,
I'd still need to update docs.
Stefan
Documentation/gitmodules.txt | 11 ++++++-----
builtin/submodule--helper.c | 36 +++++++++++++++++++++++++++++-------
git-submodule.sh | 34 +++++++++++++++++++++++++---------
t/t7406-submodule-update.sh | 29 +++++++++++++++++++++++++++++
4 files changed, 89 insertions(+), 21 deletions(-)
diff --git a/Documentation/gitmodules.txt b/Documentation/gitmodules.txt
index 4d63def206..fe42dbdb3e 100644
--- a/Documentation/gitmodules.txt
+++ b/Documentation/gitmodules.txt
@@ -50,11 +50,12 @@ submodule.<name>.update::
submodule.<name>.branch::
A remote branch name for tracking updates in the upstream submodule.
- If the option is not specified, it defaults to 'master'. A special
- value of `.` is used to indicate that the name of the branch in the
- submodule should be the same name as the current branch in the
- current repository. See the `--remote` documentation in
- linkgit:git-submodule[1] for details.
+ A special value of `.` is used to indicate that the name of the
+ branch in the submodule should be the same name as the current
+ branch in the current repository. See the `--remote` documentation
+ in linkgit:git-submodule[1] for details.
+ If not set, the default for `git submodule update --remote` is
+ to update the submodule to the superproject's recorded object id.
submodule.<name>.fetchRecurseSubmodules::
This option can be used to control recursive fetching of this
diff --git a/builtin/submodule--helper.c b/builtin/submodule--helper.c
index 40844870cf..6413f2b410 100644
--- a/builtin/submodule--helper.c
+++ b/builtin/submodule--helper.c
@@ -1889,7 +1889,7 @@ static int resolve_relative_path(int argc, const char **argv, const char *prefix
return 0;
}
-static const char *remote_submodule_branch(const char *path)
+static const char *remote_submodule_branch(const char *path, int default_pinned)
{
const struct submodule *sub;
const char *branch = NULL;
@@ -1904,8 +1904,12 @@ static const char *remote_submodule_branch(const char *path)
branch = sub->branch;
free(key);
- if (!branch)
- return "master";
+ if (!branch) {
+ if (default_pinned)
+ return "";
+ else
+ return "master";
+ }
if (!strcmp(branch, ".")) {
const char *refname = resolve_ref_unsafe("HEAD", 0, NULL, NULL);
@@ -1932,12 +1936,30 @@ static int resolve_remote_submodule_branch(int argc, const char **argv,
{
const char *ret;
struct strbuf sb = STRBUF_INIT;
- if (argc != 2)
- die("submodule--helper remote-branch takes exactly one arguments, got %d", argc);
+ int default_pinned = 0;
+
+ struct option remote_options[] = {
+ OPT_SET_INT(0, "default-master", &default_pinned,
+ N_("unconfigured submodules default to master branch"), 0),
+ OPT_SET_INT(0, "default-pinned", &default_pinned,
+ N_("unconfigured submodules default to superproject object id"), 1),
+ OPT_END()
+ };
+
+ const char *const git_submodule_helper_usage[] = {
+ N_("git submodule--helper remote-branch [--default-{master, pinned}] -- <path>"),
+ NULL
+ };
+
+ argc = parse_options(argc, argv, prefix, remote_options,
+ git_submodule_helper_usage, 0);
+
+ if (argc != 1)
+ die("submodule--helper remote-branch takes exactly one path, got %d", argc);
- ret = remote_submodule_branch(argv[1]);
+ ret = remote_submodule_branch(argv[0], default_pinned);
if (!ret)
- die("submodule %s doesn't exist", argv[1]);
+ die("submodule %s doesn't exist", argv[0]);
printf("%s", ret);
strbuf_release(&sb);
diff --git a/git-submodule.sh b/git-submodule.sh
index 1b568e29b9..829b90ea97 100755
--- a/git-submodule.sh
+++ b/git-submodule.sh
@@ -431,6 +431,7 @@ fetch_in_submodule () (
#
cmd_update()
{
+ remote_default_option=
# parse $args after "submodule ... update".
while test $# -ne 0
do
@@ -450,6 +451,12 @@ cmd_update()
--remote)
remote=1
;;
+ --remote-default-to-master)
+ remote_default_option=--default-master
+ ;;
+ --remote-pinned)
+ remote_default_option=--default-pinned
+ ;;
-N|--no-fetch)
nofetch=1
;;
@@ -555,17 +562,26 @@ cmd_update()
if test -n "$remote"
then
- branch=$(git submodule--helper remote-branch "$sm_path")
- if test -z "$nofetch"
+ if test -z $remote_default_option
then
- # Fetch remote before determining tracking $sha1
- fetch_in_submodule "$sm_path" $depth ||
- die "$(eval_gettext "Unable to fetch in submodule path '\$sm_path'")"
+ say "--remote needs clarification: --remote-pinned or --remote-default-to-master?"
+ say "assuming --remote-default-to-master for now"
+ remote_default_option=--default-master
+ fi
+ branch=$(git submodule--helper remote-branch ${remote_default_option} -- "$sm_path")
+ if test -n "$branch"
+ then
+ if test -z "$nofetch"
+ then
+ # Fetch remote before determining tracking $sha1
+ fetch_in_submodule "$sm_path" $depth ||
+ die "$(eval_gettext "Unable to fetch in submodule path '\$sm_path'")"
+ fi
+ remote_name=$(sanitize_submodule_env; cd "$sm_path" && get_default_remote)
+ sha1=$(sanitize_submodule_env; cd "$sm_path" &&
+ git rev-parse --verify "${remote_name}/${branch}") ||
+ die "$(eval_gettext "Unable to find current \${remote_name}/\${branch} revision in submodule path '\$sm_path'")"
fi
- remote_name=$(sanitize_submodule_env; cd "$sm_path" && get_default_remote)
- sha1=$(sanitize_submodule_env; cd "$sm_path" &&
- git rev-parse --verify "${remote_name}/${branch}") ||
- die "$(eval_gettext "Unable to find current \${remote_name}/\${branch} revision in submodule path '\$sm_path'")"
fi
if test "$subsha1" != "$sha1" || test -n "$force"
diff --git a/t/t7406-submodule-update.sh b/t/t7406-submodule-update.sh
index 10dc91620a..3019978211 100755
--- a/t/t7406-submodule-update.sh
+++ b/t/t7406-submodule-update.sh
@@ -260,6 +260,35 @@ test_expect_success 'submodule update --remote should fetch upstream changes wit
)
'
+test_expect_success 'submodule update --remote should not fetch upstream when no branch is set' '
+ (
+ cd super &&
+ test_might_fail git config --unset -f .gitmodules submodule.submodule.branch &&
+ git add .gitmodules &&
+ git commit --allow-empty -m "submodules: pin in superproject branch" &&
+ git -C ../submodule checkout -f master
+ ) &&
+ (
+ cd submodule &&
+ echo line4b >>file &&
+ git add file &&
+ test_tick &&
+ git commit -m "upstream line4b"
+ ) &&
+ (
+ cd super &&
+
+ git submodule update --remote-pinned --remote --force submodule &&
+ git status --porcelain --untracked=no --ignore-submodules=none >actual &&
+ test_must_be_empty actual &&
+
+ git submodule update --remote-default-to-master --remote --force submodule &&
+ git -C submodule log -1 --oneline >actual &&
+ git -C ../submodule log -1 --oneline >expect &&
+ test_cmp expect actual
+ )
+'
+
test_expect_success 'local config should override .gitmodules branch' '
(cd submodule &&
git checkout test-branch &&
--
2.19.0
^ permalink raw reply related [relevance 26%]
* Re: [BUG] Segfault in "git submodule"
2018-10-01 19:19 7% ` Stefan Beller
@ 2018-10-01 21:31 4% ` Raymond Jennings
2018-10-02 4:10 2% ` Raymond Jennings
0 siblings, 1 reply; 200+ results
From: Raymond Jennings @ 2018-10-01 21:31 UTC (permalink / raw)
To: sbeller; +Cc: pclouds, avarab, git
Yes, git 2.16.4 to be exact.
I upgraded to 2.19 after ~arch keywording the package on gentoo and
that fixed it.
On Mon, Oct 1, 2018 at 12:19 PM Stefan Beller <sbeller@google.com> wrote:
>
> On Sat, Sep 29, 2018 at 9:43 AM Raymond Jennings <shentino@gmail.com> wrote:
> >
> > I have a repo, but it appears to be specific to staging area state.
> > It only segfaults when I have a certain file deleted.
> >
> > Where do you want me to upload it?
> > On Sat, Sep 29, 2018 at 8:34 AM Duy Nguyen <pclouds@gmail.com> wrote:
> > >
> > > On Sat, Sep 29, 2018 at 5:31 PM Ævar Arnfjörð Bjarmason
> > > <avarab@gmail.com> wrote:
> > > > > #1 refs_resolve_ref_unsafe (refs=0x0,
> > > > > refname=refname@entry=0x55e863062253 "HEAD",
> > > > > resolve_flags=resolve_flags@entry=1, oid=oid@entry=0x7ffdc834b1c0,
> > > > > flags=flags@entry=0x7ffdc834b1bc) at refs.c:1493
> > >
> > > refs is NULL. It looks like somebody fails to get the right submodule
> > > ref store (or tries to get it but fails to check if it may return
> > > NULL)
>
> This is spot on.
>
> Raymond, are you on Git v2.16.0 by any chance?
> (and if now which version are you on).
>
> I suspect 2.16, as that is a version of Git, in which there happens to be
> a call into the refs subsystem in submodule--helper.c in line 624.
>
> Is it possible to upgrade Git (to v2.18.0 or later) or cherry-pick
> 74b6bda32f (submodule: check for NULL return of get_submodule_ref_store(),
> 2018-03-28) into your copy of Git?
>
> Thanks,
> Stefan
^ permalink raw reply [relevance 4%]
* Re: [PATCH v5 9/9] submodule: support reading .gitmodules when it's not in the working tree
2018-10-01 15:45 7% ` Antonio Ospite
@ 2018-10-01 19:42 7% ` Stefan Beller
0 siblings, 0 replies; 200+ results
From: Stefan Beller @ 2018-10-01 19:42 UTC (permalink / raw)
To: Antonio Ospite
Cc: SZEDER Gábor, Junio C Hamano, git, Brandon Williams,
Daniel Graña, Jonathan Nieder, Richard Hartmann,
Ævar Arnfjörð Bjarmason
On Mon, Oct 1, 2018 at 8:45 AM Antonio Ospite <ao2@ao2.it> wrote:
>
> On Thu, 27 Sep 2018 11:00:52 -0700
> Stefan Beller <sbeller@google.com> wrote:
>
> > On Thu, Sep 27, 2018 at 7:44 AM Antonio Ospite <ao2@ao2.it> wrote:
> [...]
> > > OK, so the plan for v6 is:
> > >
> > > - avoid the corruption issues spotted by Gábor by removing the call
> > > to repo_read_gitmodules in builtin/grep.c (this still does not fix
> > > the potential problem with nested submodules).
> > >
>
> Actually that is not enough to fix the inconsistent access to the
> object store: the functions is_submodule_active() and
> repo_submodule_init() too end up calling config_from_gitmodules() and
> need protecting as well, so I am going to put them under the git read
> lock and leave repo_read_gitmodules() in place for now.
>
>
> I am asking because the whole point of this patchset is to *enable* the
> use of submodules without .gitmodules in the working tree of the
> superproject. :)
I was imprecise and meant to
s/.gitmodules/mechanism to configure the name <-> path mapping/
In this series, the .gitmodules may not be present in the working tree,
but it is still there in the repo. Later we may want to rename that file
or put it into a magic branch, and I'd still find it a good idea.
What I find a bad idea is to have only a gitlink and a repo put
into that path and expect that it magically works as then it is not
a submodule, but some "halfway there thing". We need to have
an explicit "yes this is a submodule" statement, (which currently
comes from the .gitmodules file in the working tree), and I am not
attached to where it comes from, but that it exists.
^ permalink raw reply [relevance 7%]
* Re: [PATCH] grep: provide a noop --recursive option
@ 2018-10-01 19:23 5% ` Stefan Beller
2018-10-05 8:15 2% ` Christoph Berg
2018-10-05 8:19 5% ` Junio C Hamano
0 siblings, 2 replies; 200+ results
From: Stefan Beller @ 2018-10-01 19:23 UTC (permalink / raw)
To: Ævar Arnfjörð Bjarmason; +Cc: git, Junio C Hamano, myon
On Sat, Sep 29, 2018 at 7:55 AM Ævar Arnfjörð Bjarmason
<avarab@gmail.com> wrote:
>
> This --recursive (-r) option does nothing, and is purely here to
> appease people who have "grep -r ..." burned into their muscle memory.
>
> Requested-by: Christoph Berg <myon@debian.org>
> Signed-off-by: Ævar Arnfjörð Bjarmason <avarab@gmail.com>
> ---
>
> On Sat, Sep 29, 2018 at 4:10 PM Christoph Berg <myon@debian.org> wrote:
> >
> > I often use "grep -r $pattern" to recursively grep a source tree. If
> > that takes too long, I hit ^C and tag "git" in front of the command
> > line and re-run it. git then complains "error: unknown switch `r'"
> > because "git grep" is naturally recursive.
> >
> > Could we have "git grep -r" accept the argument for compatibility?
> > Other important grep switches like "-i" are compatible, adding -r
> > would improve usability.
>
> I don't have an opinion on this either way, it doesn't scratch my
> itch, but hey, why not. Here's a patch to implement it.
>
> Documentation/git-grep.txt | 6 ++++++
> builtin/grep.c | 3 +++
> t/t7810-grep.sh | 8 ++++++++
> 3 files changed, 17 insertions(+)
>
> diff --git a/Documentation/git-grep.txt b/Documentation/git-grep.txt
> index a3049af1a3..a1aea8be4e 100644
> --- a/Documentation/git-grep.txt
> +++ b/Documentation/git-grep.txt
> @@ -290,6 +290,12 @@ providing this option will cause it to die.
> Do not output matched lines; instead, exit with status 0 when
> there is a match and with non-zero status when there isn't.
>
> +-r::
> +--recursive::
> + This option does nothing. git-grep is always recursive. This
> + noop option is provided for compatibility with the muscle
> + memory of people used to grep(1).
git-grep is always file/tree recursive, but there is --recurse-submodules
which is off by default. Instead of providing a short alias to a noop,
we could use -r for submodules. (And if you happen to have no
submodules, this is a noop for you)
^ permalink raw reply [relevance 5%]
* Re: [BUG] Segfault in "git submodule"
@ 2018-10-01 19:19 7% ` Stefan Beller
2018-10-01 21:31 4% ` Raymond Jennings
0 siblings, 1 reply; 200+ results
From: Stefan Beller @ 2018-10-01 19:19 UTC (permalink / raw)
To: shentino; +Cc: Duy Nguyen, Ævar Arnfjörð Bjarmason, git
On Sat, Sep 29, 2018 at 9:43 AM Raymond Jennings <shentino@gmail.com> wrote:
>
> I have a repo, but it appears to be specific to staging area state.
> It only segfaults when I have a certain file deleted.
>
> Where do you want me to upload it?
> On Sat, Sep 29, 2018 at 8:34 AM Duy Nguyen <pclouds@gmail.com> wrote:
> >
> > On Sat, Sep 29, 2018 at 5:31 PM Ævar Arnfjörð Bjarmason
> > <avarab@gmail.com> wrote:
> > > > #1 refs_resolve_ref_unsafe (refs=0x0,
> > > > refname=refname@entry=0x55e863062253 "HEAD",
> > > > resolve_flags=resolve_flags@entry=1, oid=oid@entry=0x7ffdc834b1c0,
> > > > flags=flags@entry=0x7ffdc834b1bc) at refs.c:1493
> >
> > refs is NULL. It looks like somebody fails to get the right submodule
> > ref store (or tries to get it but fails to check if it may return
> > NULL)
This is spot on.
Raymond, are you on Git v2.16.0 by any chance?
(and if now which version are you on).
I suspect 2.16, as that is a version of Git, in which there happens to be
a call into the refs subsystem in submodule--helper.c in line 624.
Is it possible to upgrade Git (to v2.18.0 or later) or cherry-pick
74b6bda32f (submodule: check for NULL return of get_submodule_ref_store(),
2018-03-28) into your copy of Git?
Thanks,
Stefan
^ permalink raw reply [relevance 7%]
* Re: [PATCH v5 9/9] submodule: support reading .gitmodules when it's not in the working tree
2018-09-27 18:00 7% ` Stefan Beller
@ 2018-10-01 15:45 7% ` Antonio Ospite
2018-10-01 19:42 7% ` Stefan Beller
0 siblings, 1 reply; 200+ results
From: Antonio Ospite @ 2018-10-01 15:45 UTC (permalink / raw)
To: Stefan Beller
Cc: SZEDER Gábor, Junio C Hamano, git, Brandon Williams,
Daniel Graña, Jonathan Nieder, Richard Hartmann,
Ævar Arnfjörð Bjarmason
On Thu, 27 Sep 2018 11:00:52 -0700
Stefan Beller <sbeller@google.com> wrote:
> On Thu, Sep 27, 2018 at 7:44 AM Antonio Ospite <ao2@ao2.it> wrote:
[...]
> > OK, so the plan for v6 is:
> >
> > - avoid the corruption issues spotted by Gábor by removing the call
> > to repo_read_gitmodules in builtin/grep.c (this still does not fix
> > the potential problem with nested submodules).
> >
Actually that is not enough to fix the inconsistent access to the
object store: the functions is_submodule_active() and
repo_submodule_init() too end up calling config_from_gitmodules() and
need protecting as well, so I am going to put them under the git read
lock and leave repo_read_gitmodules() in place for now.
Removing unneeded code can go in a possible stand-alone patch.
> > - add a new test-tool which better exercises the new
> > config_from_gitmodules code,
>
> Sounds good.
>
> >
> > - add also a test_expect_failure test to document the use case that
> > cannot be supported yet: nested submodules without .gitmodules in
> > their working tree.
>
> Personally I would want to live in a world where we don't *have* to nor
> *want* to support submodules without .gitmodules in the respective
> superproject.
>
Just to double check: are you referring to *nested* submodules in the
sentence above?
I am asking because the whole point of this patchset is to *enable* the
use of submodules without .gitmodules in the working tree of the
superproject. :)
It's just that current limitations in git do not allow to support this
for *nested* submodules yet.
> We did support some use cases historically that I would make sure to
> continue to support, but I am not sure how much effort we want to spend
> on supporting further use cases of incomplete submodules.
>
> Feel free to do so, as such tests help to document the boundaries.
>
Let's see how v6 turns out.
Thanks,
Antonio
--
Antonio Ospite
https://ao2.it
https://twitter.com/ao2it
A: Because it messes up the order in which people normally read text.
See http://en.wikipedia.org/wiki/Posting_style
Q: Why is top-posting such a bad thing?
^ permalink raw reply [relevance 7%]
* Re: [PATCH] FYI / RFC: submodules: introduce repo-like workflow
2018-09-28 19:26 6% ` Ævar Arnfjörð Bjarmason
@ 2018-09-28 20:23 6% ` Stefan Beller
0 siblings, 0 replies; 200+ results
From: Stefan Beller @ 2018-09-28 20:23 UTC (permalink / raw)
To: Ævar Arnfjörð Bjarmason; +Cc: git, Jonathan Nieder
On Fri, Sep 28, 2018 at 12:26 PM Ævar Arnfjörð Bjarmason
<avarab@gmail.com> wrote:
>
>
> On Thu, Sep 27 2018, Stefan Beller wrote:
>
> > Internally we have rolled out this as an experiment for
> > "submodules replacing the repo tool[1]". The repo tool is described as:
> >
> > Repo unifies Git repositories when necessary, performs uploads to the
> > Gerrit revision control system, and automates parts of the Android
> > development workflow. Repo is not meant to replace Git, only to make
> > it easier to work with Git in the context of Android. The repo command
> > is an executable Python script that you can put anywhere in your path.
> >
> > In working with the Android source files, you use Repo for
> > across-network operations. For example, with a single Repo command you
> > can download files from multiple repositories into your local working
> > directory.
> >
> > In most situations, you can use Git instead of Repo, or mix Repo and
> > Git commands to form complex commands.
> >
> > [1] https://source.android.com/setup/develop/
>
> Some questions just out of curiosity, not for this patch in particular:
>
> Those docs seem to describe the situation without this patch, with this
> patch is the repo tool fully replaced?
Yeah I started by describing the status quo. Some points to add:
* The repo UX looks very similar to perforce, as that was used internally
at Google at the time a lot. Git wasn't around for long in the days of
early Android. Now that Git is well known, new grads joining are more likely
to know Git than perforce.
* repo has no tests because it started as a small script
* maintenance/releases of repo are not ideal.
So yes, one of our long term goals is to replace repo with Git-only
(on the client side) workflow.
> How are you planning to migrate from repo to this on a repository-data
> basis
We don't plan to migrate the existing clients. You'd have to re-clone.
>, does gerrit also populate .gitmodules files appropriately, which
> your clone --recurse-submodules will pick up, but repo will just ignore,
> so you can use the two in parallel?
Gerrit has native support to update submodules in a superproject.
So if you submit code to a project that is a submodule of a superproject
the superproject updates its gitlink. (and if you use a topic based workflow
to submit to two projects, it will show up as an atomic update in the
superproject by having just one commit updating two gitlinks)
Also Gerrit has a plugin "supermanifest" [1], which tracks repo
manifests and mirrors changes from the manifest into the
superproject, e.g. adding a new project to the manifest will
add a new submodule to the superproject.
[1] https://gerrit.googlesource.com/plugins/supermanifest/
With both Gerrits internal superproject subscription and that plugin
it should be possible to use repo or git-submodule in parallel (on an
organisational level, i.e. you choose one of them, and your coworker
chooses the other one)
> Now "repo init -u" takes a URL to a manifest of repositories to stitch
> together, I've understood from past conversations (but am not sure) that
> this is used e.g. by downstream Android vendors so they can use what
> Google's using + whatever they have in-house, i.e. make the manifest the
> set of open source repos plus some (e.g. drivers specific to their
> device). How is that sort of workflow going to work where you
> (presumably) have do that via .gitmodules + commit entries in trees?
The manifest is tracked in its own manifest repo, so today they fork
that project, with the superproject you'd fork the superproject and modify
the .gitmodules file as needed.
> They run their own Gerrit install with some magic to sync back & forth?
>
> I assume that now the recursive "checkout" relies on all the origin/HEAD
> symbolic refs pointing to "master", but how is this going to deal with
> incorporating a repo whose main branch has a different name?
Change the name? Or pin it via sha1.
> E.g. "trunk" or "blead"? Perhaps some interaction with
> checkout.defaultRemote + submodule.<name>.branch=<NAME> could make "git
> checkout :mainbranch" DWYM.
>
> > * Fetching changes ("repo sync") needs to fetch all repositories, as there
> > is no central place that tracks what has changed. In a superproject
> > git fetch can determine which submodules need fetching. In Androids
> > case the daily change is only in a few repositories (think 10s), so
> > migrating to a superproject would save an order of magnitude in fetch
> > traffic for daily updates of developers.
>
> Interesting that in all this time with the reliance on a central server
> repo wasn't already asking some custom API "what repos changed since xyz
> time" to narrow that down, but hey, .gitmodules + commit entries in
> trees will do it for you.
It's complicated. Shawn really grew to dislike the repo tool and wanted
to have it gone as quickly as possible, so no hacks that extend its life. ;-)
I would prefer to keep that stance as any hack in there would grow into
a nightmare down the road.
> > * Sometimes when the dependencies are not on a clear repository boundary
> > one would like to have git-bisect available across the different
> > repositories, which repo cannot provide due to its design.
>
> I assume that you're not upgrading independently to e.g. every single
> linux commit, just stable releases, so does bisecting deal with knowing
> that e.g. a breakage occurred when linux.git was updated from v4.10 to
> v4.12, and then to go within the repo itself and bisect from there, or
> is that done manually?
We currently have no submodule support in bisect, but that is open to
future contribution. But with relations between submodules via the
superproject it is at least possible to bisect precisely, whereas in repo
the projects are unrelated to each other and it becomes a political
issue if A broke B or B broke A.
Thanks for taking time to read and think through the concept here,
Stefan
^ permalink raw reply [relevance 6%]
* Re: [PATCH] FYI / RFC: submodules: introduce repo-like workflow
2018-09-27 22:16 19% [PATCH] FYI / RFC: submodules: introduce repo-like workflow Stefan Beller
2018-09-27 22:27 7% ` Jonathan Nieder
@ 2018-09-28 19:26 6% ` Ævar Arnfjörð Bjarmason
2018-09-28 20:23 6% ` Stefan Beller
1 sibling, 1 reply; 200+ results
From: Ævar Arnfjörð Bjarmason @ 2018-09-28 19:26 UTC (permalink / raw)
To: Stefan Beller; +Cc: git, jrnieder
On Thu, Sep 27 2018, Stefan Beller wrote:
> Internally we have rolled out this as an experiment for
> "submodules replacing the repo tool[1]". The repo tool is described as:
>
> Repo unifies Git repositories when necessary, performs uploads to the
> Gerrit revision control system, and automates parts of the Android
> development workflow. Repo is not meant to replace Git, only to make
> it easier to work with Git in the context of Android. The repo command
> is an executable Python script that you can put anywhere in your path.
>
> In working with the Android source files, you use Repo for
> across-network operations. For example, with a single Repo command you
> can download files from multiple repositories into your local working
> directory.
>
> In most situations, you can use Git instead of Repo, or mix Repo and
> Git commands to form complex commands.
>
> [1] https://source.android.com/setup/develop/
Some questions just out of curiosity, not for this patch in particular:
Those docs seem to describe the situation without this patch, with this
patch is the repo tool fully replaced?
How are you planning to migrate from repo to this on a repository-data
basis, does gerrit also populate .gitmodules files appropriately, which
your clone --recurse-submodules will pick up, but repo will just ignore,
so you can use the two in parallel?
Now "repo init -u" takes a URL to a manifest of repositories to stitch
together, I've understood from past conversations (but am not sure) that
this is used e.g. by downstream Android vendors so they can use what
Google's using + whatever they have in-house, i.e. make the manifest the
set of open source repos plus some (e.g. drivers specific to their
device). How is that sort of workflow going to work where you
(presumably) have do that via .gitmodules + commit entries in trees?
They run their own Gerrit install with some magic to sync back & forth?
I assume that now the recursive "checkout" relies on all the origin/HEAD
symbolic refs pointing to "master", but how is this going to deal with
incorporating a repo whose main branch has a different name?
E.g. "trunk" or "blead"? Perhaps some interaction with
checkout.defaultRemote + submodule.<name>.branch=<NAME> could make "git
checkout :mainbranch" DWYM.
> * Fetching changes ("repo sync") needs to fetch all repositories, as there
> is no central place that tracks what has changed. In a superproject
> git fetch can determine which submodules need fetching. In Androids
> case the daily change is only in a few repositories (think 10s), so
> migrating to a superproject would save an order of magnitude in fetch
> traffic for daily updates of developers.
Interesting that in all this time with the reliance on a central server
repo wasn't already asking some custom API "what repos changed since xyz
time" to narrow that down, but hey, .gitmodules + commit entries in
trees will do it for you.
> * Sometimes when the dependencies are not on a clear repository boundary
> one would like to have git-bisect available across the different
> repositories, which repo cannot provide due to its design.
I assume that you're not upgrading independently to e.g. every single
linux commit, just stable releases, so does bisecting deal with knowing
that e.g. a breakage occurred when linux.git was updated from v4.10 to
v4.12, and then to go within the repo itself and bisect from there, or
is that done manually?
^ permalink raw reply [relevance 6%]
* Re: [PATCH] FYI / RFC: submodules: introduce repo-like workflow
2018-09-27 22:27 7% ` Jonathan Nieder
@ 2018-09-28 18:08 2% ` Junio C Hamano
0 siblings, 0 replies; 200+ results
From: Junio C Hamano @ 2018-09-28 18:08 UTC (permalink / raw)
To: Jonathan Nieder; +Cc: Stefan Beller, git
Jonathan Nieder <jrnieder@gmail.com> writes:
> (dropping cc-s to my internal address that I don't use on this list
> and to git-core@google.com which bounces)
> Hi,
>
> Stefan Beller wrote:
>
>> Internally we have rolled out this as an experiment for
>> "submodules replacing the repo tool[1]". The repo tool is described as:
>>
>> Repo unifies Git repositories when necessary, performs uploads to the
>> Gerrit revision control system, and automates parts of the Android
>> development workflow. Repo is not meant to replace Git, only to make
>> it easier to work with Git in the context of Android. The repo command
>> is an executable Python script that you can put anywhere in your path.
> [...]
>> Submodules can also be understood as unifying Git repositories, even more,
>> in the superproject the submodules have relationships between their
>> versions, which the repo tool only provides in releases.
>>
>> The repo tool does not provide these relationships between versions, but
>> all the repositories (in case of Android think of ~1000 git repositories)
>> are put in their place without depending on each other.
>>
>> This comes with a couple of advantages and disadvantages:
>
> Thanks for describing this background.
Thanks for this. I probably won't be reading this before other
topics, but I've often found that changes from google were lacking
the backstory to make sense of them as a coherent whole.
For example, a patch that says "Instead of detaching at the commit
recorded in the superproject, check out the branch with the same
name, even if it points at a different commit. Here is the write up
of what it does in the Documentation/ and code" is hard to judge
beyond checking if the code does what it claims to do and if the
docs describe what the code does---without backstory like this that
talks about how individual small pieces fit within the plan for the
whole thing, judging if that "what it claims to do" is even sensible
is impossible.
^ permalink raw reply [relevance 2%]
* Re: [PATCH] FYI / RFC: submodules: introduce repo-like workflow
2018-09-27 22:16 19% [PATCH] FYI / RFC: submodules: introduce repo-like workflow Stefan Beller
@ 2018-09-27 22:27 7% ` Jonathan Nieder
2018-09-28 18:08 2% ` Junio C Hamano
2018-09-28 19:26 6% ` Ævar Arnfjörð Bjarmason
1 sibling, 1 reply; 200+ results
From: Jonathan Nieder @ 2018-09-27 22:27 UTC (permalink / raw)
To: Stefan Beller; +Cc: git
(dropping cc-s to my internal address that I don't use on this list
and to git-core@google.com which bounces)
Hi,
Stefan Beller wrote:
> Internally we have rolled out this as an experiment for
> "submodules replacing the repo tool[1]". The repo tool is described as:
>
> Repo unifies Git repositories when necessary, performs uploads to the
> Gerrit revision control system, and automates parts of the Android
> development workflow. Repo is not meant to replace Git, only to make
> it easier to work with Git in the context of Android. The repo command
> is an executable Python script that you can put anywhere in your path.
[...]
> Submodules can also be understood as unifying Git repositories, even more,
> in the superproject the submodules have relationships between their
> versions, which the repo tool only provides in releases.
>
> The repo tool does not provide these relationships between versions, but
> all the repositories (in case of Android think of ~1000 git repositories)
> are put in their place without depending on each other.
>
> This comes with a couple of advantages and disadvantages:
Thanks for describing this background.
[...]
> This changes the following commands in the superproject:
>
> checkout -B/-b create branches in subs, too
> checkout (-f): update branch in submodule (create if needed) and check
> it out; Pass the argument down literally if it is a branch
> name (e.g. "checkout -f master" will run a
> "checkout -f master" in the submodule as well)
> clone: see checkout
> reset --hard: see checkout
As you mentioned, I've been using this submodule.repoLike=true mode
for my own use for a while. You did a nice job of explaining on how
it fits into a Gerrit-driven workflow; I'd like to add that I find it
pleasant in non-Gerrit-driven contexts as well.
The primary difference from repoLike=false is that this makes the
normal state to have branches checked out in submodules. For example,
if I run
git checkout --recurse-submodules -B master origin/master
then this will create and check out a "master" branch in all
submodules instead of only in the superproject. This helps avoid some
issues in Git's submodule handling where submodule commits can be
pruned if they have not been checked out in a while because there is
no ref pointing to them.
Some next steps:
- now that we have a repository object, some of the implementation can
be simplified and made more robust. I expect that will also make
these patches easier to review
- also in the direction of reviewability, at that point we may want to
split this into multiple patches
- gitsubmodules.txt and config.txt should describe the new option, to
help new users understand what this new repoLike workflow does
- there are some edge cases in the UX that get... messy that I should
describe in another message
All that said, thanks for sending this out, and I'd be happy to hear
from any interested people --- feedback from anyone adventurous enough
to try this out would be very welcome.
Happy hacking,
Jonathan
^ permalink raw reply [relevance 7%]
* [PATCH] FYI / RFC: submodules: introduce repo-like workflow
@ 2018-09-27 22:16 19% Stefan Beller
2018-09-27 22:27 7% ` Jonathan Nieder
2018-09-28 19:26 6% ` Ævar Arnfjörð Bjarmason
0 siblings, 2 replies; 200+ results
From: Stefan Beller @ 2018-09-27 22:16 UTC (permalink / raw)
To: git, git-core; +Cc: jrnieder, Stefan Beller, Jonathan Nieder
Internally we have rolled out this as an experiment for
"submodules replacing the repo tool[1]". The repo tool is described as:
Repo unifies Git repositories when necessary, performs uploads to the
Gerrit revision control system, and automates parts of the Android
development workflow. Repo is not meant to replace Git, only to make
it easier to work with Git in the context of Android. The repo command
is an executable Python script that you can put anywhere in your path.
In working with the Android source files, you use Repo for
across-network operations. For example, with a single Repo command you
can download files from multiple repositories into your local working
directory.
In most situations, you can use Git instead of Repo, or mix Repo and
Git commands to form complex commands.
[1] https://source.android.com/setup/develop/
Submodules can also be understood as unifying Git repositories, even more,
in the superproject the submodules have relationships between their
versions, which the repo tool only provides in releases.
The repo tool does not provide these relationships between versions, but
all the repositories (in case of Android think of ~1000 git repositories)
are put in their place without depending on each other.
This comes with a couple of advantages and disadvantages:
* Many users are familiar with Git, but not submodules. Each repository
can be used independently with Git and there is no need to update the
superproject or the repo manifest for a change in a repository.
* It is easy to work with repositories with no version-control-dependencies
if there are dependencies in the code. In case of Android the
repositories are bound at natural boundaries. For example the linux
kernel is one repository, as then upstream work is made easy for this
repository. So it is desirable to keep an easy-as-repo workflow.
* Fetching changes ("repo sync") needs to fetch all repositories, as there
is no central place that tracks what has changed. In a superproject
git fetch can determine which submodules need fetching. In Androids
case the daily change is only in a few repositories (think 10s), so
migrating to a superproject would save an order of magnitude in fetch
traffic for daily updates of developers.
* Sometimes when the dependencies are not on a clear repository boundary
one would like to have git-bisect available across the different
repositories, which repo cannot provide due to its design.
Internally we have the Gerrit as a central point, where the source of
truth is found for a given repository.
This patch adds a new mode to submodule handling, where the superproject
controls the existence of the submodule (just as current submodule
handling), but the submodule HEAD is not detached, but following the same
branch name as the superproject.
Current situation visualized:
superproject
HEAD -> "<branch name>" -> OID
|
submodule v
HEAD --------------------> OID
The OID in the submodule is controlled via the HEAD in the submodule that
is set accordingly to the gitlink in the superproject. Confusion can arise
when the (detached) HEAD in the submodule doesn't match the superprojects
gitlink.
This patch visualized:
superproject
HEAD -> "<branch name>" -> OID
|
submodule v
HEAD -> "<branch name>" -> OID
As there is a central point of truth in our setup (our Gerrit installation)
which keeps the superproject and the submodule branches in sync, this
ought to look the same for the user; removing the "detached HEAD" in the
submodule. git-status will still notice if there is an OID mismatch between
the gitlink and the submodules branch, but that is a race condition and
should be caught by Gerrit.
This changes the following commands in the superproject:
checkout -B/-b create branches in subs, too
checkout (-f): update branch in submodule (create if needed) and check
it out; Pass the argument down literally if it is a branch
name (e.g. "checkout -f master" will run a
"checkout -f master" in the submodule as well)
clone: see checkout
reset --hard: see checkout
Change-Id: I69b361e5bd9d57226a0976fd36968cf9aeb52ae0
Signed-off-by: Jonathan Nieder <jrn@google.com>
Signed-off-by: Stefan Beller <sbeller@google.com>
---
Documentation/git-checkout.txt | 13 +++--
Makefile | 1 +
branch.c | 84 +++++++++++++++++++++++++++++++++
builtin.h | 1 +
builtin/branch.c | 35 ++++++++++++++
builtin/checkout.c | 76 ++++++++++++++++++++++++------
builtin/clone.c | 12 ++++-
builtin/reset.c | 41 ++++++++++++++--
entry.c | 43 +++++++++++++----
git-submodule.sh | 24 +++++++++-
git.c | 1 +
submodule-move-head.c | 81 ++++++++++++++++++++++++++++++++
submodule-move-head.h | 22 +++++++++
submodule.c | 14 ++++++
submodule.h | 9 +++-
t/lib-submodule-update.sh | 86 +++++++++++++++++++++++++++++++++-
t/t1013-read-tree-submodule.sh | 1 +
t/t2013-checkout-submodule.sh | 4 ++
t/t7112-reset-submodule.sh | 5 ++
unpack-trees.c | 20 ++++++--
unpack-trees.h | 10 ++++
21 files changed, 538 insertions(+), 45 deletions(-)
create mode 100644 submodule-move-head.c
create mode 100644 submodule-move-head.h
diff --git a/Documentation/git-checkout.txt b/Documentation/git-checkout.txt
index 9db02928c4..19b0c2a272 100644
--- a/Documentation/git-checkout.txt
+++ b/Documentation/git-checkout.txt
@@ -268,13 +268,12 @@ section of linkgit:git-add[1] to learn how to operate the `--patch` mode.
worktree.
--[no-]recurse-submodules::
- Using --recurse-submodules will update the content of all initialized
- submodules according to the commit recorded in the superproject. If
- local modifications in a submodule would be overwritten the checkout
- will fail unless `-f` is used. If nothing (or --no-recurse-submodules)
- is used, the work trees of submodules will not be updated.
- Just like linkgit:git-submodule[1], this will detach the
- submodules HEAD.
+ Using --recurse-submodules will update the content and current
+ branch of all initialized submodules in addition to the
+ superproject. If local modifications in a submodule would be
+ overwritten, the checkout will fail unless `-f` is used. If
+ nothing (or --no-recurse-submodules) is used, the work trees of
+ submodules will not be updated.
<branch>::
Branch to checkout; if it refers to a branch (i.e., a name that,
diff --git a/Makefile b/Makefile
index 13e1c52478..6824dbcccf 100644
--- a/Makefile
+++ b/Makefile
@@ -975,6 +975,7 @@ LIB_OBJS += streaming.o
LIB_OBJS += string-list.o
LIB_OBJS += submodule.o
LIB_OBJS += submodule-config.o
+LIB_OBJS += submodule-move-head.o
LIB_OBJS += sub-process.o
LIB_OBJS += symlinks.o
LIB_OBJS += tag.o
diff --git a/branch.c b/branch.c
index 776f55fc66..e4538f2b36 100644
--- a/branch.c
+++ b/branch.c
@@ -1,9 +1,13 @@
#include "git-compat-util.h"
#include "cache.h"
#include "config.h"
+#include "repository.h"
+#include "submodule.h"
#include "branch.h"
#include "refs.h"
#include "refspec.h"
+#include "tree-walk.h"
+#include "run-command.h"
#include "remote.h"
#include "commit.h"
#include "worktree.h"
@@ -242,6 +246,76 @@ N_("\n"
"will track its remote counterpart, you may want to use\n"
"\"git push -u\" to set the upstream config as you push.");
+static void create_branch_in_submodule(const char *name, const char *start_name,
+ const char *start_ref, const struct object_id *start_oid,
+ int force, int reflog, int clobber_head,
+ int quiet, enum branch_track track,
+ int checking_out_branch,
+ const char *sub_path, const struct object_id *entry_oid)
+{
+ struct child_process cp = CHILD_PROCESS_INIT;
+
+ trace_printf("create_branch_in_submodule %s", sub_path);
+
+ prepare_submodule_repo_env(&cp.env_array);
+ cp.git_cmd = 1;
+ cp.dir = sub_path;
+ argv_array_push(&cp.args, "branch--helper");
+ argv_array_pushf(&cp.args, "--name=%s", name);
+ if (checking_out_branch) {
+ argv_array_pushf(&cp.args, "--start_name=%s",
+ start_ref);
+ } else {
+ argv_array_pushf(&cp.args, "--start_name=%s",
+ oid_to_hex(entry_oid));
+ }
+ argv_array_pushf(&cp.args, "--force=%d", force);
+ argv_array_pushf(&cp.args, "--reflog=%d", reflog);
+ argv_array_pushf(&cp.args, "--clobber_head=%d", clobber_head);
+ argv_array_pushf(&cp.args, "--quiet=%d", quiet);
+ argv_array_pushf(&cp.args, "--track=%d", track);
+
+ if (run_command(&cp))
+ fprintf(stderr, "process for submodule '%s' failed", sub_path);
+ child_process_clear(&cp);
+}
+
+static void create_branch_in_submodules(const char *name, const char *start_name,
+ const char *start_ref, const struct object_id *start_oid,
+ int force, int reflog, int clobber_head,
+ int quiet, enum branch_track track, struct strbuf *rec_path)
+{
+
+ int checking_out_branch = start_ref && starts_with(start_ref, "refs/heads/");
+ void *buf;
+ struct tree_desc tree;
+ struct name_entry entry;
+ int rec_path_len = rec_path->len;
+
+ buf = fill_tree_descriptor(&tree, start_oid);
+ if (!buf)
+ die("could not read %s for checkout", start_name);
+
+ while (tree_entry(&tree, &entry)) {
+
+ if (rec_path->len > 0)
+ strbuf_addch(rec_path, '/');
+ strbuf_addstr(rec_path, entry.path);
+
+ trace_printf("create_branch_in_submodules %s %o", rec_path->buf, entry.mode);
+
+ if (S_ISGITLINK(entry.mode) && is_submodule_active(the_repository, rec_path->buf))
+ create_branch_in_submodule(name, start_name, start_ref,
+ start_oid, force, reflog, clobber_head,
+ quiet, track, checking_out_branch, rec_path->buf, entry.oid);
+ else if (S_ISDIR(entry.mode)) {
+ create_branch_in_submodules(name, start_name, start_ref, entry.oid, force, reflog, clobber_head, quiet, track, rec_path);
+ }
+ strbuf_setlen(rec_path, rec_path_len);
+ }
+ free(buf);
+}
+
void create_branch(const char *name, const char *start_name,
int force, int clobber_head_ok, int reflog,
int quiet, enum branch_track track)
@@ -250,6 +324,7 @@ void create_branch(const char *name, const char *start_name,
struct object_id oid;
char *real_ref;
struct strbuf ref = STRBUF_INIT;
+ struct strbuf sub_path = STRBUF_INIT;
int forcing = 0;
int dont_change_ref = 0;
int explicit_tracking = 0;
@@ -307,6 +382,14 @@ void create_branch(const char *name, const char *start_name,
if (reflog)
log_all_ref_updates = LOG_REFS_NORMAL;
+ /*
+ * NEEDSWORK: Doesn't handle errors part-way through very well.
+ */
+ trace_printf("create_branch need to update subs: %d", should_update_submodules());
+ if (behave_google_repo_like() && should_update_submodules())
+ create_branch_in_submodules(name, start_name, real_ref, &oid,
+ force, reflog, clobber_head_ok, quiet, track, &sub_path);
+
if (!dont_change_ref) {
struct ref_transaction *transaction;
struct strbuf err = STRBUF_INIT;
@@ -333,6 +416,7 @@ void create_branch(const char *name, const char *start_name,
setup_tracking(ref.buf + 11, real_ref, track, quiet);
strbuf_release(&ref);
+ strbuf_release(&sub_path);
free(real_ref);
}
diff --git a/builtin.h b/builtin.h
index 962f0489ab..31030f248f 100644
--- a/builtin.h
+++ b/builtin.h
@@ -135,6 +135,7 @@ extern int cmd_archive(int argc, const char **argv, const char *prefix);
extern int cmd_bisect__helper(int argc, const char **argv, const char *prefix);
extern int cmd_blame(int argc, const char **argv, const char *prefix);
extern int cmd_branch(int argc, const char **argv, const char *prefix);
+extern int cmd_branch_helper(int argc, const char **argv, const char *prefix);
extern int cmd_bundle(int argc, const char **argv, const char *prefix);
extern int cmd_cat_file(int argc, const char **argv, const char *prefix);
extern int cmd_checkout(int argc, const char **argv, const char *prefix);
diff --git a/builtin/branch.c b/builtin/branch.c
index c396c41533..b6db7ec5bf 100644
--- a/builtin/branch.c
+++ b/builtin/branch.c
@@ -36,6 +36,11 @@ static const char * const builtin_branch_usage[] = {
NULL
};
+static const char * const builtin_branch_helper_usage[] = {
+ N_("git branch--helper"),
+ NULL
+};
+
static const char *head;
static struct object_id head_oid;
@@ -578,6 +583,36 @@ static int edit_branch_description(const char *branch_name)
return 0;
}
+int cmd_branch_helper(int argc, const char **argv, const char *prefix)
+{
+ const char *name = NULL, *start_name = NULL;
+ int force = 0, reflog = 0, clobber_head = 0, quiet = 0;
+ enum branch_track track = BRANCH_TRACK_NEVER;
+
+ struct option options[] = {
+ OPT_STRING(0, "name", &name, N_(""), N_("")),
+ OPT_STRING(0, "start_name", &start_name, N_(""), N_("")),
+
+ OPT_INTEGER(0, "force", &force, N_("")),
+ OPT_INTEGER(0, "reflog", &reflog, N_("")),
+ OPT_INTEGER(0, "clobber_head", &clobber_head, N_("")),
+ OPT_INTEGER(0, "quiet", &quiet, N_("")),
+
+ /* implicit int -> enum conversion */
+ OPT_INTEGER(0, "track", &track, N_("")),
+ OPT_END(),
+ };
+
+ argc = parse_options(argc, argv, prefix, options,
+ builtin_branch_helper_usage, 0);
+ if (argc > 0)
+ die (_("branchhelper doesn't know about %s"), argv[0]);
+
+ create_branch(name, start_name, force, clobber_head, reflog,
+ quiet, track);
+ return 0;
+}
+
int cmd_branch(int argc, const char **argv, const char *prefix)
{
int delete = 0, rename = 0, copy = 0, force = 0, list = 0;
diff --git a/builtin/checkout.c b/builtin/checkout.c
index b30b48767e..054d0ef60e 100644
--- a/builtin/checkout.c
+++ b/builtin/checkout.c
@@ -3,6 +3,7 @@
#include "checkout.h"
#include "lockfile.h"
#include "parse-options.h"
+#include "repository.h"
#include "refs.h"
#include "object-store.h"
#include "commit.h"
@@ -22,6 +23,7 @@
#include "ll-merge.h"
#include "resolve-undo.h"
#include "submodule-config.h"
+#include "submodule-move-head.h"
#include "submodule.h"
#include "advice.h"
@@ -421,11 +423,24 @@ static void describe_detached_head(const char *msg, struct commit *commit)
strbuf_release(&sb);
}
-static int reset_tree(struct tree *tree, const struct checkout_opts *o,
+struct branch_info {
+ const char *name; /* The short name used */
+ const char *path; /* The full name of a real branch */
+ struct commit *commit; /* The named commit */
+ /*
+ * if not null the branch is detached because it's already
+ * checked out in this checkout
+ */
+ char *checkout;
+};
+
+static int reset_tree(struct branch_info *b, const struct checkout_opts *o,
int worktree, int *writeout_error)
{
struct unpack_trees_options opts;
+ struct submodule_move_head_options move_head_opts;
struct tree_desc tree_desc;
+ struct tree *tree = get_commit_tree(b->commit);
memset(&opts, 0, sizeof(opts));
opts.head_idx = -1;
@@ -437,6 +452,16 @@ static int reset_tree(struct tree *tree, const struct checkout_opts *o,
opts.verbose_update = o->show_progress;
opts.src_index = &the_index;
opts.dst_index = &the_index;
+
+ if (behave_google_repo_like()) {
+ opts.move_head = unpack_trees_move_head;
+ memset(&move_head_opts, 0, sizeof(move_head_opts));
+ move_head_opts.force = 1;
+ move_head_opts.new_ref = b->path;
+ move_head_opts.target_ref = o->force_detach ? NULL : b->path;
+ opts.unpack_data = &move_head_opts;
+ }
+
parse_tree(tree);
init_tree_desc(&tree_desc, tree->buffer, tree->size);
switch (unpack_trees(1, &tree_desc, &opts)) {
@@ -456,17 +481,6 @@ static int reset_tree(struct tree *tree, const struct checkout_opts *o,
}
}
-struct branch_info {
- const char *name; /* The short name used */
- const char *path; /* The full name of a real branch */
- struct commit *commit; /* The named commit */
- /*
- * if not null the branch is detached because it's already
- * checked out in this checkout
- */
- char *checkout;
-};
-
static void setup_branch_path(struct branch_info *branch)
{
struct strbuf buf = STRBUF_INIT;
@@ -584,7 +598,7 @@ static int merge_working_tree(const struct checkout_opts *opts,
resolve_undo_clear();
if (opts->force) {
- ret = reset_tree(get_commit_tree(new_branch_info->commit),
+ ret = reset_tree(new_branch_info,
opts, 1, writeout_error);
if (ret)
return ret;
@@ -592,6 +606,7 @@ static int merge_working_tree(const struct checkout_opts *opts,
struct tree_desc trees[2];
struct tree *tree;
struct unpack_trees_options topts;
+ struct submodule_move_head_options mopts;
memset(&topts, 0, sizeof(topts));
topts.head_idx = -1;
@@ -607,6 +622,16 @@ static int merge_working_tree(const struct checkout_opts *opts,
return 1;
}
+ if (behave_google_repo_like()) {
+ topts.move_head = unpack_trees_move_head;
+ memset(&mopts, 0, sizeof(mopts));
+ mopts.old_ref = old_branch_info->path;
+ mopts.new_ref = new_branch_info->path;
+ mopts.target_ref = opts->force_detach
+ ? NULL : new_branch_info->path;
+ topts.unpack_data = &mopts;
+ }
+
/* 2-way merge to the new branch */
topts.initial_checkout = is_cache_unborn();
topts.update = 1;
@@ -672,7 +697,7 @@ static int merge_working_tree(const struct checkout_opts *opts,
o.verbosity = 0;
work = write_tree_from_memory(&o);
- ret = reset_tree(get_commit_tree(new_branch_info->commit),
+ ret = reset_tree(new_branch_info,
opts, 1,
writeout_error);
if (ret)
@@ -687,7 +712,7 @@ static int merge_working_tree(const struct checkout_opts *opts,
&result);
if (ret < 0)
exit(128);
- ret = reset_tree(get_commit_tree(new_branch_info->commit),
+ ret = reset_tree(new_branch_info,
opts, 0,
writeout_error);
strbuf_release(&o.obuf);
@@ -722,6 +747,18 @@ static void report_tracking(struct branch_info *new_branch_info)
strbuf_release(&sb);
}
+static void create_symref_in_submodules(const char *symref, const char *target, const char *logmsg)
+{
+ int i = 0;
+ for (i = 0; i < active_nr; i++) {
+ const struct cache_entry *ce = active_cache[i];
+ if (!S_ISGITLINK(ce->ce_mode) || !is_submodule_active(the_repository, ce->name))
+ continue;
+
+ create_symref_in_submodule(ce->name, symref, target, logmsg);
+ }
+}
+
static void update_refs_for_switch(const struct checkout_opts *opts,
struct branch_info *old_branch_info,
struct branch_info *new_branch_info)
@@ -732,6 +769,9 @@ static void update_refs_for_switch(const struct checkout_opts *opts,
if (opts->new_orphan_branch) {
char *refname;
+ if (should_update_submodules())
+ die("--orphan --recurse-submodules is not implemented");
+
refname = mkpathdup("refs/heads/%s", opts->new_orphan_branch);
if (opts->new_branch_log &&
!should_autocreate_reflog(refname)) {
@@ -784,6 +824,12 @@ static void update_refs_for_switch(const struct checkout_opts *opts,
describe_detached_head(_("HEAD is now at"), new_branch_info->commit);
}
} else if (new_branch_info->path) { /* Switch branches. */
+ /*
+ * NEEDSWORK: We don't handle attachment on checkout <branch>
+ * yet.
+ */
+ if (opts->new_branch && should_update_submodules() && behave_google_repo_like())
+ create_symref_in_submodules("HEAD", new_branch_info->path, msg.buf);
if (create_symref("HEAD", new_branch_info->path, msg.buf) < 0)
die(_("unable to update HEAD"));
if (!opts->quiet) {
diff --git a/builtin/clone.c b/builtin/clone.c
index 15b142d646..d5936ea1c7 100644
--- a/builtin/clone.c
+++ b/builtin/clone.c
@@ -28,6 +28,7 @@
#include "run-command.h"
#include "connected.h"
#include "packfile.h"
+#include "submodule.h"
#include "list-objects-filter-options.h"
#include "object-store.h"
@@ -738,7 +739,6 @@ static int checkout(int submodule_progress)
if (!starts_with(head, "refs/heads/"))
die(_("HEAD not found below refs/heads!"));
}
- free(head);
/* We need to be in the new work tree for the checkout */
setup_work_tree();
@@ -767,8 +767,17 @@ static int checkout(int submodule_progress)
oid_to_hex(&oid), "1", NULL);
if (!err && (option_recurse_submodules.nr > 0)) {
+ const char *branch;
struct argv_array args = ARGV_ARRAY_INIT;
argv_array_pushl(&args, "submodule", "update", "--init", "--recursive", NULL);
+ if (behave_google_repo_like()) {
+ if (!strcmp(head, "HEAD"))
+ ; /* detach HEAD in submodules, too. */
+ else if (skip_prefix(head, "refs/heads/", &branch))
+ argv_array_pushl(&args, "--checkout-branch", branch, NULL);
+ else
+ BUG("HEAD not found below refs/heads!");
+ }
if (option_shallow_submodules == 1)
argv_array_push(&args, "--depth=1");
@@ -786,6 +795,7 @@ static int checkout(int submodule_progress)
argv_array_clear(&args);
}
+ free(head);
return err;
}
diff --git a/builtin/reset.c b/builtin/reset.c
index 11cd0dcb8c..9cc8c4db52 100644
--- a/builtin/reset.c
+++ b/builtin/reset.c
@@ -24,6 +24,7 @@
#include "cache-tree.h"
#include "submodule.h"
#include "submodule-config.h"
+#include "submodule-move-head.h"
static const char * const git_reset_usage[] = {
N_("git reset [--mixed | --soft | --hard | --merge | --keep] [-q] [<commit>]"),
@@ -42,12 +43,16 @@ static inline int is_merge(void)
return !access(git_path_merge_head(the_repository), F_OK);
}
-static int reset_index(const struct object_id *oid, int reset_type, int quiet)
+static int reset_index(const char *rev, const struct object_id *oid, int reset_type, int quiet)
{
- int i, nr = 0;
+ int i, nr = 0, flags = 0;
struct tree_desc desc[2];
struct tree *tree;
struct unpack_trees_options opts;
+ struct submodule_move_head_options mopts;
+ char *current_branch = NULL;
+ struct object_id discard;
+ char *new_ref = NULL;
int ret = -1;
memset(&opts, 0, sizeof(opts));
@@ -65,6 +70,31 @@ static int reset_index(const struct object_id *oid, int reset_type, int quiet)
break;
case HARD:
opts.update = 1;
+
+ if (behave_google_repo_like()) {
+ /*
+ * Submodule handling:
+ * - unless we are detached, attach HEAD in submodules
+ * - if rev is a branch name, use that branch instead of oid in
+ * submodules.
+ */
+ current_branch = resolve_refdup("HEAD", 0, NULL, &flags);
+ if (!(flags & REF_ISSYMREF))
+ current_branch = NULL;
+ if (dwim_ref(rev, strlen(rev), &discard, &new_ref) != 1 ||
+ !starts_with(new_ref, "refs/heads/")) {
+ free(new_ref);
+ new_ref = NULL;
+ }
+
+ opts.move_head = unpack_trees_move_head;
+
+ memset(&mopts, 0, sizeof(mopts));
+ mopts.force = 1;
+ mopts.new_ref = new_ref;
+ mopts.target_ref = current_branch;
+ opts.unpack_data = &mopts;
+ }
/* fallthrough */
default:
opts.reset = 1;
@@ -101,6 +131,8 @@ static int reset_index(const struct object_id *oid, int reset_type, int quiet)
out:
for (i = 0; i < nr; i++)
free((void *)desc[i].buffer);
+ free(current_branch);
+ free(new_ref);
return ret;
}
@@ -317,6 +349,7 @@ int cmd_reset(int argc, const char **argv, const char *prefix)
oidcpy(&oid, the_hash_algo->empty_tree);
} else if (!pathspec.nr) {
struct commit *commit;
+
if (get_oid_committish(rev, &oid))
die(_("Failed to resolve '%s' as a valid revision."), rev);
commit = lookup_commit_reference(the_repository, &oid);
@@ -379,9 +412,9 @@ int cmd_reset(int argc, const char **argv, const char *prefix)
refresh_index(&the_index, flags, NULL, NULL,
_("Unstaged changes after reset:"));
} else {
- int err = reset_index(&oid, reset_type, quiet);
+ int err = reset_index(rev, &oid, reset_type, quiet);
if (reset_type == KEEP && !err)
- err = reset_index(&oid, MIXED, quiet);
+ err = reset_index(rev, &oid, MIXED, quiet);
if (err)
die(_("Could not reset index file to revision '%s'."), rev);
}
diff --git a/entry.c b/entry.c
index 5d136c5d55..eee5bc498e 100644
--- a/entry.c
+++ b/entry.c
@@ -2,11 +2,23 @@
#include "blob.h"
#include "object-store.h"
#include "dir.h"
+#include "unpack-trees.h"
#include "streaming.h"
#include "submodule.h"
#include "progress.h"
#include "fsmonitor.h"
+/* NEEDSWORK: share code with unpack-trees.c */
+static int move_head(const struct unpack_trees_options *o, const char *path, const char *old, const char *new, unsigned flags)
+{
+ if (behave_google_repo_like()) {
+ if (!o || !o->move_head)
+ return submodule_move_head(path, old, new, flags);
+ return o->move_head(o, path, old, new, flags);
+ } else
+ return submodule_move_head(path, old, new, flags);
+}
+
static void create_directories(const char *path, int path_len,
const struct checkout *state)
{
@@ -251,7 +263,7 @@ int finish_delayed_checkout(struct checkout *state)
return errs;
}
-static int write_entry(struct cache_entry *ce,
+static int write_entry(struct unpack_trees_options *o, struct cache_entry *ce,
char *path, const struct checkout *state, int to_tempfile)
{
unsigned int ce_mode_s_ifmt = ce->ce_mode & S_IFMT;
@@ -357,7 +369,7 @@ static int write_entry(struct cache_entry *ce,
return error("cannot create submodule directory %s", path);
sub = submodule_from_ce(ce);
if (sub)
- return submodule_move_head(ce->name,
+ return move_head(o, ce->name,
NULL, oid_to_hex(&ce->oid),
state->force ? SUBMODULE_MOVE_HEAD_FORCE : 0);
break;
@@ -435,14 +447,15 @@ static void mark_colliding_entries(const struct checkout *state,
* its name is returned in topath[], which must be able to hold at
* least TEMPORARY_FILENAME_LENGTH bytes long.
*/
-int checkout_entry(struct cache_entry *ce,
- const struct checkout *state, char *topath)
+int unpack_trees_checkout_entry(struct unpack_trees_options *o,
+ struct cache_entry *ce,
+ const struct checkout *state, char *topath)
{
static struct strbuf path = STRBUF_INIT;
struct stat st;
if (topath)
- return write_entry(ce, topath, state, 1);
+ return write_entry(o, ce, topath, state, 1);
strbuf_reset(&path);
strbuf_add(&path, state->base_dir, state->base_dir_len);
@@ -466,10 +479,10 @@ int checkout_entry(struct cache_entry *ce,
if (!(st.st_mode & S_IFDIR))
unlink_or_warn(ce->name);
- return submodule_move_head(ce->name,
+ return move_head(o, ce->name,
NULL, oid_to_hex(&ce->oid), 0);
} else
- return submodule_move_head(ce->name,
+ return move_head(o, ce->name,
"HEAD", oid_to_hex(&ce->oid),
state->force ? SUBMODULE_MOVE_HEAD_FORCE : 0);
}
@@ -506,5 +519,19 @@ int checkout_entry(struct cache_entry *ce,
return 0;
create_directories(path.buf, path.len, state);
- return write_entry(ce, path.buf, state, 0);
+ return write_entry(o, ce, path.buf, state, 0);
+}
+
+/*
+ * Write the contents from ce out to the working tree.
+ *
+ * When topath[] is not NULL, instead of writing to the working tree
+ * file named by ce, a temporary file is created by this function and
+ * its name is returned in topath[], which must be able to hold at
+ * least TEMPORARY_FILENAME_LENGTH bytes long.
+ */
+int checkout_entry(struct cache_entry *ce,
+ const struct checkout *state, char *topath)
+{
+ return unpack_trees_checkout_entry(NULL, ce, state, topath);
}
diff --git a/git-submodule.sh b/git-submodule.sh
index 1b568e29b9..5e6fcbe06e 100755
--- a/git-submodule.sh
+++ b/git-submodule.sh
@@ -424,6 +424,18 @@ fetch_in_submodule () (
esac
)
+# usage: checkout_in_submoodule "$sm_path" "$command" "${branch:-}" "$rev"
+checkout_in_submodule () (
+ sanitize_submodule_env &&
+ cd "$1" &&
+ if test -n "$3"
+ then
+ $2 -B "$3" "$4"
+ else
+ $2 "$4"
+ fi
+)
+
#
# Update each submodule path to correct revision, using clone and checkout as needed
#
@@ -476,6 +488,11 @@ cmd_update()
--recursive)
recursive=1
;;
+ --checkout-branch)
+ update="checkout"
+ checkout_dest=$2
+ shift
+ ;;
--checkout)
update="checkout"
;;
@@ -592,6 +609,11 @@ cmd_update()
die "$(eval_gettext "Fetched in submodule path '\$displaypath', but it did not contain \$sha1. Direct fetching of that commit failed.")"
fi
+ if test -n "$checkout_dest" && test "$update_module" != checkout
+ then
+ die "Cannot use --checkout-branch with update mode '$update_module'"
+ fi
+
must_die_on_failure=
case "$update_module" in
checkout)
@@ -621,7 +643,7 @@ cmd_update()
die "$(eval_gettext "Invalid update mode '$update_module' for submodule path '$path'")"
esac
- if (sanitize_submodule_env; cd "$sm_path" && $command "$sha1")
+ if checkout_in_submodule "$sm_path" "$command" "$checkout_dest" "$sha1"
then
say "$say_msg"
elif test -n "$must_die_on_failure"
diff --git a/git.c b/git.c
index a6f4b44af5..3b9c112a36 100644
--- a/git.c
+++ b/git.c
@@ -447,6 +447,7 @@ static struct cmd_struct commands[] = {
{ "bisect--helper", cmd_bisect__helper, RUN_SETUP },
{ "blame", cmd_blame, RUN_SETUP },
{ "branch", cmd_branch, RUN_SETUP | DELAY_PAGER_CONFIG },
+ { "branch--helper", cmd_branch_helper, RUN_SETUP },
{ "bundle", cmd_bundle, RUN_SETUP_GENTLY | NO_PARSEOPT },
{ "cat-file", cmd_cat_file, RUN_SETUP },
{ "check-attr", cmd_check_attr, RUN_SETUP },
diff --git a/submodule-move-head.c b/submodule-move-head.c
new file mode 100644
index 0000000000..4c8f49066f
--- /dev/null
+++ b/submodule-move-head.c
@@ -0,0 +1,81 @@
+#include "cache.h"
+#include "submodule-move-head.h"
+#include "repository.h"
+#include "submodule.h"
+#include "refs.h"
+#include "unpack-trees.h"
+#include "run-command.h"
+
+void create_symref_in_submodule(const char *path, const char *symref, const char *target, const char *logmsg)
+{
+ struct child_process cp = CHILD_PROCESS_INIT;
+
+ /* NEEDSWORK: What about sub-submodules? */
+ prepare_submodule_repo_env(&cp.env_array);
+ cp.git_cmd = 1;
+ cp.dir = path;
+ argv_array_pushl(&cp.args, "symbolic-ref", "-m", logmsg, symref, target, NULL);
+
+ if (run_command(&cp))
+ die("process for submodule '%s' failed", path);
+}
+
+static void create_ref_in_submodule(const char *path, const char *ref, const char *value)
+{
+ struct child_process cp = CHILD_PROCESS_INIT;
+
+ /* NEEDSWORK: set a reasonable reflog message. */
+ prepare_submodule_repo_env(&cp.env_array);
+ cp.git_cmd = 1;
+ cp.dir = path;
+ argv_array_pushl(&cp.args, "update-ref", ref, value, sha1_to_hex(null_sha1), NULL);
+
+ if (run_command(&cp))
+ die("process for submodule '%s' failed", path);
+}
+
+static int ref_exists_in_submodule(const char *submodule_path, const char *refname)
+{
+ struct ref_store *refs = get_submodule_ref_store(submodule_path);
+ if (!refs)
+ return 0;
+ return refs_resolve_ref_unsafe(refs, refname, RESOLVE_REF_READING, NULL, NULL) != NULL;
+}
+
+int unpack_trees_move_head(const struct unpack_trees_options *opt, const char *path, const char *old, const char *new, unsigned flags)
+{
+ struct submodule_move_head_options *o = opt->unpack_data;
+ const char *new_ref = o->new_ref;
+ const char *target_ref = o->target_ref;
+ const char *old_commit = old;
+ const char *new_commit = new;
+
+ /*
+ * NEEDSWORK:
+ * - set log message
+ * - what about sub-submodules?
+ */
+
+ if (!is_submodule_active(the_repository, path))
+ return 0;
+
+ if (old) {
+ if (o->force)
+ old_commit = "HEAD";
+ else if (o->old_ref && ref_exists_in_submodule(path, o->old_ref))
+ old_commit = o->old_ref;
+ }
+ if (new_ref && new && ref_exists_in_submodule(path, new_ref))
+ new_commit = new_ref;
+
+ if (target_ref)
+ flags |= SUBMODULE_MOVE_HEAD_SKIP_REF_UPDATE;
+ if (submodule_move_head(path, old_commit, new_commit, flags) < 0)
+ return -1;
+ if (new && target_ref && !(flags & SUBMODULE_MOVE_HEAD_DRY_RUN)) {
+ if (!ref_exists_in_submodule(path, target_ref))
+ create_ref_in_submodule(path, target_ref, new);
+ create_symref_in_submodule(path, "HEAD", target_ref, "msg");
+ }
+ return 0;
+}
diff --git a/submodule-move-head.h b/submodule-move-head.h
new file mode 100644
index 0000000000..80e2679038
--- /dev/null
+++ b/submodule-move-head.h
@@ -0,0 +1,22 @@
+#ifndef SUBMODULE_MOVE_HEAD_H
+#define SUBMODULE_MOVE_HEAD_H
+
+struct unpack_trees_options;
+
+/* NEEDSWORK: document */
+struct submodule_move_head_options {
+ int force;
+ const char *old_ref;
+ const char *new_ref;
+ const char *target_ref;
+};
+
+/*
+ * For use as unpack_trees_options.move_head. Parameters should be a
+ * struct submodule_move_head_options * in unpack_trees_options.unpack_data.
+ */
+extern int unpack_trees_move_head(const struct unpack_trees_options *opt, const char *path, const char *old, const char *new, unsigned flags);
+
+extern void create_symref_in_submodule(const char *path, const char *symref, const char *target, const char *logmsg);
+
+#endif /* SUBMODULE_MOVE_HEAD_H */
diff --git a/submodule.c b/submodule.c
index b53cb6e9c4..81eace805f 100644
--- a/submodule.c
+++ b/submodule.c
@@ -30,6 +30,17 @@ static int initialized_fetch_ref_tips;
static struct oid_array ref_tips_before_fetch;
static struct oid_array ref_tips_after_fetch;
+int behave_google_repo_like(void)
+{
+ static int google_repo_like = -1;
+
+ if (google_repo_like == -1)
+ git_config_get_bool("submodule.repolike", &google_repo_like);
+
+ return google_repo_like;
+}
+
+
/*
* Check if the .gitmodules file is unmerged. Parsing of the .gitmodules file
* will be disabled because we can't guess what might be configured in
@@ -1676,6 +1687,9 @@ int submodule_move_head(const char *path,
if (!(flags & SUBMODULE_MOVE_HEAD_DRY_RUN)) {
if (new_head) {
+ if (flags & SUBMODULE_MOVE_HEAD_SKIP_REF_UPDATE)
+ goto out;
+
child_process_init(&cp);
/* also set the HEAD accordingly */
cp.git_cmd = 1;
diff --git a/submodule.h b/submodule.h
index e452919aa4..3346b1a12e 100644
--- a/submodule.h
+++ b/submodule.h
@@ -120,8 +120,11 @@ int push_unpushed_submodules(struct oid_array *commits,
*/
int submodule_to_gitdir(struct strbuf *buf, const char *submodule);
-#define SUBMODULE_MOVE_HEAD_DRY_RUN (1<<0)
-#define SUBMODULE_MOVE_HEAD_FORCE (1<<1)
+#define SUBMODULE_MOVE_HEAD_DRY_RUN (1<<0)
+#define SUBMODULE_MOVE_HEAD_FORCE (1<<1)
+#define SUBMODULE_MOVE_HEAD_SKIP_REF_UPDATE (1<<2)
+
+/* NULL values for old or new represent the "missing" state. */
int submodule_move_head(const char *path,
const char *old,
const char *new_head,
@@ -146,4 +149,6 @@ void absorb_git_dir_into_superproject(const char *prefix,
*/
const char *get_superproject_working_tree(void);
+int behave_google_repo_like(void);
+
#endif
diff --git a/t/lib-submodule-update.sh b/t/lib-submodule-update.sh
index 016391723c..71d5b506de 100755
--- a/t/lib-submodule-update.sh
+++ b/t/lib-submodule-update.sh
@@ -657,7 +657,54 @@ test_submodule_recursing_with_args_common() {
test_submodule_content sub1 origin/add_sub1
)
'
- test_expect_success "$command: submodule branch is not changed, detach HEAD instead" '
+
+ if test "$KNOWN_FAILURE_SUBMODULE_REFS_NOT_UPDATED" = 1
+ then
+ RESULT=failure
+ else
+ RESULT=success
+ fi
+ if test "$KNOWN_DIFFERENCE_SUBMODULE_REFS_NOT_UPDATED" = "read-tree"
+ then
+ test_expect_$RESULT "$command: submodule branch is not changed, detach HEAD" '
+ prolog &&
+ reset_work_tree_to_interested add_sub1 &&
+ (
+ cd submodule_update &&
+ git -C sub1 checkout -b keep_branch &&
+ git -C sub1 rev-parse HEAD >expect &&
+ git branch -t modify_sub1 origin/modify_sub1 &&
+ $command modify_sub1 &&
+ test_superproject_content origin/modify_sub1 &&
+ test_submodule_content sub1 origin/modify_sub1 &&
+ git -C sub1 rev-parse keep_branch >actual &&
+ test_cmp expect actual &&
+ test_must_fail git -C sub1 symbolic-ref HEAD >actual
+ )
+ '
+ elif test "$KNOWN_DIFFERENCE_SUBMODULE_REFS_NOT_UPDATED" = "checkout -B current"
+ then
+ test_expect_$RESULT "$command: submodule branch is changed to 'current'" '
+ prolog &&
+ reset_work_tree_to_interested add_sub1 &&
+ (
+ cd submodule_update &&
+ git -C sub1 checkout -b keep_branch &&
+ git -C sub1 rev-parse HEAD >expect &&
+ git branch -t modify_sub1 origin/modify_sub1 &&
+ $command modify_sub1 &&
+ test_superproject_content origin/modify_sub1 &&
+ test_submodule_content sub1 origin/modify_sub1 &&
+ git -C sub1 rev-parse keep_branch >actual &&
+ test_cmp expect actual &&
+ echo refs/heads/current >expect &&
+ git -C sub1 symbolic-ref HEAD >actual &&
+ test_cmp expect actual
+ )
+ '
+ elif test "$KNOWN_DIFFERENCE_SUBMODULE_REFS_NOT_UPDATED" = "reset"
+ then
+ test_expect_$RESULT "$command: submodule branch is changed to superproject, resetting to target" '
prolog &&
reset_work_tree_to_interested add_sub1 &&
(
@@ -665,14 +712,49 @@ test_submodule_recursing_with_args_common() {
git -C sub1 checkout -b keep_branch &&
git -C sub1 rev-parse HEAD >expect &&
git branch -t modify_sub1 origin/modify_sub1 &&
+ git checkout -b newbranch &&
$command modify_sub1 &&
+
+ # we modified the wt
test_superproject_content origin/modify_sub1 &&
test_submodule_content sub1 origin/modify_sub1 &&
+
+ # keep_branch does not change
git -C sub1 rev-parse keep_branch >actual &&
test_cmp expect actual &&
- test_must_fail git -C sub1 symbolic-ref HEAD
+
+ # the submodule is attached to the same branch as the superproject
+ git -C sub1 symbolic-ref HEAD >actual &&
+ echo refs/heads/newbranch >expect &&
+ test_cmp expect actual
)
'
+ else
+ test_expect_$RESULT "$command: submodule branch is changed" '
+ prolog &&
+ reset_work_tree_to_interested add_sub1 &&
+ (
+ cd submodule_update &&
+ git -C sub1 checkout -b keep_branch &&
+ git -C sub1 rev-parse HEAD >expect &&
+ git branch -t modify_sub1 origin/modify_sub1 &&
+ $command modify_sub1 &&
+
+ # modified wt
+ test_superproject_content origin/modify_sub1 &&
+ test_submodule_content sub1 origin/modify_sub1 &&
+
+ # unrelated keep_branch is fine
+ git -C sub1 rev-parse keep_branch >actual &&
+ test_cmp expect actual &&
+
+ # submodule ref is checked out
+ git -C sub1 symbolic-ref HEAD >actual &&
+ echo refs/heads/modify_sub1 >expect &&
+ test_cmp expect actual
+ )
+ '
+ fi
# Replacing a tracked file with a submodule produces a checked out submodule
test_expect_success "$command: replace tracked file with submodule checks out submodule" '
diff --git a/t/t1013-read-tree-submodule.sh b/t/t1013-read-tree-submodule.sh
index 91a6fafcb4..de59ebd121 100755
--- a/t/t1013-read-tree-submodule.sh
+++ b/t/t1013-read-tree-submodule.sh
@@ -7,6 +7,7 @@ test_description='read-tree can handle submodules'
KNOWN_FAILURE_DIRECTORY_SUBMODULE_CONFLICTS=1
KNOWN_FAILURE_SUBMODULE_OVERWRITE_IGNORED_UNTRACKED=1
+KNOWN_DIFFERENCE_SUBMODULE_REFS_NOT_UPDATED="read-tree"
test_submodule_switch_recursing_with_args "read-tree -u -m"
diff --git a/t/t2013-checkout-submodule.sh b/t/t2013-checkout-submodule.sh
index 8f86b5f4b2..940f0fca20 100755
--- a/t/t2013-checkout-submodule.sh
+++ b/t/t2013-checkout-submodule.sh
@@ -66,6 +66,10 @@ test_expect_success '"checkout <submodule>" honors submodule.*.ignore from .git/
KNOWN_FAILURE_DIRECTORY_SUBMODULE_CONFLICTS=1
test_submodule_switch_recursing_with_args "checkout"
+KNOWN_DIFFERENCE_SUBMODULE_REFS_NOT_UPDATED="checkout -B current"
+test_submodule_switch_recursing_with_args "checkout -B current"
+unset KNOWN_DIFFERENCE_SUBMODULE_REFS_NOT_UPDATED
+
test_submodule_forced_switch_recursing_with_args "checkout -f"
test_submodule_switch "git checkout"
diff --git a/t/t7112-reset-submodule.sh b/t/t7112-reset-submodule.sh
index a1cb9ff858..c31acc6578 100755
--- a/t/t7112-reset-submodule.sh
+++ b/t/t7112-reset-submodule.sh
@@ -8,13 +8,18 @@ test_description='reset can handle submodules'
KNOWN_FAILURE_SUBMODULE_RECURSIVE_NESTED=1
KNOWN_FAILURE_DIRECTORY_SUBMODULE_CONFLICTS=1
KNOWN_FAILURE_SUBMODULE_OVERWRITE_IGNORED_UNTRACKED=1
+KNOWN_DIFFERENCE_SUBMODULE_REFS_NOT_UPDATED="reset"
+KNOWN_FAILURE_SUBMODULE_REFS_NOT_UPDATED=1
test_submodule_switch_recursing_with_args "reset --keep"
+unset KNOWN_FAILURE_SUBMODULE_REFS_NOT_UPDATED
test_submodule_forced_switch_recursing_with_args "reset --hard"
+KNOWN_FAILURE_SUBMODULE_REFS_NOT_UPDATED=1
test_submodule_switch "git reset --keep"
+unset KNOWN_FAILURE_SUBMODULE_REFS_NOT_UPDATED
test_submodule_switch "git reset --merge"
test_submodule_forced_switch "git reset --hard"
diff --git a/unpack-trees.c b/unpack-trees.c
index 51bfac6aa0..4d4a85226e 100644
--- a/unpack-trees.c
+++ b/unpack-trees.c
@@ -254,6 +254,16 @@ static void display_error_msgs(struct unpack_trees_options *o)
fprintf(stderr, _("Aborting\n"));
}
+static int move_head(const struct unpack_trees_options *o, const char *path, const char *old, const char *new, unsigned flags)
+{
+ if (behave_google_repo_like()) {
+ if (!o->move_head)
+ return submodule_move_head(path, old, new, flags);
+ return o->move_head(o, path, old, new, flags);
+ } else
+ return submodule_move_head(path, old, new, flags);
+}
+
static int check_submodule_move_head(const struct cache_entry *ce,
const char *old_id,
const char *new_id,
@@ -268,7 +278,7 @@ static int check_submodule_move_head(const struct cache_entry *ce,
if (o->reset)
flags |= SUBMODULE_MOVE_HEAD_FORCE;
- if (submodule_move_head(ce->name, old_id, new_id, flags))
+ if (move_head(o, ce->name, old_id, new_id, flags))
return o->gently ? -1 :
add_rejected_path(o, ERROR_WOULD_LOSE_SUBMODULE, ce->name);
return 0;
@@ -304,12 +314,12 @@ static void load_gitmodules_file(struct index_state *index,
* Unlink the last component and schedule the leading directories for
* removal, such that empty directories get removed.
*/
-static void unlink_entry(const struct cache_entry *ce)
+static void unlink_entry(const struct unpack_trees_options *o, const struct cache_entry *ce)
{
const struct submodule *sub = submodule_from_ce(ce);
if (sub) {
/* state.force is set at the caller. */
- submodule_move_head(ce->name, "HEAD", NULL,
+ move_head(o, ce->name, "HEAD", NULL,
SUBMODULE_MOVE_HEAD_FORCE);
}
if (!check_leading_path(ce->name, ce_namelen(ce)))
@@ -408,7 +418,7 @@ static int check_updates(struct unpack_trees_options *o)
if (ce->ce_flags & CE_WT_REMOVE) {
display_progress(progress, ++cnt);
if (o->update && !o->dry_run)
- unlink_entry(ce);
+ unlink_entry(o, ce);
}
}
remove_marked_cache_entries(index);
@@ -450,7 +460,7 @@ static int check_updates(struct unpack_trees_options *o)
display_progress(progress, ++cnt);
ce->ce_flags &= ~CE_UPDATE;
if (o->update && !o->dry_run) {
- errs |= checkout_entry(ce, &state, NULL);
+ errs |= unpack_trees_checkout_entry(o, ce, &state, NULL);
}
}
}
diff --git a/unpack-trees.h b/unpack-trees.h
index 0135080a7b..3f1a205ad0 100644
--- a/unpack-trees.h
+++ b/unpack-trees.h
@@ -9,12 +9,15 @@
#define MAX_UNPACK_TREES 8
struct cache_entry;
+struct tree_desc;
struct unpack_trees_options;
struct exclude_list;
typedef int (*merge_fn_t)(const struct cache_entry * const *src,
struct unpack_trees_options *options);
+typedef int (*submodule_move_head_fn)(const struct unpack_trees_options *o, const char *path, const char *old, const char *new, unsigned flags);
+
enum unpack_trees_error_types {
ERROR_WOULD_OVERWRITE = 0,
ERROR_NOT_UPTODATE_FILE,
@@ -65,6 +68,7 @@ struct unpack_trees_options {
struct dir_struct *dir;
struct pathspec *pathspec;
merge_fn_t fn;
+ submodule_move_head_fn move_head;
const char *msgs[NB_UNPACK_TREES_ERROR_TYPES];
struct argv_array msgs_to_free;
/*
@@ -86,6 +90,12 @@ struct unpack_trees_options {
struct exclude_list *el; /* for internal use */
};
+/* defined in entry.c, for internal use */
+int unpack_trees_checkout_entry(struct unpack_trees_options *o,
+ struct cache_entry *ce,
+ const struct checkout *state,
+ char *topath);
+
int unpack_trees(unsigned n, struct tree_desc *t,
struct unpack_trees_options *options);
--
2.19.0
^ permalink raw reply related [relevance 19%]
* Re: [PATCH] submodule: Alllow staged changes for get_superproject_working_tree
@ 2018-09-27 19:42 7% ` Stefan Beller
0 siblings, 0 replies; 200+ results
From: Stefan Beller @ 2018-09-27 19:42 UTC (permalink / raw)
To: Sam McKelvie; +Cc: git
On Thu, Sep 27, 2018 at 11:12 AM Sam McKelvie <sammck@gmail.com> wrote:
>
> Invoking 'git rev-parse --show-superproject-working-tree' exits with
>
> "fatal: BUG: returned path string doesn't match cwd?"
>
> when the superproject has an unmerged entry for the current submodule,
> instead of displaying the superproject's working tree.
>
> The problem is due to the fact that when a merge of the submodule reference
> is in progress, "git ls-files --stage —full-name <submodule-relative-path>”
> returns three seperate entries for the submodule (one for each stage) rather
> than a single entry; e.g.,
>
> $ git ls-files --stage --full-name submodule-child-test
> 160000 dbbd2766fa330fa741ea59bb38689fcc2d283ac5 1 submodule-child-test
> 160000 f174d1dbfe863a59692c3bdae730a36f2a788c51 2 submodule-child-test
> 160000 e6178f3a58b958543952e12824aa2106d560f21d 3 submodule-child-test
>
> The code in get_superproject_working_tree() expected exactly one entry to
> be returned; this patch makes it use the first entry if multiple entries
> are returned.
>
> Test t1500-rev-parse is extended to cover this case.
>
> Signed-off-by: Sam McKelvie <sammck@gmail.com>
Thanks for following up, this patch is
Reviewed-by: Stefan Beller <sbeller@google.com>
Thanks for adding the test as well!
Stefan
^ permalink raw reply [relevance 7%]
* Re: [PATCH v5 9/9] submodule: support reading .gitmodules when it's not in the working tree
2018-09-27 14:44 7% ` Antonio Ospite
@ 2018-09-27 18:00 7% ` Stefan Beller
2018-10-01 15:45 7% ` Antonio Ospite
0 siblings, 1 reply; 200+ results
From: Stefan Beller @ 2018-09-27 18:00 UTC (permalink / raw)
To: Antonio Ospite
Cc: SZEDER Gábor, Junio C Hamano, git, Brandon Williams,
Daniel Graña, Jonathan Nieder, Richard Hartmann,
Ævar Arnfjörð Bjarmason
On Thu, Sep 27, 2018 at 7:44 AM Antonio Ospite <ao2@ao2.it> wrote:
>
> If you end up touching get_oid() please CC me.
noted. I am not sure I'll touch it anytime soon, though.
>
> Are you suggesting to look into super-prefix for any reason in
> particular?
No, I misread the intent of that part of your message
> >
> > So I think based on ff6f1f564c4 it is safe to remove all calls to
> > repo_read_gitmodules.
> >
>
> Thanks for confirming.
>
> OK, so the plan for v6 is:
>
> - avoid the corruption issues spotted by Gábor by removing the call
> to repo_read_gitmodules in builtin/grep.c (this still does not fix
> the potential problem with nested submodules).
>
> - add a new test-tool which better exercises the new
> config_from_gitmodules code,
Sounds good.
>
> - add also a test_expect_failure test to document the use case that
> cannot be supported yet: nested submodules without .gitmodules in
> their working tree.
Personally I would want to live in a world where we don't *have* to nor
*want* to support submodules without .gitmodules in the respective
superproject.
We did support some use cases historically that I would make sure to
continue to support, but I am not sure how much effort we want to spend
on supporting further use cases of incomplete submodules.
Feel free to do so, as such tests help to document the boundaries.
Stefan
^ permalink raw reply [relevance 7%]
* Re: [PATCH v5 9/9] submodule: support reading .gitmodules when it's not in the working tree
@ 2018-09-27 14:49 4% ` Antonio Ospite
0 siblings, 0 replies; 200+ results
From: Antonio Ospite @ 2018-09-27 14:49 UTC (permalink / raw)
To: Junio C Hamano
Cc: SZEDER Gábor, git, Brandon Williams, Daniel Graña,
Jonathan Nieder, Richard Hartmann, Stefan Beller,
Ævar Arnfjörð Bjarmason
On Fri, 21 Sep 2018 09:19:45 -0700
Junio C Hamano <gitster@pobox.com> wrote:
> Antonio Ospite <ao2@ao2.it> writes:
>
> > Protecting the problematic submodules function could work for now, but
> > I'd like to have more comments, my proposal is:
> >
> > diff --git a/builtin/grep.c b/builtin/grep.c
> > index 601f801158..52b45de749 100644
> > --- a/builtin/grep.c
> > +++ b/builtin/grep.c
> > @@ -427,6 +427,11 @@ static int grep_submodule(struct grep_opt *opt, struct repository *superproject,
> > if (repo_submodule_init(&submodule, superproject, path))
> > return 0;
> >
> > + grep_read_lock();
> > + /*
> > + * NEEDSWORK: repo_read_gitmodules accesses the object store which is
> > + * global, thus it needs to be protected.
> > + */
> > repo_read_gitmodules(&submodule);
> >
> > /*
> > @@ -439,7 +444,6 @@ static int grep_submodule(struct grep_opt *opt, struct repository *superproject,
> > * store is no longer global and instead is a member of the repository
> > * object.
> > */
> > - grep_read_lock();
> > add_to_alternates_memory(submodule.objects->objectdir);
> > grep_read_unlock();
>
> I think this is in line with how the grep codepath protects itself
> when doing anything that accesses the object store.
>
Thanks for the comment.
However, after confirming with Stefan Beller, I think we are going to
solve the corruption issue by removing this call to
repo_read_gitmodules(), which is not strictly necessary:
https://public-inbox.org/git/CAGZ79kZaomuE3p1puznM1x+hu-w4O+ZqeGUODBDj=-R3Z1hDzg@mail.gmail.com/
Thanks,
Antonio
--
Antonio Ospite
https://ao2.it
https://twitter.com/ao2it
A: Because it messes up the order in which people normally read text.
See http://en.wikipedia.org/wiki/Posting_style
Q: Why is top-posting such a bad thing?
^ permalink raw reply [relevance 4%]
* Re: [PATCH v5 9/9] submodule: support reading .gitmodules when it's not in the working tree
2018-09-24 21:00 7% ` Stefan Beller
@ 2018-09-27 14:44 7% ` Antonio Ospite
2018-09-27 18:00 7% ` Stefan Beller
0 siblings, 1 reply; 200+ results
From: Antonio Ospite @ 2018-09-27 14:44 UTC (permalink / raw)
To: Stefan Beller
Cc: SZEDER Gábor, Junio C Hamano, git, Brandon Williams,
Daniel Graña, Jonathan Nieder, Richard Hartmann,
Ævar Arnfjörð Bjarmason
Hi Stefan,
On Mon, 24 Sep 2018 14:00:50 -0700
Stefan Beller <sbeller@google.com> wrote:
> On Mon, Sep 24, 2018 at 3:20 AM Antonio Ospite <ao2@ao2.it> wrote:
>
[...]
> > This is a limitation of the object store in git, there is no equivalent
> > of get_oid() to get the oid from a specific repository and this affects
> > config_with_options too when the config source is a blob.
>
> Not yet, as there is a big push to pass-through an object-store object
> or similar recently and rely less on global variables.
> I am not sure I get to this code, though.
>
If you end up touching get_oid() please CC me.
> > This does not affect commands called via "git -C submodule_dir cmd"
> > because in that case the chdir happens before the_repository is set up,
> > for instance "git-submodule $SOMETHING --recursive" commands seem to
> > change the working directory before the recursion.
>
> For this it may be worth looking into the option
> --super-prefix=<path>
> Currently for internal use only. Set a prefix which gives a
> path from above a repository down to its root. One use is
> to give submodules context about the superproject that
> invoked it.
>
My comment wanted to highlight that there are NO problems in the
mentioned cases:
- git -C submodule_dir cmd
- git submodule cmd --recursive
Are you suggesting to look into super-prefix for any reason in
particular?
[...]
> > The test suite passes even after removing repo_read_gitmodules()
> > entirely from builtin/grep.c, but I am still not confident that I get
> > all the implication of why that call was originally added in commit
> > f9ee2fcdfa (grep: recurse in-process using 'struct repository',
> > 2017-08-02).
>
> If you checkout that commit and remove the call to repo_read_gitmodules
> and then call git-grep in a superproject with nested submodules, you
> get a segfault.
>
> On master (and deleting out that line) you do not get the segfault,
> I think praise goes to ff6f1f564c4 (submodule-config: lazy-load a
> repository's .gitmodules file, 2017-08-03) which happened shortly
> after f9ee2fcdfa.
>
> It showcased that it worked by converting ls-files, but left out grep.
>
> So I think based on ff6f1f564c4 it is safe to remove all calls to
> repo_read_gitmodules.
>
Thanks for confirming.
> > Anyways, even if we removed the call we would prevent the problem from
> > happening in the test suite, but not in the real world, in case non-leaf
> > submodules without .gitmodules in their working tree.
>
> Quite frankly I think grep was just overlooked in review of
> https://public-inbox.org/git/20170803182000.179328-14-bmwill@google.com/
>
OK, so the plan for v6 is:
- avoid the corruption issues spotted by Gábor by removing the call
to repo_read_gitmodules in builtin/grep.c (this still does not fix
the potential problem with nested submodules).
- add a new test-tool which better exercises the new
config_from_gitmodules code,
- add also a test_expect_failure test to document the use case that
cannot be supported yet: nested submodules without .gitmodules in
their working tree.
Thanks,
Antonio
--
Antonio Ospite
https://ao2.it
https://twitter.com/ao2it
A: Because it messes up the order in which people normally read text.
See http://en.wikipedia.org/wiki/Posting_style
Q: Why is top-posting such a bad thing?
^ permalink raw reply [relevance 7%]
* Re: On shipping more of our technical docs as manpages
@ 2018-09-26 23:21 4% ` Stefan Beller
0 siblings, 0 replies; 200+ results
From: Stefan Beller @ 2018-09-26 23:21 UTC (permalink / raw)
To: Ævar Arnfjörð Bjarmason; +Cc: Jeff King, Junio C Hamano, git
On Wed, Sep 26, 2018 at 1:44 PM Ævar Arnfjörð Bjarmason
<avarab@gmail.com> wrote:
> > And if we were going to generate something external, would it make more
> > sense to write in a structured format like doxygen? I am not a big fan
> > of it myself, but at least from there you can generate a more richly
> > interconnected set of documentation.
>
> It's useful to have a single authoritative source for all documentation
> that's easy to search through.
If that is the case I would propose to keep it all in header files and
organize the headers.
> That includes stuff like perl585delta(1) which we'd stick in
> Documentation/RelNotes, and "Internals and C Language Interface". Most
> of what we'd put in Documentation/technical/api-* & headers is in
> perlapi(1).
This seems cool, but was also a recent introduction?
perl400delta seems to yield nothing for me (which may be because
I do not have an old version of perl installed?)
>
> Sometimes it's obvious where to look, but as someone who uses most of
> these APIs very occasionally (because I contribute less) I keep
> (re-)discovering various internal APIs.
Sometimes I have the same feeling. Maybe more structure in the
source files would help (e.g. datastructures/{strbuf, string-list}.h
and objects/{packfile.h, sha1*} ?
> Every perl installation also ships perlapi and a MB or so of obscure
> internal docs to everyone, which makes it easier to find via Google et
> al, which I find useful. So I guess I'm more on the side fence of
> dropping a hypothetical gitapi-oid-array into /usr/share/man than you
> are.
>
Personally I would not want to ship our internal docs everywhere
as it seems like overly wasteful. But then, only my early days
of childhood were guided by Internet that is not available anywhere
and everywhere. Today I would just take for granted that I can lookup
things in github/git/git when I am in not at my regular desk.
> ANYWAY
>
> This E-mail is much longer than I intended, sorry about that. I didn't
> just mean to bitch about how we ship docs, but I was wondering if there
> was a desire to move to something like what I've outlined above, or
> whether the status quo was mostly by design and intended.
>
> I just thought I'd write this rather lengthy E-Mails because this is one
> of the rare areas where you can fully describe an idea in E-Mail without
> writing any patches.
>
> If the consensus is that something like the exhaustive index "perldoc
> perl" provides wouldn't be useful for git I can just drop this, but if
> people are enthusiastic about having that it would be useful to work on
> it...
I agree with Junio, as that the discoverability is unrelated to where to store
the actual docs, Documentation/technical/* has the advantage that it
only has "good" stuff, i.e. some topic that someone cared enough to
write about and it is easy to guess if it is relevant to your need.
In *.h, we have a lot of false positives (xdiff-interface.h or cache.h
just pollute the searching space when looking for suitable storage
classes.)
So I wonder if we'd want to have a list (similar as in
command-list.txt) that describes the header files and gives
some classification of them, for example one class could be the
data structures (strbuf, stringlist, hashmap), algorithms
(diff, range-diff), OS specific stuff (run-command, trace, alloc)
or Git specific things (apply, fetch, submodule)
^ permalink raw reply [relevance 4%]
* Re: [PATCH v4 4/9] submodule: move global changed_submodule_names into fetch submodule struct
2018-09-25 19:47 16% ` [PATCH v4 4/9] submodule: move global changed_submodule_names into fetch submodule struct Stefan Beller
@ 2018-09-26 22:19 5% ` Junio C Hamano
0 siblings, 0 replies; 200+ results
From: Junio C Hamano @ 2018-09-26 22:19 UTC (permalink / raw)
To: Stefan Beller; +Cc: git
Stefan Beller <sbeller@google.com> writes:
> The `changed_submodule_names` are only used for fetching, so let's make it
> part of the struct that is passed around for fetching submodules.
>
> Signed-off-by: Stefan Beller <sbeller@google.com>
> ---
> submodule.c | 42 +++++++++++++++++++++++-------------------
> 1 file changed, 23 insertions(+), 19 deletions(-)
Yup, the less file-scope static we have and the more of them moved
to a struct, the closer we get to be able to use multiple of them at
the same time, which is a very nice step in the right direction.
>
> diff --git a/submodule.c b/submodule.c
> index 22c64bd8559..17103379ba4 100644
> --- a/submodule.c
> +++ b/submodule.c
> @@ -25,7 +25,7 @@
> #include "commit-reach.h"
>
> static int config_update_recurse_submodules = RECURSE_SUBMODULES_OFF;
> -static struct string_list changed_submodule_names = STRING_LIST_INIT_DUP;
> +
> static int initialized_fetch_ref_tips;
> static struct oid_array ref_tips_before_fetch;
> static struct oid_array ref_tips_after_fetch;
> @@ -1110,7 +1110,22 @@ void check_for_new_submodule_commits(struct object_id *oid)
> oid_array_append(&ref_tips_after_fetch, oid);
> }
>
> -static void calculate_changed_submodule_paths(void)
> +struct submodule_parallel_fetch {
> + int count;
> + struct argv_array args;
> + struct repository *r;
> + const char *prefix;
> + int command_line_option;
> + int default_option;
> + int quiet;
> + int result;
> +
> + struct string_list changed_submodule_names;
> +};
> +#define SPF_INIT {0, ARGV_ARRAY_INIT, NULL, NULL, 0, 0, 0, 0, STRING_LIST_INIT_DUP }
> +
> +static void calculate_changed_submodule_paths(
> + struct submodule_parallel_fetch *spf)
> {
> struct argv_array argv = ARGV_ARRAY_INIT;
> struct string_list changed_submodules = STRING_LIST_INIT_DUP;
> @@ -1148,7 +1163,8 @@ static void calculate_changed_submodule_paths(void)
> continue;
>
> if (!submodule_has_commits(path, commits))
> - string_list_append(&changed_submodule_names, name->string);
> + string_list_append(&spf->changed_submodule_names,
> + name->string);
> }
>
> free_submodules_oids(&changed_submodules);
> @@ -1185,18 +1201,6 @@ int submodule_touches_in_range(struct object_id *excl_oid,
> return ret;
> }
>
> -struct submodule_parallel_fetch {
> - int count;
> - struct argv_array args;
> - struct repository *r;
> - const char *prefix;
> - int command_line_option;
> - int default_option;
> - int quiet;
> - int result;
> -};
> -#define SPF_INIT {0, ARGV_ARRAY_INIT, NULL, NULL, 0, 0, 0, 0}
> -
> static int get_fetch_recurse_config(const struct submodule *submodule,
> struct submodule_parallel_fetch *spf)
> {
> @@ -1257,7 +1261,7 @@ static int get_next_submodule(struct child_process *cp,
> case RECURSE_SUBMODULES_ON_DEMAND:
> if (!submodule ||
> !string_list_lookup(
> - &changed_submodule_names,
> + &spf->changed_submodule_names,
> submodule->name))
> continue;
> default_argv = "on-demand";
> @@ -1349,8 +1353,8 @@ int fetch_populated_submodules(struct repository *r,
> argv_array_push(&spf.args, "--recurse-submodules-default");
> /* default value, "--submodule-prefix" and its value are added later */
>
> - calculate_changed_submodule_paths();
> - string_list_sort(&changed_submodule_names);
> + calculate_changed_submodule_paths(&spf);
> + string_list_sort(&spf.changed_submodule_names);
> run_processes_parallel(max_parallel_jobs,
> get_next_submodule,
> fetch_start_failure,
> @@ -1359,7 +1363,7 @@ int fetch_populated_submodules(struct repository *r,
>
> argv_array_clear(&spf.args);
> out:
> - string_list_clear(&changed_submodule_names, 1);
> + string_list_clear(&spf.changed_submodule_names, 1);
> return spf.result;
> }
^ permalink raw reply [relevance 5%]
* Re: [PATCH v4 2/9] submodule.c: fix indentation
2018-09-25 19:47 24% ` [PATCH v4 2/9] submodule.c: fix indentation Stefan Beller
@ 2018-09-26 22:18 4% ` Junio C Hamano
0 siblings, 0 replies; 200+ results
From: Junio C Hamano @ 2018-09-26 22:18 UTC (permalink / raw)
To: Stefan Beller; +Cc: git
Stefan Beller <sbeller@google.com> writes:
> The submodule subsystem is really bad at staying within 80 characters.
> Fix it while we are here.
>
> Signed-off-by: Stefan Beller <sbeller@google.com>
> ---
Makes sense.
> submodule.c | 9 ++++++---
> 1 file changed, 6 insertions(+), 3 deletions(-)
>
> diff --git a/submodule.c b/submodule.c
> index b53cb6e9c47..0de9e2800ad 100644
> --- a/submodule.c
> +++ b/submodule.c
> @@ -1244,7 +1244,8 @@ static int get_next_submodule(struct child_process *cp,
> if (!submodule) {
> const char *name = default_name_or_path(ce->name);
> if (name) {
> - default_submodule.path = default_submodule.name = name;
> + default_submodule.path = name;
> + default_submodule.name = name;
> submodule = &default_submodule;
> }
> }
> @@ -1254,8 +1255,10 @@ static int get_next_submodule(struct child_process *cp,
> default:
> case RECURSE_SUBMODULES_DEFAULT:
> case RECURSE_SUBMODULES_ON_DEMAND:
> - if (!submodule || !unsorted_string_list_lookup(&changed_submodule_names,
> - submodule->name))
> + if (!submodule ||
> + !unsorted_string_list_lookup(
> + &changed_submodule_names,
> + submodule->name))
> continue;
> default_argv = "on-demand";
> break;
^ permalink raw reply [relevance 4%]
* Re: [PATCH v4 3/9] submodule.c: sort changed_submodule_names before searching it
2018-09-25 19:47 17% ` [PATCH v4 3/9] submodule.c: sort changed_submodule_names before searching it Stefan Beller
@ 2018-09-26 22:14 4% ` Junio C Hamano
0 siblings, 0 replies; 200+ results
From: Junio C Hamano @ 2018-09-26 22:14 UTC (permalink / raw)
To: Stefan Beller; +Cc: git
Stefan Beller <sbeller@google.com> writes:
> We can string_list_insert() to maintain sorted-ness of the
> list as we find new items, or we can string_list_append() to
> build an unsorted list and sort it at the end just once.
>
> To pick which one is more appropriate, we notice the fact
> that we discover new items more or less in the already
> sorted order. That makes "append then sort" more
> appropriate.
I somehow thought that we agreed that the second paragraph above did
not make much sense in the previous review round.
... goes and looks ...
https://public-inbox.org/git/CAGZ79kbavjVbTqXsmtjW6=jhkq47_p3mc6=92xOp4_mfhqDtvw@mail.gmail.com/
That was two review cycles ago, I guess.
^ permalink raw reply [relevance 4%]
* [PATCH v4 9/9] builtin/fetch: check for submodule updates for non branch fetches
2018-09-25 19:47 8% [PATCH v4 0/9] fetch: make sure submodule oids are fetched Stefan Beller
` (6 preceding siblings ...)
2018-09-25 19:47 23% ` [PATCH v4 8/9] fetch: retry fetching submodules if needed objects were not fetched Stefan Beller
@ 2018-09-25 19:47 21% ` Stefan Beller
7 siblings, 0 replies; 200+ results
From: Stefan Beller @ 2018-09-25 19:47 UTC (permalink / raw)
To: git; +Cc: Stefan Beller
Gerrit, the code review tool, has a different workflow than our mailing
list based approach. Usually users upload changes to a Gerrit server and
continuous integration and testing happens by bots. Sometimes however a
user wants to checkout a change locally and look at it locally. For this
use case, Gerrit offers a command line snippet to copy and paste to your
terminal, which looks like
git fetch https://<host>/gerrit refs/changes/<id> &&
git checkout FETCH_HEAD
For Gerrit changes that contain changing submodule gitlinks, it would be
easy to extend both the fetch and checkout with the '--recurse-submodules'
flag, such that this command line snippet would produce the state of a
change locally.
However the functionality added in the previous patch, which would
ensure that we fetch the objects in the submodule that the gitlink pointed
at, only works for remote tracking branches so far, not for FETCH_HEAD.
Make sure that fetching a superproject to its FETCH_HEAD, also respects
the existence checks for objects in the submodule recursion.
Signed-off-by: Stefan Beller <sbeller@google.com>
---
builtin/fetch.c | 5 ++++-
t/t5526-fetch-submodules.sh | 2 +-
2 files changed, 5 insertions(+), 2 deletions(-)
diff --git a/builtin/fetch.c b/builtin/fetch.c
index e3b03ad3bd3..f2d9e548bf0 100644
--- a/builtin/fetch.c
+++ b/builtin/fetch.c
@@ -894,11 +894,14 @@ static int store_updated_refs(const char *raw_url, const char *remote_name,
rc |= update_local_ref(ref, what, rm, ¬e,
summary_width);
free(ref);
- } else
+ } else {
+ check_for_new_submodule_commits(&rm->old_oid);
format_display(¬e, '*',
*kind ? kind : "branch", NULL,
*what ? what : "HEAD",
"FETCH_HEAD", summary_width);
+ }
+
if (note.len) {
if (verbosity >= 0 && !shown_url) {
fprintf(stderr, _("From %.*s\n"),
diff --git a/t/t5526-fetch-submodules.sh b/t/t5526-fetch-submodules.sh
index af12c50e7dd..a509eabb044 100755
--- a/t/t5526-fetch-submodules.sh
+++ b/t/t5526-fetch-submodules.sh
@@ -615,7 +615,7 @@ test_expect_success "fetch new commits on-demand when they are not reachable" '
git update-ref refs/changes/2 $D &&
(
cd downstream &&
- git fetch --recurse-submodules --recurse-submodules-default on-demand origin refs/changes/2:refs/heads/my_branch &&
+ git fetch --recurse-submodules origin refs/changes/2 &&
git -C submodule cat-file -t $C &&
git checkout --recurse-submodules FETCH_HEAD
)
--
2.19.0.605.g01d371f741-goog
^ permalink raw reply related [relevance 21%]
* [PATCH v4 8/9] fetch: retry fetching submodules if needed objects were not fetched
2018-09-25 19:47 8% [PATCH v4 0/9] fetch: make sure submodule oids are fetched Stefan Beller
` (5 preceding siblings ...)
2018-09-25 19:47 26% ` [PATCH v4 7/9] submodule: fetch in submodules git directory instead of in worktree Stefan Beller
@ 2018-09-25 19:47 23% ` Stefan Beller
2018-09-25 19:47 21% ` [PATCH v4 9/9] builtin/fetch: check for submodule updates for non branch fetches Stefan Beller
7 siblings, 0 replies; 200+ results
From: Stefan Beller @ 2018-09-25 19:47 UTC (permalink / raw)
To: git; +Cc: Stefan Beller
Currently when git-fetch is asked to recurse into submodules, it dispatches
a plain "git-fetch -C <submodule-dir>" (with some submodule related options
such as prefix and recusing strategy, but) without any information of the
remote or the tip that should be fetched.
This works surprisingly well in some workflows (such as using submodules
as a third party library), while not so well in other scenarios, such
as in a Gerrit topic-based workflow, that can tie together changes
(potentially across repositories) on the server side. One of the parts
of such a Gerrit workflow is to download a change when wanting to examine
it, and you'd want to have its submodule changes that are in the same
topic downloaded as well. However these submodule changes reside in their
own repository in their own ref (refs/changes/<int>).
Retry fetching a submodule by object name if the object id that the
superproject points to, cannot be found.
This retrying does not happen when the "git fetch" done at the
superproject is not storing the fetched results in remote
tracking branches (i.e. instead just recording them to
FETCH_HEAD) in this step. A later patch will fix this.
builtin/fetch used to only inspect submodules when they were fetched
"on-demand", as in either on/off case it was clear whether the submodule
needs to be fetched. However to know whether we need to try fetching the
object ids, we need to identify the object names, which is done in this
function check_for_new_submodule_commits(), so we'll also run that code
in case the submodule recursion is set to "on".
Signed-off-by: Stefan Beller <sbeller@google.com>
---
builtin/fetch.c | 9 +-
submodule.c | 185 ++++++++++++++++++++++++++++++------
t/t5526-fetch-submodules.sh | 16 ++++
3 files changed, 177 insertions(+), 33 deletions(-)
diff --git a/builtin/fetch.c b/builtin/fetch.c
index 0696abfc2a1..e3b03ad3bd3 100644
--- a/builtin/fetch.c
+++ b/builtin/fetch.c
@@ -707,8 +707,7 @@ static int update_local_ref(struct ref *ref,
what = _("[new ref]");
}
- if ((recurse_submodules != RECURSE_SUBMODULES_OFF) &&
- (recurse_submodules != RECURSE_SUBMODULES_ON))
+ if (recurse_submodules != RECURSE_SUBMODULES_OFF)
check_for_new_submodule_commits(&ref->new_oid);
r = s_update_ref(msg, ref, 0);
format_display(display, r ? '!' : '*', what,
@@ -723,8 +722,7 @@ static int update_local_ref(struct ref *ref,
strbuf_add_unique_abbrev(&quickref, ¤t->object.oid, DEFAULT_ABBREV);
strbuf_addstr(&quickref, "..");
strbuf_add_unique_abbrev(&quickref, &ref->new_oid, DEFAULT_ABBREV);
- if ((recurse_submodules != RECURSE_SUBMODULES_OFF) &&
- (recurse_submodules != RECURSE_SUBMODULES_ON))
+ if (recurse_submodules != RECURSE_SUBMODULES_OFF)
check_for_new_submodule_commits(&ref->new_oid);
r = s_update_ref("fast-forward", ref, 1);
format_display(display, r ? '!' : ' ', quickref.buf,
@@ -738,8 +736,7 @@ static int update_local_ref(struct ref *ref,
strbuf_add_unique_abbrev(&quickref, ¤t->object.oid, DEFAULT_ABBREV);
strbuf_addstr(&quickref, "...");
strbuf_add_unique_abbrev(&quickref, &ref->new_oid, DEFAULT_ABBREV);
- if ((recurse_submodules != RECURSE_SUBMODULES_OFF) &&
- (recurse_submodules != RECURSE_SUBMODULES_ON))
+ if (recurse_submodules != RECURSE_SUBMODULES_OFF)
check_for_new_submodule_commits(&ref->new_oid);
r = s_update_ref("forced-update", ref, 1);
format_display(display, r ? '!' : '+', quickref.buf,
diff --git a/submodule.c b/submodule.c
index 3f791f22771..05799362e05 100644
--- a/submodule.c
+++ b/submodule.c
@@ -1127,8 +1127,12 @@ struct submodule_parallel_fetch {
int result;
struct string_list changed_submodule_names;
+ struct get_next_submodule_task **retry;
+ int retry_nr, retry_alloc;
};
-#define SPF_INIT {0, ARGV_ARRAY_INIT, NULL, NULL, 0, 0, 0, 0, STRING_LIST_INIT_DUP }
+#define SPF_INIT {0, ARGV_ARRAY_INIT, NULL, NULL, 0, 0, 0, 0, \
+ STRING_LIST_INIT_DUP, \
+ NULL, 0, 0}
static void calculate_changed_submodule_paths(
struct submodule_parallel_fetch *spf)
@@ -1233,6 +1237,56 @@ static int get_fetch_recurse_config(const struct submodule *submodule,
return spf->default_option;
}
+struct get_next_submodule_task {
+ struct repository *repo;
+ const struct submodule *sub;
+ unsigned free_sub : 1; /* Do we need to free the submodule? */
+ struct oid_array *commits;
+};
+
+static const struct submodule *get_default_submodule(const char *path)
+{
+ struct submodule *ret = NULL;
+ const char *name = default_name_or_path(path);
+
+ if (!name)
+ return NULL;
+
+ ret = xmalloc(sizeof(*ret));
+ memset(ret, 0, sizeof(*ret));
+ ret->path = name;
+ ret->name = name;
+
+ return (const struct submodule *) ret;
+}
+
+static struct get_next_submodule_task *get_next_submodule_task_create(
+ struct repository *r, const char *path)
+{
+ struct get_next_submodule_task *task = xmalloc(sizeof(*task));
+ memset(task, 0, sizeof(*task));
+
+ task->sub = submodule_from_path(r, &null_oid, path);
+ if (!task->sub) {
+ task->sub = get_default_submodule(path);
+ task->free_sub = 1;
+ }
+
+ return task;
+}
+
+static void get_next_submodule_task_release(struct get_next_submodule_task *p)
+{
+ if (p->free_sub)
+ free((void*)p->sub);
+ p->free_sub = 0;
+ p->sub = NULL;
+
+ if (p->repo)
+ repo_clear(p->repo);
+ FREE_AND_NULL(p->repo);
+}
+
static struct repository *get_submodule_repo_for(struct repository *r,
const struct submodule *sub)
{
@@ -1259,39 +1313,35 @@ static struct repository *get_submodule_repo_for(struct repository *r,
static int get_next_submodule(struct child_process *cp,
struct strbuf *err, void *data, void **task_cb)
{
- int ret = 0;
struct submodule_parallel_fetch *spf = data;
for (; spf->count < spf->r->index->cache_nr; spf->count++) {
- struct strbuf submodule_prefix = STRBUF_INIT;
+ int recurse_config;
const struct cache_entry *ce = spf->r->index->cache[spf->count];
const char *default_argv;
- const struct submodule *submodule;
- struct repository *repo;
- struct submodule default_submodule = SUBMODULE_INIT;
+ struct get_next_submodule_task *task;
if (!S_ISGITLINK(ce->ce_mode))
continue;
- submodule = submodule_from_path(spf->r, &null_oid, ce->name);
- if (!submodule) {
- const char *name = default_name_or_path(ce->name);
- if (name) {
- default_submodule.path = name;
- default_submodule.name = name;
- submodule = &default_submodule;
- }
+ task = get_next_submodule_task_create(spf->r, ce->name);
+
+ if (!task->sub) {
+ free(task);
+ continue;
}
- switch (get_fetch_recurse_config(submodule, spf))
+ recurse_config = get_fetch_recurse_config(task->sub, spf);
+
+ switch (recurse_config)
{
default:
case RECURSE_SUBMODULES_DEFAULT:
case RECURSE_SUBMODULES_ON_DEMAND:
- if (!submodule ||
+ if (!task->sub ||
!string_list_lookup(
&spf->changed_submodule_names,
- submodule->name))
+ task->sub->name))
continue;
default_argv = "on-demand";
break;
@@ -1302,12 +1352,12 @@ static int get_next_submodule(struct child_process *cp,
continue;
}
- strbuf_addf(&submodule_prefix, "%s%s/", spf->prefix, ce->name);
- repo = get_submodule_repo_for(spf->r, submodule);
- if (repo) {
+ task->repo = get_submodule_repo_for(spf->r, task->sub);
+ if (task->repo) {
+ struct strbuf submodule_prefix = STRBUF_INIT;
child_process_init(cp);
prepare_submodule_repo_env_in_gitdir(&cp->env_array);
- cp->dir = xstrdup(repo->gitdir);
+ cp->dir = task->repo->gitdir;
cp->git_cmd = 1;
if (!spf->quiet)
strbuf_addf(err, "Fetching submodule %s%s\n",
@@ -1316,18 +1366,51 @@ static int get_next_submodule(struct child_process *cp,
argv_array_pushv(&cp->args, spf->args.argv);
argv_array_push(&cp->args, default_argv);
argv_array_push(&cp->args, "--submodule-prefix");
+
+ strbuf_addf(&submodule_prefix, "%s%s/",
+ spf->prefix,
+ task->sub->path);
argv_array_push(&cp->args, submodule_prefix.buf);
- repo_clear(repo);
- free(repo);
- ret = 1;
- }
- strbuf_release(&submodule_prefix);
- if (ret) {
spf->count++;
+ *task_cb = task;
+
+ strbuf_release(&submodule_prefix);
return 1;
+ } else {
+ get_next_submodule_task_release(task);
+ free(task);
}
}
+
+ if (spf->retry_nr) {
+ struct get_next_submodule_task *task = spf->retry[spf->retry_nr - 1];
+ struct strbuf submodule_prefix = STRBUF_INIT;
+ spf->retry_nr--;
+
+ strbuf_addf(&submodule_prefix, "%s%s/", spf->prefix, task->sub->path);
+
+ child_process_init(cp);
+ prepare_submodule_repo_env_in_gitdir(&cp->env_array);
+ cp->git_cmd = 1;
+ cp->dir = task->repo->gitdir;
+
+ argv_array_init(&cp->args);
+ argv_array_pushv(&cp->args, spf->args.argv);
+ argv_array_push(&cp->args, "on-demand");
+ argv_array_push(&cp->args, "--submodule-prefix");
+ argv_array_push(&cp->args, submodule_prefix.buf);
+
+ /* NEEDSWORK: have get_default_remote from s--h */
+ argv_array_push(&cp->args, "origin");
+ oid_array_for_each_unique(task->commits,
+ append_oid_to_argv, &cp->args);
+
+ *task_cb = task;
+ strbuf_release(&submodule_prefix);
+ return 1;
+ }
+
return 0;
}
@@ -1335,20 +1418,68 @@ static int fetch_start_failure(struct strbuf *err,
void *cb, void *task_cb)
{
struct submodule_parallel_fetch *spf = cb;
+ struct get_next_submodule_task *task = task_cb;
spf->result = 1;
+ get_next_submodule_task_release(task);
return 0;
}
+static int commit_exists_in_sub(const struct object_id *oid, void *data)
+{
+ struct repository *subrepo = data;
+
+ enum object_type type = oid_object_info(subrepo, oid, NULL);
+
+ return type != OBJ_COMMIT;
+}
+
static int fetch_finish(int retvalue, struct strbuf *err,
void *cb, void *task_cb)
{
struct submodule_parallel_fetch *spf = cb;
+ struct get_next_submodule_task *task = task_cb;
+ const struct submodule *sub;
+
+ struct string_list_item *it;
+ struct oid_array *commits;
if (retvalue)
spf->result = 1;
+ if (!task)
+ return 0;
+
+ sub = task->sub;
+ if (!sub)
+ goto out;
+
+ it = string_list_lookup(&spf->changed_submodule_names, sub->name);
+ if (!it)
+ goto out;
+
+ commits = it->util;
+ oid_array_filter(commits,
+ commit_exists_in_sub,
+ task->repo);
+
+ /* Are there commits that do not exist? */
+ if (commits->nr) {
+ /* We already tried fetching them, do not try again. */
+ if (task->commits)
+ return 0;
+
+ task->commits = commits;
+ ALLOC_GROW(spf->retry, spf->retry_nr + 1, spf->retry_alloc);
+ spf->retry[spf->retry_nr] = task;
+ spf->retry_nr++;
+ return 0;
+ }
+
+out:
+ get_next_submodule_task_release(task);
+
return 0;
}
diff --git a/t/t5526-fetch-submodules.sh b/t/t5526-fetch-submodules.sh
index 42692219a1a..af12c50e7dd 100755
--- a/t/t5526-fetch-submodules.sh
+++ b/t/t5526-fetch-submodules.sh
@@ -605,4 +605,20 @@ test_expect_success "fetch new commits when submodule got renamed" '
test_cmp expect actual
'
+test_expect_success "fetch new commits on-demand when they are not reachable" '
+ git checkout --detach &&
+ C=$(git -C submodule commit-tree -m "new change outside refs/heads" HEAD^{tree}) &&
+ git -C submodule update-ref refs/changes/1 $C &&
+ git update-index --cacheinfo 160000 $C submodule &&
+ git commit -m "updated submodule outside of refs/heads" &&
+ D=$(git rev-parse HEAD) &&
+ git update-ref refs/changes/2 $D &&
+ (
+ cd downstream &&
+ git fetch --recurse-submodules --recurse-submodules-default on-demand origin refs/changes/2:refs/heads/my_branch &&
+ git -C submodule cat-file -t $C &&
+ git checkout --recurse-submodules FETCH_HEAD
+ )
+'
+
test_done
--
2.19.0.605.g01d371f741-goog
^ permalink raw reply related [relevance 23%]
* [PATCH v4 7/9] submodule: fetch in submodules git directory instead of in worktree
2018-09-25 19:47 8% [PATCH v4 0/9] fetch: make sure submodule oids are fetched Stefan Beller
` (4 preceding siblings ...)
2018-09-25 19:47 24% ` [PATCH v4 6/9] repository: repo_submodule_init to take a submodule struct Stefan Beller
@ 2018-09-25 19:47 26% ` Stefan Beller
2018-09-25 19:47 23% ` [PATCH v4 8/9] fetch: retry fetching submodules if needed objects were not fetched Stefan Beller
2018-09-25 19:47 21% ` [PATCH v4 9/9] builtin/fetch: check for submodule updates for non branch fetches Stefan Beller
7 siblings, 0 replies; 200+ results
From: Stefan Beller @ 2018-09-25 19:47 UTC (permalink / raw)
To: git; +Cc: Stefan Beller
This patch started as a refactoring to make 'get_next_submodule' more
readable, but upon doing so, I realized that "git fetch" of the submodule
actually doesn't need to be run in the submodules worktree. So let's run
it in its git dir instead.
That should pave the way towards fetching submodules that are currently
not checked out.
This patch leaks the cp->dir in get_next_submodule, as any further
callback in run_processes_parallel doesn't have access to the child
process any more. In an early iteration of this patch, the function
get_submodule_repo_for directly returned the string containing the
git directory, which would be a better design choice for this patch.
However the next patch both fixes the memory leak of cp->dir and also has
a use case for using the full repository handle of the submodule, so
it makes sense to introduce the get_submodule_repo_for here already.
Signed-off-by: Stefan Beller <sbeller@google.com>
---
submodule.c | 51 +++++++++++++++++++++++++++----------
t/t5526-fetch-submodules.sh | 7 ++++-
2 files changed, 44 insertions(+), 14 deletions(-)
diff --git a/submodule.c b/submodule.c
index dd478ed70bf..3f791f22771 100644
--- a/submodule.c
+++ b/submodule.c
@@ -481,6 +481,12 @@ void prepare_submodule_repo_env(struct argv_array *out)
DEFAULT_GIT_DIR_ENVIRONMENT);
}
+static void prepare_submodule_repo_env_in_gitdir(struct argv_array *out)
+{
+ prepare_submodule_repo_env_no_git_dir(out);
+ argv_array_pushf(out, "%s=.", GIT_DIR_ENVIRONMENT);
+}
+
/* Helper function to display the submodule header line prior to the full
* summary output. If it can locate the submodule objects directory it will
* attempt to lookup both the left and right commits and put them into the
@@ -1227,6 +1233,29 @@ static int get_fetch_recurse_config(const struct submodule *submodule,
return spf->default_option;
}
+static struct repository *get_submodule_repo_for(struct repository *r,
+ const struct submodule *sub)
+{
+ struct repository *ret = xmalloc(sizeof(*ret));
+
+ if (repo_submodule_init(ret, r, sub)) {
+ /*
+ * No entry in .gitmodules? Technically not a submodule,
+ * but historically we supported repositories that happen to be
+ * in-place where a gitlink is. Keep supporting them.
+ */
+ struct strbuf gitdir = STRBUF_INIT;
+ strbuf_repo_worktree_path(&gitdir, r, "%s/.git", sub->path);
+ if (repo_init(ret, gitdir.buf, NULL)) {
+ strbuf_release(&gitdir);
+ return NULL;
+ }
+ strbuf_release(&gitdir);
+ }
+
+ return ret;
+}
+
static int get_next_submodule(struct child_process *cp,
struct strbuf *err, void *data, void **task_cb)
{
@@ -1234,12 +1263,11 @@ static int get_next_submodule(struct child_process *cp,
struct submodule_parallel_fetch *spf = data;
for (; spf->count < spf->r->index->cache_nr; spf->count++) {
- struct strbuf submodule_path = STRBUF_INIT;
- struct strbuf submodule_git_dir = STRBUF_INIT;
struct strbuf submodule_prefix = STRBUF_INIT;
const struct cache_entry *ce = spf->r->index->cache[spf->count];
- const char *git_dir, *default_argv;
+ const char *default_argv;
const struct submodule *submodule;
+ struct repository *repo;
struct submodule default_submodule = SUBMODULE_INIT;
if (!S_ISGITLINK(ce->ce_mode))
@@ -1274,16 +1302,12 @@ static int get_next_submodule(struct child_process *cp,
continue;
}
- strbuf_repo_worktree_path(&submodule_path, spf->r, "%s", ce->name);
- strbuf_addf(&submodule_git_dir, "%s/.git", submodule_path.buf);
strbuf_addf(&submodule_prefix, "%s%s/", spf->prefix, ce->name);
- git_dir = read_gitfile(submodule_git_dir.buf);
- if (!git_dir)
- git_dir = submodule_git_dir.buf;
- if (is_directory(git_dir)) {
+ repo = get_submodule_repo_for(spf->r, submodule);
+ if (repo) {
child_process_init(cp);
- cp->dir = strbuf_detach(&submodule_path, NULL);
- prepare_submodule_repo_env(&cp->env_array);
+ prepare_submodule_repo_env_in_gitdir(&cp->env_array);
+ cp->dir = xstrdup(repo->gitdir);
cp->git_cmd = 1;
if (!spf->quiet)
strbuf_addf(err, "Fetching submodule %s%s\n",
@@ -1293,10 +1317,11 @@ static int get_next_submodule(struct child_process *cp,
argv_array_push(&cp->args, default_argv);
argv_array_push(&cp->args, "--submodule-prefix");
argv_array_push(&cp->args, submodule_prefix.buf);
+
+ repo_clear(repo);
+ free(repo);
ret = 1;
}
- strbuf_release(&submodule_path);
- strbuf_release(&submodule_git_dir);
strbuf_release(&submodule_prefix);
if (ret) {
spf->count++;
diff --git a/t/t5526-fetch-submodules.sh b/t/t5526-fetch-submodules.sh
index 6c2f9b2ba26..42692219a1a 100755
--- a/t/t5526-fetch-submodules.sh
+++ b/t/t5526-fetch-submodules.sh
@@ -566,7 +566,12 @@ test_expect_success 'fetching submodule into a broken repository' '
test_must_fail git -C dst status &&
test_must_fail git -C dst diff &&
- test_must_fail git -C dst fetch --recurse-submodules
+
+ # git-fetch cannot find the git directory of the submodule,
+ # so it will do nothing, successfully, as it cannot distinguish between
+ # this broken submodule and a submodule that was just set active but
+ # not cloned yet
+ git -C dst fetch --recurse-submodules
'
test_expect_success "fetch new commits when submodule got renamed" '
--
2.19.0.605.g01d371f741-goog
^ permalink raw reply related [relevance 26%]
* [PATCH v4 6/9] repository: repo_submodule_init to take a submodule struct
2018-09-25 19:47 8% [PATCH v4 0/9] fetch: make sure submodule oids are fetched Stefan Beller
` (3 preceding siblings ...)
2018-09-25 19:47 17% ` [PATCH v4 5/9] submodule.c: do not copy around submodule list Stefan Beller
@ 2018-09-25 19:47 24% ` Stefan Beller
2018-09-25 19:47 26% ` [PATCH v4 7/9] submodule: fetch in submodules git directory instead of in worktree Stefan Beller
` (2 subsequent siblings)
7 siblings, 0 replies; 200+ results
From: Stefan Beller @ 2018-09-25 19:47 UTC (permalink / raw)
To: git; +Cc: Stefan Beller
When constructing a struct repository for a submodule for some revision
of the superproject where the submodule is not contained in the index,
it may not be present in the working tree currently either. In that
siutation giving a 'path' argument is not useful. Upgrade the
repo_submodule_init function to take a struct submodule instead.
While we are at it, overhaul the repo_submodule_init function by renaming
the submodule repository struct, which is to be initialized, to a name
that is not confused with the struct submodule as easily.
Also move its documentation into the header file.
Signed-off-by: Stefan Beller <sbeller@google.com>
---
builtin/grep.c | 17 ++++++++++-------
builtin/ls-files.c | 12 +++++++-----
builtin/submodule--helper.c | 2 +-
repository.c | 27 ++++++++++-----------------
repository.h | 11 +++++++++--
5 files changed, 37 insertions(+), 32 deletions(-)
diff --git a/builtin/grep.c b/builtin/grep.c
index 601f801158f..81c53c862b1 100644
--- a/builtin/grep.c
+++ b/builtin/grep.c
@@ -418,16 +418,19 @@ static int grep_submodule(struct grep_opt *opt, struct repository *superproject,
const struct object_id *oid,
const char *filename, const char *path)
{
- struct repository submodule;
+ struct repository subrepo;
+ const struct submodule *sub = submodule_from_path(superproject,
+ &null_oid, path);
+
int hit;
if (!is_submodule_active(superproject, path))
return 0;
- if (repo_submodule_init(&submodule, superproject, path))
+ if (repo_submodule_init(&subrepo, superproject, sub))
return 0;
- repo_read_gitmodules(&submodule);
+ repo_read_gitmodules(&subrepo);
/*
* NEEDSWORK: This adds the submodule's object directory to the list of
@@ -440,7 +443,7 @@ static int grep_submodule(struct grep_opt *opt, struct repository *superproject,
* object.
*/
grep_read_lock();
- add_to_alternates_memory(submodule.objects->objectdir);
+ add_to_alternates_memory(subrepo.objects->objectdir);
grep_read_unlock();
if (oid) {
@@ -465,14 +468,14 @@ static int grep_submodule(struct grep_opt *opt, struct repository *superproject,
init_tree_desc(&tree, data, size);
hit = grep_tree(opt, pathspec, &tree, &base, base.len,
- object->type == OBJ_COMMIT, &submodule);
+ object->type == OBJ_COMMIT, &subrepo);
strbuf_release(&base);
free(data);
} else {
- hit = grep_cache(opt, &submodule, pathspec, 1);
+ hit = grep_cache(opt, &subrepo, pathspec, 1);
}
- repo_clear(&submodule);
+ repo_clear(&subrepo);
return hit;
}
diff --git a/builtin/ls-files.c b/builtin/ls-files.c
index 7f9919a3623..4d1649c1b3a 100644
--- a/builtin/ls-files.c
+++ b/builtin/ls-files.c
@@ -206,17 +206,19 @@ static void show_files(struct repository *repo, struct dir_struct *dir);
static void show_submodule(struct repository *superproject,
struct dir_struct *dir, const char *path)
{
- struct repository submodule;
+ struct repository subrepo;
+ const struct submodule *sub = submodule_from_path(superproject,
+ &null_oid, path);
- if (repo_submodule_init(&submodule, superproject, path))
+ if (repo_submodule_init(&subrepo, superproject, sub))
return;
- if (repo_read_index(&submodule) < 0)
+ if (repo_read_index(&subrepo) < 0)
die("index file corrupt");
- show_files(&submodule, dir);
+ show_files(&subrepo, dir);
- repo_clear(&submodule);
+ repo_clear(&subrepo);
}
static void show_ce(struct repository *repo, struct dir_struct *dir,
diff --git a/builtin/submodule--helper.c b/builtin/submodule--helper.c
index 40844870cfb..1164e0979bd 100644
--- a/builtin/submodule--helper.c
+++ b/builtin/submodule--helper.c
@@ -2037,7 +2037,7 @@ static int ensure_core_worktree(int argc, const char **argv, const char *prefix)
if (!sub)
BUG("We could get the submodule handle before?");
- if (repo_submodule_init(&subrepo, the_repository, path))
+ if (repo_submodule_init(&subrepo, the_repository, sub))
die(_("could not get a repository handle for submodule '%s'"), path);
if (!repo_config_get_string(&subrepo, "core.worktree", &cw)) {
diff --git a/repository.c b/repository.c
index 5dd14867181..aabe64ee5d9 100644
--- a/repository.c
+++ b/repository.c
@@ -166,30 +166,23 @@ int repo_init(struct repository *repo,
return -1;
}
-/*
- * Initialize 'submodule' as the submodule given by 'path' in parent repository
- * 'superproject'.
- * Return 0 upon success and a non-zero value upon failure.
- */
-int repo_submodule_init(struct repository *submodule,
+int repo_submodule_init(struct repository *subrepo,
struct repository *superproject,
- const char *path)
+ const struct submodule *sub)
{
- const struct submodule *sub;
struct strbuf gitdir = STRBUF_INIT;
struct strbuf worktree = STRBUF_INIT;
int ret = 0;
- sub = submodule_from_path(superproject, &null_oid, path);
if (!sub) {
ret = -1;
goto out;
}
- strbuf_repo_worktree_path(&gitdir, superproject, "%s/.git", path);
- strbuf_repo_worktree_path(&worktree, superproject, "%s", path);
+ strbuf_repo_worktree_path(&gitdir, superproject, "%s/.git", sub->path);
+ strbuf_repo_worktree_path(&worktree, superproject, "%s", sub->path);
- if (repo_init(submodule, gitdir.buf, worktree.buf)) {
+ if (repo_init(subrepo, gitdir.buf, worktree.buf)) {
/*
* If initilization fails then it may be due to the submodule
* not being populated in the superproject's worktree. Instead
@@ -201,16 +194,16 @@ int repo_submodule_init(struct repository *submodule,
strbuf_repo_git_path(&gitdir, superproject,
"modules/%s", sub->name);
- if (repo_init(submodule, gitdir.buf, NULL)) {
+ if (repo_init(subrepo, gitdir.buf, NULL)) {
ret = -1;
goto out;
}
}
- submodule->submodule_prefix = xstrfmt("%s%s/",
- superproject->submodule_prefix ?
- superproject->submodule_prefix :
- "", path);
+ subrepo->submodule_prefix = xstrfmt("%s%s/",
+ superproject->submodule_prefix ?
+ superproject->submodule_prefix :
+ "", sub->path);
out:
strbuf_release(&gitdir);
diff --git a/repository.h b/repository.h
index 9f16c42c1ed..a9c7a5baa54 100644
--- a/repository.h
+++ b/repository.h
@@ -116,9 +116,16 @@ void repo_set_worktree(struct repository *repo, const char *path);
void repo_set_hash_algo(struct repository *repo, int algo);
void initialize_the_repository(void);
int repo_init(struct repository *r, const char *gitdir, const char *worktree);
-int repo_submodule_init(struct repository *submodule,
+
+/*
+ * Initialize the repository 'subrepo' as the submodule given by the
+ * struct submodule 'sub' in parent repository 'superproject'.
+ * Return 0 upon success and a non-zero value upon failure.
+ */
+struct submodule;
+int repo_submodule_init(struct repository *subrepo,
struct repository *superproject,
- const char *path);
+ const struct submodule *sub);
void repo_clear(struct repository *repo);
/*
--
2.19.0.605.g01d371f741-goog
^ permalink raw reply related [relevance 24%]
* [PATCH v4 5/9] submodule.c: do not copy around submodule list
2018-09-25 19:47 8% [PATCH v4 0/9] fetch: make sure submodule oids are fetched Stefan Beller
` (2 preceding siblings ...)
2018-09-25 19:47 16% ` [PATCH v4 4/9] submodule: move global changed_submodule_names into fetch submodule struct Stefan Beller
@ 2018-09-25 19:47 17% ` Stefan Beller
2018-09-25 19:47 24% ` [PATCH v4 6/9] repository: repo_submodule_init to take a submodule struct Stefan Beller
` (3 subsequent siblings)
7 siblings, 0 replies; 200+ results
From: Stefan Beller @ 2018-09-25 19:47 UTC (permalink / raw)
To: git; +Cc: Stefan Beller
'calculate_changed_submodule_paths' uses a local list to compute the
changed submodules, and then produces the result by copying appropriate
items into the result list.
Instead use the result list directly and prune items afterwards
using string_list_remove_empty_items.
By doing so we'll have access to the util pointer for longer that
contains the commits that we need to fetch, which will be
useful in a later patch.
Signed-off-by: Stefan Beller <sbeller@google.com>
---
submodule.c | 19 ++++++++++---------
1 file changed, 10 insertions(+), 9 deletions(-)
diff --git a/submodule.c b/submodule.c
index 17103379ba4..dd478ed70bf 100644
--- a/submodule.c
+++ b/submodule.c
@@ -1128,8 +1128,7 @@ static void calculate_changed_submodule_paths(
struct submodule_parallel_fetch *spf)
{
struct argv_array argv = ARGV_ARRAY_INIT;
- struct string_list changed_submodules = STRING_LIST_INIT_DUP;
- const struct string_list_item *name;
+ struct string_list_item *name;
/* No need to check if there are no submodules configured */
if (!submodule_from_path(the_repository, NULL, NULL))
@@ -1146,9 +1145,9 @@ static void calculate_changed_submodule_paths(
* Collect all submodules (whether checked out or not) for which new
* commits have been recorded upstream in "changed_submodule_names".
*/
- collect_changed_submodules(&changed_submodules, &argv);
+ collect_changed_submodules(&spf->changed_submodule_names, &argv);
- for_each_string_list_item(name, &changed_submodules) {
+ for_each_string_list_item(name, &spf->changed_submodule_names) {
struct oid_array *commits = name->util;
const struct submodule *submodule;
const char *path = NULL;
@@ -1162,12 +1161,14 @@ static void calculate_changed_submodule_paths(
if (!path)
continue;
- if (!submodule_has_commits(path, commits))
- string_list_append(&spf->changed_submodule_names,
- name->string);
+ if (submodule_has_commits(path, commits)) {
+ oid_array_clear(commits);
+ *name->string = '\0';
+ }
}
- free_submodules_oids(&changed_submodules);
+ string_list_remove_empty_items(&spf->changed_submodule_names, 1);
+
argv_array_clear(&argv);
oid_array_clear(&ref_tips_before_fetch);
oid_array_clear(&ref_tips_after_fetch);
@@ -1363,7 +1364,7 @@ int fetch_populated_submodules(struct repository *r,
argv_array_clear(&spf.args);
out:
- string_list_clear(&spf.changed_submodule_names, 1);
+ free_submodules_oids(&spf.changed_submodule_names);
return spf.result;
}
--
2.19.0.605.g01d371f741-goog
^ permalink raw reply related [relevance 17%]
* [PATCH v4 4/9] submodule: move global changed_submodule_names into fetch submodule struct
2018-09-25 19:47 8% [PATCH v4 0/9] fetch: make sure submodule oids are fetched Stefan Beller
2018-09-25 19:47 24% ` [PATCH v4 2/9] submodule.c: fix indentation Stefan Beller
2018-09-25 19:47 17% ` [PATCH v4 3/9] submodule.c: sort changed_submodule_names before searching it Stefan Beller
@ 2018-09-25 19:47 16% ` Stefan Beller
2018-09-26 22:19 5% ` Junio C Hamano
2018-09-25 19:47 17% ` [PATCH v4 5/9] submodule.c: do not copy around submodule list Stefan Beller
` (4 subsequent siblings)
7 siblings, 1 reply; 200+ results
From: Stefan Beller @ 2018-09-25 19:47 UTC (permalink / raw)
To: git; +Cc: Stefan Beller
The `changed_submodule_names` are only used for fetching, so let's make it
part of the struct that is passed around for fetching submodules.
Signed-off-by: Stefan Beller <sbeller@google.com>
---
submodule.c | 42 +++++++++++++++++++++++-------------------
1 file changed, 23 insertions(+), 19 deletions(-)
diff --git a/submodule.c b/submodule.c
index 22c64bd8559..17103379ba4 100644
--- a/submodule.c
+++ b/submodule.c
@@ -25,7 +25,7 @@
#include "commit-reach.h"
static int config_update_recurse_submodules = RECURSE_SUBMODULES_OFF;
-static struct string_list changed_submodule_names = STRING_LIST_INIT_DUP;
+
static int initialized_fetch_ref_tips;
static struct oid_array ref_tips_before_fetch;
static struct oid_array ref_tips_after_fetch;
@@ -1110,7 +1110,22 @@ void check_for_new_submodule_commits(struct object_id *oid)
oid_array_append(&ref_tips_after_fetch, oid);
}
-static void calculate_changed_submodule_paths(void)
+struct submodule_parallel_fetch {
+ int count;
+ struct argv_array args;
+ struct repository *r;
+ const char *prefix;
+ int command_line_option;
+ int default_option;
+ int quiet;
+ int result;
+
+ struct string_list changed_submodule_names;
+};
+#define SPF_INIT {0, ARGV_ARRAY_INIT, NULL, NULL, 0, 0, 0, 0, STRING_LIST_INIT_DUP }
+
+static void calculate_changed_submodule_paths(
+ struct submodule_parallel_fetch *spf)
{
struct argv_array argv = ARGV_ARRAY_INIT;
struct string_list changed_submodules = STRING_LIST_INIT_DUP;
@@ -1148,7 +1163,8 @@ static void calculate_changed_submodule_paths(void)
continue;
if (!submodule_has_commits(path, commits))
- string_list_append(&changed_submodule_names, name->string);
+ string_list_append(&spf->changed_submodule_names,
+ name->string);
}
free_submodules_oids(&changed_submodules);
@@ -1185,18 +1201,6 @@ int submodule_touches_in_range(struct object_id *excl_oid,
return ret;
}
-struct submodule_parallel_fetch {
- int count;
- struct argv_array args;
- struct repository *r;
- const char *prefix;
- int command_line_option;
- int default_option;
- int quiet;
- int result;
-};
-#define SPF_INIT {0, ARGV_ARRAY_INIT, NULL, NULL, 0, 0, 0, 0}
-
static int get_fetch_recurse_config(const struct submodule *submodule,
struct submodule_parallel_fetch *spf)
{
@@ -1257,7 +1261,7 @@ static int get_next_submodule(struct child_process *cp,
case RECURSE_SUBMODULES_ON_DEMAND:
if (!submodule ||
!string_list_lookup(
- &changed_submodule_names,
+ &spf->changed_submodule_names,
submodule->name))
continue;
default_argv = "on-demand";
@@ -1349,8 +1353,8 @@ int fetch_populated_submodules(struct repository *r,
argv_array_push(&spf.args, "--recurse-submodules-default");
/* default value, "--submodule-prefix" and its value are added later */
- calculate_changed_submodule_paths();
- string_list_sort(&changed_submodule_names);
+ calculate_changed_submodule_paths(&spf);
+ string_list_sort(&spf.changed_submodule_names);
run_processes_parallel(max_parallel_jobs,
get_next_submodule,
fetch_start_failure,
@@ -1359,7 +1363,7 @@ int fetch_populated_submodules(struct repository *r,
argv_array_clear(&spf.args);
out:
- string_list_clear(&changed_submodule_names, 1);
+ string_list_clear(&spf.changed_submodule_names, 1);
return spf.result;
}
--
2.19.0.605.g01d371f741-goog
^ permalink raw reply related [relevance 16%]
* [PATCH v4 3/9] submodule.c: sort changed_submodule_names before searching it
2018-09-25 19:47 8% [PATCH v4 0/9] fetch: make sure submodule oids are fetched Stefan Beller
2018-09-25 19:47 24% ` [PATCH v4 2/9] submodule.c: fix indentation Stefan Beller
@ 2018-09-25 19:47 17% ` Stefan Beller
2018-09-26 22:14 4% ` Junio C Hamano
2018-09-25 19:47 16% ` [PATCH v4 4/9] submodule: move global changed_submodule_names into fetch submodule struct Stefan Beller
` (5 subsequent siblings)
7 siblings, 1 reply; 200+ results
From: Stefan Beller @ 2018-09-25 19:47 UTC (permalink / raw)
To: git; +Cc: Stefan Beller
We can string_list_insert() to maintain sorted-ness of the
list as we find new items, or we can string_list_append() to
build an unsorted list and sort it at the end just once.
To pick which one is more appropriate, we notice the fact
that we discover new items more or less in the already
sorted order. That makes "append then sort" more
appropriate.
Signed-off-by: Stefan Beller <sbeller@google.com>
---
submodule.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/submodule.c b/submodule.c
index 0de9e2800ad..22c64bd8559 100644
--- a/submodule.c
+++ b/submodule.c
@@ -1256,7 +1256,7 @@ static int get_next_submodule(struct child_process *cp,
case RECURSE_SUBMODULES_DEFAULT:
case RECURSE_SUBMODULES_ON_DEMAND:
if (!submodule ||
- !unsorted_string_list_lookup(
+ !string_list_lookup(
&changed_submodule_names,
submodule->name))
continue;
@@ -1350,6 +1350,7 @@ int fetch_populated_submodules(struct repository *r,
/* default value, "--submodule-prefix" and its value are added later */
calculate_changed_submodule_paths();
+ string_list_sort(&changed_submodule_names);
run_processes_parallel(max_parallel_jobs,
get_next_submodule,
fetch_start_failure,
--
2.19.0.605.g01d371f741-goog
^ permalink raw reply related [relevance 17%]
* [PATCH v4 2/9] submodule.c: fix indentation
2018-09-25 19:47 8% [PATCH v4 0/9] fetch: make sure submodule oids are fetched Stefan Beller
@ 2018-09-25 19:47 24% ` Stefan Beller
2018-09-26 22:18 4% ` Junio C Hamano
2018-09-25 19:47 17% ` [PATCH v4 3/9] submodule.c: sort changed_submodule_names before searching it Stefan Beller
` (6 subsequent siblings)
7 siblings, 1 reply; 200+ results
From: Stefan Beller @ 2018-09-25 19:47 UTC (permalink / raw)
To: git; +Cc: Stefan Beller
The submodule subsystem is really bad at staying within 80 characters.
Fix it while we are here.
Signed-off-by: Stefan Beller <sbeller@google.com>
---
submodule.c | 9 ++++++---
1 file changed, 6 insertions(+), 3 deletions(-)
diff --git a/submodule.c b/submodule.c
index b53cb6e9c47..0de9e2800ad 100644
--- a/submodule.c
+++ b/submodule.c
@@ -1244,7 +1244,8 @@ static int get_next_submodule(struct child_process *cp,
if (!submodule) {
const char *name = default_name_or_path(ce->name);
if (name) {
- default_submodule.path = default_submodule.name = name;
+ default_submodule.path = name;
+ default_submodule.name = name;
submodule = &default_submodule;
}
}
@@ -1254,8 +1255,10 @@ static int get_next_submodule(struct child_process *cp,
default:
case RECURSE_SUBMODULES_DEFAULT:
case RECURSE_SUBMODULES_ON_DEMAND:
- if (!submodule || !unsorted_string_list_lookup(&changed_submodule_names,
- submodule->name))
+ if (!submodule ||
+ !unsorted_string_list_lookup(
+ &changed_submodule_names,
+ submodule->name))
continue;
default_argv = "on-demand";
break;
--
2.19.0.605.g01d371f741-goog
^ permalink raw reply related [relevance 24%]
* [PATCH v4 0/9] fetch: make sure submodule oids are fetched
@ 2018-09-25 19:47 8% Stefan Beller
2018-09-25 19:47 24% ` [PATCH v4 2/9] submodule.c: fix indentation Stefan Beller
` (7 more replies)
0 siblings, 8 replies; 200+ results
From: Stefan Beller @ 2018-09-25 19:47 UTC (permalink / raw)
To: git; +Cc: Stefan Beller
v4:
* Per Ævars comment, moved the docs for oid_array_filter into
Documentation/technical/...
* addressed all outstanding comment as noted in "What's cooking" email,
* below is a range-diff against the currently queued version of the
series.
v3:
* I discovered some issues with v2 after sending,
which is why I rewrote the later patches completely
and now we pass around a "task" struct that contains everything to know
about the things to work on and what needs free()ing afterwards.
* as it is no longer string list based, this drops adding string_list_{pop, last}
v2:
* extended commit messages,
* plugged a memory leak
* rewrote the patch "sha1-array: provide oid_array_filter" to be much more like
object_array_fiter
* fixed a typo pointed out by Ramsay.
The range diff is below.
Thanks,
Stefan
v1:
Currently when git-fetch is asked to recurse into submodules, it dispatches
a plain "git-fetch -C <submodule-dir>" (and some submodule related options
such as prefix and recusing strategy, but) without any information of the
remote or the tip that should be fetched.
This works surprisingly well in some workflows, not so well in others,
which this series aims to fix.
The first patches provide new basic functionality and do some refactoring;
the interesting part is in the two last patches.
This was discussed in
https://public-inbox.org/git/20180808221752.195419-1-sbeller@google.com/
and I think I addressed all feedback so far.
Stefan Beller (9):
sha1-array: provide oid_array_filter
submodule.c: fix indentation
submodule.c: sort changed_submodule_names before searching it
submodule: move global changed_submodule_names into fetch submodule
struct
submodule.c: do not copy around submodule list
repository: repo_submodule_init to take a submodule struct
submodule: fetch in submodules git directory instead of in worktree
fetch: retry fetching submodules if needed objects were not fetched
builtin/fetch: check for submodule updates for non branch fetches
Documentation/technical/api-oid-array.txt | 5 +
builtin/fetch.c | 14 +-
builtin/grep.c | 17 +-
builtin/ls-files.c | 12 +-
builtin/submodule--helper.c | 2 +-
repository.c | 27 +--
repository.h | 11 +-
sha1-array.c | 17 ++
sha1-array.h | 3 +
submodule.c | 275 +++++++++++++++++-----
t/t5526-fetch-submodules.sh | 23 +-
11 files changed, 311 insertions(+), 95 deletions(-)
git range-diff origin/sb/submodule-recursive-fetch-gets-the-tip...
1: 6fecf7cd01a < -: ----------- string-list: add string_list_{pop, last} functions
2: 7007a318a68 < -: ----------- sha1-array: provide oid_array_filter
[ ... snip ... I rebased onto: ]
-: ----------- > 215: fe8321ec057 Second batch post 2.19
-: ----------- > 216: a9b49d4cfe9 sha1-array: provide oid_array_filter
3: 807429234ac ! 217: 813205700d1 submodule.c: fix indentation
@@ -6,7 +6,6 @@
Fix it while we are here.
Signed-off-by: Stefan Beller <sbeller@google.com>
- Signed-off-by: Junio C Hamano <gitster@pobox.com>
diff --git a/submodule.c b/submodule.c
--- a/submodule.c
4: f6fa5273af9 ! 218: b4aa77f72ba submodule.c: sort changed_submodule_names before searching it
@@ -12,7 +12,6 @@
appropriate.
Signed-off-by: Stefan Beller <sbeller@google.com>
- Signed-off-by: Junio C Hamano <gitster@pobox.com>
diff --git a/submodule.c b/submodule.c
--- a/submodule.c
5: adf7a2fd203 ! 219: df4da0e18e4 submodule: move global changed_submodule_names into fetch submodule struct
@@ -6,13 +6,12 @@
part of the struct that is passed around for fetching submodules.
Signed-off-by: Stefan Beller <sbeller@google.com>
- Signed-off-by: Junio C Hamano <gitster@pobox.com>
diff --git a/submodule.c b/submodule.c
--- a/submodule.c
+++ b/submodule.c
@@
- #include "object-store.h"
+ #include "commit-reach.h"
static int config_update_recurse_submodules = RECURSE_SUBMODULES_OFF;
-static struct string_list changed_submodule_names = STRING_LIST_INIT_DUP;
6: 56c9398589a ! 220: b9f5d06c134 submodule.c: do not copy around submodule list
@@ -9,12 +9,11 @@
Instead use the result list directly and prune items afterwards
using string_list_remove_empty_items.
- By doin so we'll have access to the util pointer for longer that
+ By doing so we'll have access to the util pointer for longer that
contains the commits that we need to fetch, which will be
useful in a later patch.
Signed-off-by: Stefan Beller <sbeller@google.com>
- Signed-off-by: Junio C Hamano <gitster@pobox.com>
diff --git a/submodule.c b/submodule.c
--- a/submodule.c
-: ----------- > 221: e5ff287a0a2 repository: repo_submodule_init to take a submodule struct
7: 9f70a5f32c9 ! 222: 8acd734b5f5 submodule: fetch in submodules git directory instead of in worktree
@@ -3,14 +3,24 @@
submodule: fetch in submodules git directory instead of in worktree
This patch started as a refactoring to make 'get_next_submodule' more
- readable, but upon doing so, I realized that git-fetch actually doesn't
- need to be run in the worktree. So let's run it in the git dir instead.
+ readable, but upon doing so, I realized that "git fetch" of the submodule
+ actually doesn't need to be run in the submodules worktree. So let's run
+ it in its git dir instead.
That should pave the way towards fetching submodules that are currently
not checked out.
+ This patch leaks the cp->dir in get_next_submodule, as any further
+ callback in run_processes_parallel doesn't have access to the child
+ process any more. In an early iteration of this patch, the function
+ get_submodule_repo_for directly returned the string containing the
+ git directory, which would be a better design choice for this patch.
+
+ However the next patch both fixes the memory leak of cp->dir and also has
+ a use case for using the full repository handle of the submodule, so
+ it makes sense to introduce the get_submodule_repo_for here already.
+
Signed-off-by: Stefan Beller <sbeller@google.com>
- Signed-off-by: Junio C Hamano <gitster@pobox.com>
diff --git a/submodule.c b/submodule.c
--- a/submodule.c
@@ -32,24 +42,26 @@
return spf->default_option;
}
-+static const char *get_submodule_git_dir(struct repository *r, const char *path)
++static struct repository *get_submodule_repo_for(struct repository *r,
++ const struct submodule *sub)
+{
-+ struct repository subrepo;
-+ const char *ret;
++ struct repository *ret = xmalloc(sizeof(*ret));
+
-+ if (repo_submodule_init(&subrepo, r, path)) {
-+ /* no entry in .gitmodules? */
++ if (repo_submodule_init(ret, r, sub)) {
++ /*
++ * No entry in .gitmodules? Technically not a submodule,
++ * but historically we supported repositories that happen to be
++ * in-place where a gitlink is. Keep supporting them.
++ */
+ struct strbuf gitdir = STRBUF_INIT;
-+ strbuf_repo_worktree_path(&gitdir, r, "%s/.git", path);
-+ if (repo_init(&subrepo, gitdir.buf, NULL)) {
++ strbuf_repo_worktree_path(&gitdir, r, "%s/.git", sub->path);
++ if (repo_init(ret, gitdir.buf, NULL)) {
+ strbuf_release(&gitdir);
+ return NULL;
+ }
++ strbuf_release(&gitdir);
+ }
+
-+ ret = xstrdup(subrepo.gitdir);
-+ repo_clear(&subrepo);
-+
+ return ret;
+}
+
@@ -64,7 +76,13 @@
- struct strbuf submodule_git_dir = STRBUF_INIT;
struct strbuf submodule_prefix = STRBUF_INIT;
const struct cache_entry *ce = spf->r->index->cache[spf->count];
- const char *git_dir, *default_argv;
+- const char *git_dir, *default_argv;
++ const char *default_argv;
+ const struct submodule *submodule;
++ struct repository *repo;
+ struct submodule default_submodule = SUBMODULE_INIT;
+
+ if (!S_ISGITLINK(ce->ce_mode))
@@
continue;
}
@@ -76,18 +94,23 @@
- if (!git_dir)
- git_dir = submodule_git_dir.buf;
- if (is_directory(git_dir)) {
-+ git_dir = get_submodule_git_dir(spf->r, ce->name);
-+ if (git_dir) {
++ repo = get_submodule_repo_for(spf->r, submodule);
++ if (repo) {
child_process_init(cp);
- cp->dir = strbuf_detach(&submodule_path, NULL);
- prepare_submodule_repo_env(&cp->env_array);
+ prepare_submodule_repo_env_in_gitdir(&cp->env_array);
-+ cp->dir = git_dir;
++ cp->dir = xstrdup(repo->gitdir);
cp->git_cmd = 1;
if (!spf->quiet)
strbuf_addf(err, "Fetching submodule %s%s\n",
@@
+ argv_array_push(&cp->args, default_argv);
+ argv_array_push(&cp->args, "--submodule-prefix");
argv_array_push(&cp->args, submodule_prefix.buf);
++
++ repo_clear(repo);
++ free(repo);
ret = 1;
}
- strbuf_release(&submodule_path);
8: bab609b4dc1 ! 223: 5752ba212a7 fetch: retry fetching submodules if sha1 were not fetched
@@ -1,9 +1,9 @@
Author: Stefan Beller <sbeller@google.com>
- fetch: retry fetching submodules if sha1 were not fetched
+ fetch: retry fetching submodules if needed objects were not fetched
Currently when git-fetch is asked to recurse into submodules, it dispatches
- a plain "git-fetch -C <submodule-dir>" (and some submodule related options
+ a plain "git-fetch -C <submodule-dir>" (with some submodule related options
such as prefix and recusing strategy, but) without any information of the
remote or the tip that should be fetched.
@@ -16,16 +16,22 @@
topic downloaded as well. However these submodule changes reside in their
own repository in their own ref (refs/changes/<int>).
- Retry fetching a submodule if the object id that the superproject points
- to, cannot be found.
+ Retry fetching a submodule by object name if the object id that the
+ superproject points to, cannot be found.
- This doesn't support fetching to FETCH_HEAD yet, but only into a local
- branch. To make fetching into FETCH_HEAD work, we need some refactoring
- in builtin/fetch.c to adjust the calls to 'check_for_new_submodule_commits'
- that is coming in the next patch.
+ This retrying does not happen when the "git fetch" done at the
+ superproject is not storing the fetched results in remote
+ tracking branches (i.e. instead just recording them to
+ FETCH_HEAD) in this step. A later patch will fix this.
+
+ builtin/fetch used to only inspect submodules when they were fetched
+ "on-demand", as in either on/off case it was clear whether the submodule
+ needs to be fetched. However to know whether we need to try fetching the
+ object ids, we need to identify the object names, which is done in this
+ function check_for_new_submodule_commits(), so we'll also run that code
+ in case the submodule recursion is set to "on".
Signed-off-by: Stefan Beller <sbeller@google.com>
- Signed-off-by: Junio C Hamano <gitster@pobox.com>
diff --git a/builtin/fetch.c b/builtin/fetch.c
--- a/builtin/fetch.c
@@ -68,68 +74,181 @@
int result;
struct string_list changed_submodule_names;
-+ struct string_list retry;
++ struct get_next_submodule_task **retry;
++ int retry_nr, retry_alloc;
};
-#define SPF_INIT {0, ARGV_ARRAY_INIT, NULL, NULL, 0, 0, 0, 0, STRING_LIST_INIT_DUP }
+#define SPF_INIT {0, ARGV_ARRAY_INIT, NULL, NULL, 0, 0, 0, 0, \
+ STRING_LIST_INIT_DUP, \
-+ STRING_LIST_INIT_NODUP}
++ NULL, 0, 0}
static void calculate_changed_submodule_paths(
struct submodule_parallel_fetch *spf)
@@
+ return spf->default_option;
+ }
+
++struct get_next_submodule_task {
++ struct repository *repo;
++ const struct submodule *sub;
++ unsigned free_sub : 1; /* Do we need to free the submodule? */
++ struct oid_array *commits;
++};
++
++static const struct submodule *get_default_submodule(const char *path)
++{
++ struct submodule *ret = NULL;
++ const char *name = default_name_or_path(path);
++
++ if (!name)
++ return NULL;
++
++ ret = xmalloc(sizeof(*ret));
++ memset(ret, 0, sizeof(*ret));
++ ret->path = name;
++ ret->name = name;
++
++ return (const struct submodule *) ret;
++}
++
++static struct get_next_submodule_task *get_next_submodule_task_create(
++ struct repository *r, const char *path)
++{
++ struct get_next_submodule_task *task = xmalloc(sizeof(*task));
++ memset(task, 0, sizeof(*task));
++
++ task->sub = submodule_from_path(r, &null_oid, path);
++ if (!task->sub) {
++ task->sub = get_default_submodule(path);
++ task->free_sub = 1;
++ }
++
++ return task;
++}
++
++static void get_next_submodule_task_release(struct get_next_submodule_task *p)
++{
++ if (p->free_sub)
++ free((void*)p->sub);
++ p->free_sub = 0;
++ p->sub = NULL;
++
++ if (p->repo)
++ repo_clear(p->repo);
++ FREE_AND_NULL(p->repo);
++}
++
+ static struct repository *get_submodule_repo_for(struct repository *r,
+ const struct submodule *sub)
+ {
+@@
+ static int get_next_submodule(struct child_process *cp,
+ struct strbuf *err, void *data, void **task_cb)
{
- int ret = 0;
+- int ret = 0;
struct submodule_parallel_fetch *spf = data;
-+ struct string_list_item *it;
for (; spf->count < spf->r->index->cache_nr; spf->count++) {
+- struct strbuf submodule_prefix = STRBUF_INIT;
+ int recurse_config;
- struct strbuf submodule_prefix = STRBUF_INIT;
const struct cache_entry *ce = spf->r->index->cache[spf->count];
- const char *git_dir, *default_argv;
-@@
- }
+ const char *default_argv;
+- const struct submodule *submodule;
+- struct repository *repo;
+- struct submodule default_submodule = SUBMODULE_INIT;
++ struct get_next_submodule_task *task;
+
+ if (!S_ISGITLINK(ce->ce_mode))
+ continue;
+
+- submodule = submodule_from_path(spf->r, &null_oid, ce->name);
+- if (!submodule) {
+- const char *name = default_name_or_path(ce->name);
+- if (name) {
+- default_submodule.path = name;
+- default_submodule.name = name;
+- submodule = &default_submodule;
+- }
++ task = get_next_submodule_task_create(spf->r, ce->name);
++
++ if (!task->sub) {
++ free(task);
++ continue;
}
- switch (get_fetch_recurse_config(submodule, spf))
-+ recurse_config = get_fetch_recurse_config(submodule, spf);
++ recurse_config = get_fetch_recurse_config(task->sub, spf);
+
+ switch (recurse_config)
{
default:
case RECURSE_SUBMODULES_DEFAULT:
+ case RECURSE_SUBMODULES_ON_DEMAND:
+- if (!submodule ||
++ if (!task->sub ||
+ !string_list_lookup(
+ &spf->changed_submodule_names,
+- submodule->name))
++ task->sub->name))
+ continue;
+ default_argv = "on-demand";
+ break;
+@@
+ continue;
+ }
+
+- strbuf_addf(&submodule_prefix, "%s%s/", spf->prefix, ce->name);
+- repo = get_submodule_repo_for(spf->r, submodule);
+- if (repo) {
++ task->repo = get_submodule_repo_for(spf->r, task->sub);
++ if (task->repo) {
++ struct strbuf submodule_prefix = STRBUF_INIT;
+ child_process_init(cp);
+ prepare_submodule_repo_env_in_gitdir(&cp->env_array);
+- cp->dir = xstrdup(repo->gitdir);
++ cp->dir = task->repo->gitdir;
+ cp->git_cmd = 1;
+ if (!spf->quiet)
+ strbuf_addf(err, "Fetching submodule %s%s\n",
@@
- strbuf_release(&submodule_prefix);
- if (ret) {
+ argv_array_pushv(&cp->args, spf->args.argv);
+ argv_array_push(&cp->args, default_argv);
+ argv_array_push(&cp->args, "--submodule-prefix");
++
++ strbuf_addf(&submodule_prefix, "%s%s/",
++ spf->prefix,
++ task->sub->path);
+ argv_array_push(&cp->args, submodule_prefix.buf);
+
+- repo_clear(repo);
+- free(repo);
+- ret = 1;
+- }
+- strbuf_release(&submodule_prefix);
+- if (ret) {
spf->count++;
-+ if (submodule != &default_submodule)
-+ /* discard const-ness: */
-+ *task_cb = (void*)submodule;
++ *task_cb = task;
++
++ strbuf_release(&submodule_prefix);
return 1;
++ } else {
++ get_next_submodule_task_release(task);
++ free(task);
}
}
+
-+retry_next:
-+
-+ if (spf->retry.nr) {
++ if (spf->retry_nr) {
++ struct get_next_submodule_task *task = spf->retry[spf->retry_nr - 1];
+ struct strbuf submodule_prefix = STRBUF_INIT;
-+ const struct submodule *sub;
++ spf->retry_nr--;
+
-+ it = string_list_last(&spf->retry);
-+ sub = submodule_from_name(spf->r, &null_oid,
-+ it->string);
++ strbuf_addf(&submodule_prefix, "%s%s/", spf->prefix, task->sub->path);
+
+ child_process_init(cp);
-+ cp->dir = get_submodule_git_dir(spf->r, sub->path);
-+ if (!cp->dir) {
-+ string_list_pop(&spf->retry, 0);
-+ goto retry_next;
-+ }
+ prepare_submodule_repo_env_in_gitdir(&cp->env_array);
+ cp->git_cmd = 1;
++ cp->dir = task->repo->gitdir;
+
-+ strbuf_addf(&submodule_prefix, "%s%s/", spf->prefix, sub->path);
+ argv_array_init(&cp->args);
+ argv_array_pushv(&cp->args, spf->args.argv);
+ argv_array_push(&cp->args, "on-demand");
@@ -138,12 +257,11 @@
+
+ /* NEEDSWORK: have get_default_remote from s--h */
+ argv_array_push(&cp->args, "origin");
-+ oid_array_for_each_unique(it->util,
++ oid_array_for_each_unique(task->commits,
+ append_oid_to_argv, &cp->args);
+
-+ *task_cb = NULL; /* make sure we do not recurse forever */
++ *task_cb = task;
+ strbuf_release(&submodule_prefix);
-+ string_list_pop(&spf->retry, 0);
+ return 1;
+ }
+
@@ -151,6 +269,14 @@
}
@@
+ void *cb, void *task_cb)
+ {
+ struct submodule_parallel_fetch *spf = cb;
++ struct get_next_submodule_task *task = task_cb;
+
+ spf->result = 1;
+
++ get_next_submodule_task_release(task);
return 0;
}
@@ -167,35 +293,46 @@
void *cb, void *task_cb)
{
struct submodule_parallel_fetch *spf = cb;
-+ struct submodule *sub = task_cb;
-+ struct repository subrepo;
++ struct get_next_submodule_task *task = task_cb;
++ const struct submodule *sub;
++
++ struct string_list_item *it;
++ struct oid_array *commits;
if (retvalue)
spf->result = 1;
-+ if (!sub)
++ if (!task)
+ return 0;
+
-+ if (repo_submodule_init(&subrepo, spf->r, sub->path) < 0)
-+ warning(_("Could not get submodule repository for submodule '%s' in repository '%s'"),
-+ sub->path, spf->r->worktree);
-+ else {
-+ struct string_list_item *it;
-+ struct oid_array *commits;
++ sub = task->sub;
++ if (!sub)
++ goto out;
+
-+ it = string_list_lookup(&spf->changed_submodule_names, sub->name);
-+ if (!it)
-+ return 0;
++ it = string_list_lookup(&spf->changed_submodule_names, sub->name);
++ if (!it)
++ goto out;
++
++ commits = it->util;
++ oid_array_filter(commits,
++ commit_exists_in_sub,
++ task->repo);
+
-+ commits = it->util;
-+ oid_array_filter(commits,
-+ commit_exists_in_sub,
-+ &subrepo);
++ /* Are there commits that do not exist? */
++ if (commits->nr) {
++ /* We already tried fetching them, do not try again. */
++ if (task->commits)
++ return 0;
+
-+ if (commits->nr)
-+ string_list_append(&spf->retry, sub->name)
-+ ->util = commits;
++ task->commits = commits;
++ ALLOC_GROW(spf->retry, spf->retry_nr + 1, spf->retry_alloc);
++ spf->retry[spf->retry_nr] = task;
++ spf->retry_nr++;
++ return 0;
+ }
++
++out:
++ get_next_submodule_task_release(task);
+
return 0;
}
9: c16d21313f6 ! 224: d02ee9ef485 builtin/fetch: check for submodule updates for non branch fetches
@@ -2,11 +2,29 @@
builtin/fetch: check for submodule updates for non branch fetches
- For Gerrit users that use submodules the invocation of fetch without a
- branch is their main use case.
+ Gerrit, the code review tool, has a different workflow than our mailing
+ list based approach. Usually users upload changes to a Gerrit server and
+ continuous integration and testing happens by bots. Sometimes however a
+ user wants to checkout a change locally and look at it locally. For this
+ use case, Gerrit offers a command line snippet to copy and paste to your
+ terminal, which looks like
+
+ git fetch https://<host>/gerrit refs/changes/<id> &&
+ git checkout FETCH_HEAD
+
+ For Gerrit changes that contain changing submodule gitlinks, it would be
+ easy to extend both the fetch and checkout with the '--recurse-submodules'
+ flag, such that this command line snippet would produce the state of a
+ change locally.
+
+ However the functionality added in the previous patch, which would
+ ensure that we fetch the objects in the submodule that the gitlink pointed
+ at, only works for remote tracking branches so far, not for FETCH_HEAD.
+
+ Make sure that fetching a superproject to its FETCH_HEAD, also respects
+ the existence checks for objects in the submodule recursion.
Signed-off-by: Stefan Beller <sbeller@google.com>
- Signed-off-by: Junio C Hamano <gitster@pobox.com>
diff --git a/builtin/fetch.c b/builtin/fetch.c
--- a/builtin/fetch.c
^ permalink raw reply [relevance 8%]
* Git Test Coverage Report (Tuesday, Sept 25)
@ 2018-09-25 18:42 4% Derrick Stolee
0 siblings, 0 replies; 200+ results
From: Derrick Stolee @ 2018-09-25 18:42 UTC (permalink / raw)
To: Git List
In an effort to ensure new code is reasonably covered by the test suite,
we now have contrib/coverage-diff.sh to combine the gcov output from
'make coverage-test ; make coverage-report' with the output from 'git
diff A B' to discover _new_ lines of code that are not covered.
This report takes the output of these results after running on four
branches:
pu: 80e728fa913dc3a1165b6dec9a7afa6052a86325
jch: 0c10634844314ab89666ed0a1c7d36dde7ac9689
next: 76f2f5c1e34c4dbef1029e2984c2892894c444ce
master: fe8321ec057f9231c26c29b364721568e58040f7
master@{1}: 2d3b1c576c85b7f5db1f418907af00ab88e0c303
I ran the test suite on each of these branches on an Ubuntu Linux VM,
and I'm missing some dependencies (like apache, svn, and perforce) so
not all tests are run.
I submit this output without comment. I'm taking a look especially at my
own lines to see where coverage can be improved.
Thanks,
-Stolee
---
master@{1}..master:
builtin/remote.c
5025425dfff ( Shulhan 2018-09-13 20:18:33 +0700
864) return error(_("No such remote: '%s'"), name);
commit-reach.c
b67f6b26e35 (Derrick Stolee 2018-09-21 08:05:26 -0700
559) continue;
b67f6b26e35 (Derrick Stolee 2018-09-21 08:05:26 -0700
569) from->objects[i].item->flags |= assign_flag;
b67f6b26e35 (Derrick Stolee 2018-09-21 08:05:26 -0700
570) continue;
b67f6b26e35 (Derrick Stolee 2018-09-21 08:05:26 -0700
576) result = 0;
b67f6b26e35 (Derrick Stolee 2018-09-21 08:05:26 -0700
577) goto cleanup;
cat: compat#mingw.c.gcov: No such file or directory
ll-merge.c
d64324cb60e (Torsten Bögershausen 2018-09-12 21:32:02 +0200
379) marker_size = DEFAULT_CONFLICT_MARKER_SIZE;
remote-curl.c
c3b9bc94b9b (Elijah Newren 2018-09-05 10:03:07 -0700
181) options.filter = xstrdup(value);
master..next:
fsck.c
fb8952077df (René Scharfe 2018-09-03 14:49:26 +0000
212) die_errno("Could not read '%s'", path);
midx.c
56ee7ff1565 (Derrick Stolee 2018-09-13 11:02:13 -0700
949) return 0;
cc6af73c029 (Derrick Stolee 2018-09-13 11:02:25 -0700
990) midx_report(_("failed to load pack-index for
packfile %s"),
cc6af73c029 (Derrick Stolee 2018-09-13 11:02:25 -0700
991) e.p->pack_name);
cc6af73c029 (Derrick Stolee 2018-09-13 11:02:25 -0700
992) break;
next..jch:
blame.c
a470beea39b (Nguyễn Thái Ngọc Duy 2018-09-21 17:57:21 +0200
113) !strcmp(r->index->cache[-1 - pos]->name, path))
a470beea39b (Nguyễn Thái Ngọc Duy 2018-09-21 17:57:21 +0200
272) int pos = index_name_pos(r->index, path, len);
a470beea39b (Nguyễn Thái Ngọc Duy 2018-09-21 17:57:21 +0200
274) mode = r->index->cache[pos]->ce_mode;
commit-graph.c
5cef295f283 (Derrick Stolee 2018-08-20 18:24:32 +0000
67) return 0;
20fd6d57996 (Derrick Stolee 2018-08-20 18:24:30 +0000
79) return 0;
config.c
4ee45bacad0 ( Ben Peart 2018-09-12 16:18:55 +0000
2301) if (is_bool)
4ee45bacad0 ( Ben Peart 2018-09-12 16:18:55 +0000
2302) return val ? 0 : 1;
4ee45bacad0 ( Ben Peart 2018-09-12 16:18:55 +0000
2304) return val;
diff.c
b78ea5fc357 (Nguyễn Thái Ngọc Duy 2018-09-21 17:57:19 +0200
4117) add_external_diff_name(o->repo, &argv, other, two);
http-push.c
2abf3503854 (Nguyễn Thái Ngọc Duy 2018-09-21 17:57:38 +0200
1928) repo_init_revisions(the_repository, &revs,
setup_git_directory());
list-objects-filter-options.c
7c8a0cecc49 (Matthew DeVore 2018-09-21 13:32:04 -0700
55) if (errbuf) {
7c8a0cecc49 (Matthew DeVore 2018-09-21 13:32:04 -0700
56) strbuf_addstr(
7c8a0cecc49 (Matthew DeVore 2018-09-21 13:32:04 -0700
60) return 1;
ba72cca605f (Matthew DeVore 2018-09-21 13:32:03 -0700 86) if
(errbuf)
list-objects-filter.c
22e9b63e620 (Matthew DeVore 2018-09-21 13:32:02 -0700
47) BUG("unknown filter_situation: %d", filter_situation);
7c8a0cecc49 (Matthew DeVore 2018-09-21 13:32:04 -0700 100) default:
7c8a0cecc49 (Matthew DeVore 2018-09-21 13:32:04 -0700
101) BUG("unknown filter_situation: %d", filter_situation);
22e9b63e620 (Matthew DeVore 2018-09-21 13:32:02 -0700
152) BUG("unknown filter_situation: %d", filter_situation);
22e9b63e620 (Matthew DeVore 2018-09-21 13:32:02 -0700
257) BUG("unknown filter_situation: %d", filter_situation);
22e9b63e620 (Matthew DeVore 2018-09-21 13:32:02 -0700
438) BUG("invalid list-objects filter choice: %d",
list-objects.c
f447a499dbb (Matthew DeVore 2018-08-13 11:14:28 -0700
197) ctx->show_object(obj, base->buf, ctx->show_data);
preload-index.c
ae9af12287b (Nguyễn Thái Ngọc Duy 2018-09-15 19:56:04 +0200
73) struct progress_data *pd = p->progress;
ae9af12287b (Nguyễn Thái Ngọc Duy 2018-09-15 19:56:04 +0200
75) pthread_mutex_lock(&pd->mutex);
ae9af12287b (Nguyễn Thái Ngọc Duy 2018-09-15 19:56:04 +0200
76) pd->n += last_nr - nr;
ae9af12287b (Nguyễn Thái Ngọc Duy 2018-09-15 19:56:04 +0200
77) display_progress(pd->progress, pd->n);
ae9af12287b (Nguyễn Thái Ngọc Duy 2018-09-15 19:56:04 +0200
78) pthread_mutex_unlock(&pd->mutex);
ae9af12287b (Nguyễn Thái Ngọc Duy 2018-09-15 19:56:04 +0200
79) last_nr = nr;
ae9af12287b (Nguyễn Thái Ngọc Duy 2018-09-15 19:56:04 +0200
93) struct progress_data *pd = p->progress;
ae9af12287b (Nguyễn Thái Ngọc Duy 2018-09-15 19:56:04 +0200
95) pthread_mutex_lock(&pd->mutex);
ae9af12287b (Nguyễn Thái Ngọc Duy 2018-09-15 19:56:04 +0200
96) display_progress(pd->progress, pd->n + last_nr);
ae9af12287b (Nguyễn Thái Ngọc Duy 2018-09-15 19:56:04 +0200
97) pthread_mutex_unlock(&pd->mutex);
ae9af12287b (Nguyễn Thái Ngọc Duy 2018-09-15 19:56:04 +0200
128) pd.progress = start_delayed_progress(_("Refreshing
index"), index->cache_nr);
ae9af12287b (Nguyễn Thái Ngọc Duy 2018-09-15 19:56:04 +0200
129) pthread_mutex_init(&pd.mutex, NULL);
ae9af12287b (Nguyễn Thái Ngọc Duy 2018-09-15 19:56:04 +0200
140) p->progress = &pd;
prio-queue.c
2d181390f3c (Derrick Stolee 2018-09-17 21:08:43 -0700
94) return queue->array[queue->nr - 1].data;
read-cache.c
ae9af12287b (Nguyễn Thái Ngọc Duy 2018-09-15 19:56:04 +0200
1535) display_progress(progress, i);
13e1b78757a (Nguyễn Thái Ngọc Duy 2018-09-12 16:18:57 +0000
1754) strbuf_setlen(name, name->len - len);
13e1b78757a (Nguyễn Thái Ngọc Duy 2018-09-12 16:18:57 +0000
1755) ep = cp + strlen((const char *)cp);
13e1b78757a (Nguyễn Thái Ngọc Duy 2018-09-12 16:18:57 +0000
1792) const unsigned char *cp = (const unsigned char *)name;
13e1b78757a (Nguyễn Thái Ngọc Duy 2018-09-12 16:18:57 +0000
1795) previous_len = previous_ce ? previous_ce->ce_namelen : 0;
13e1b78757a (Nguyễn Thái Ngọc Duy 2018-09-12 16:18:57 +0000
1796) strip_len = decode_varint(&cp);
13e1b78757a (Nguyễn Thái Ngọc Duy 2018-09-12 16:18:57 +0000
1797) if (previous_len < strip_len) {
13e1b78757a (Nguyễn Thái Ngọc Duy 2018-09-12 16:18:57 +0000
1798) if (previous_ce)
13e1b78757a (Nguyễn Thái Ngọc Duy 2018-09-12 16:18:57 +0000
1799) die(_("malformed name field in the
index, near path '%s'"),
13e1b78757a (Nguyễn Thái Ngọc Duy 2018-09-12 16:18:57 +0000
1800) previous_ce->name);
13e1b78757a (Nguyễn Thái Ngọc Duy 2018-09-12 16:18:57 +0000
1802) die(_("malformed name field in the index
in the first path"));
13e1b78757a (Nguyễn Thái Ngọc Duy 2018-09-12 16:18:57 +0000
1804) copy_len = previous_len - strip_len;
13e1b78757a (Nguyễn Thái Ngọc Duy 2018-09-12 16:18:57 +0000
1805) name = (const char *)cp;
13e1b78757a (Nguyễn Thái Ngọc Duy 2018-09-12 16:18:57 +0000
1829) memcpy(ce->name, previous_ce->name, copy_len);
13e1b78757a (Nguyễn Thái Ngọc Duy 2018-09-12 16:18:57 +0000
1830) memcpy(ce->name + copy_len, name, len + 1 - copy_len);
13e1b78757a (Nguyễn Thái Ngọc Duy 2018-09-12 16:18:57 +0000
1831) *ent_size = (name - ((char *)ondisk)) + len + 1 - copy_len;
9928a34be6b ( Ben Peart 2018-09-12 16:18:59 +0000
1954) munmap((void *)p->mmap, p->mmap_size);
4ee45bacad0 ( Ben Peart 2018-09-12 16:18:55 +0000
1955) die("index file corrupt");
ad570c680bc ( Ben Peart 2018-09-12 16:18:56 +0000
1996) mem_pool_init(&istate->ce_mem_pool,
ad570c680bc ( Ben Peart 2018-09-12 16:18:56 +0000 2034)static
void *load_cache_entries_thread(void *_data)
ad570c680bc ( Ben Peart 2018-09-12 16:18:56 +0000 2036) struct
load_cache_entries_thread_data *p = _data;
ad570c680bc ( Ben Peart 2018-09-12 16:18:56 +0000 2038)
p->consumed += load_cache_entry_block(p->istate, p->ce_mem_pool,
13e1b78757a (Nguyễn Thái Ngọc Duy 2018-09-12 16:18:57 +0000
2039) p->offset, p->nr, p->mmap, p->start_offset, p->previous_ce);
ad570c680bc ( Ben Peart 2018-09-12 16:18:56 +0000 2040) return
NULL;
ad570c680bc ( Ben Peart 2018-09-12 16:18:56 +0000 2043)static
unsigned long load_cache_entries_threaded(int nr_threads, struct
index_state *istate,
ad570c680bc ( Ben Peart 2018-09-12 16:18:56 +0000 2046) struct
strbuf previous_name_buf = STRBUF_INIT, *previous_name;
ad570c680bc ( Ben Peart 2018-09-12 16:18:56 +0000 2053) if
(istate->name_hash_initialized)
ad570c680bc ( Ben Peart 2018-09-12 16:18:56 +0000
2054) BUG("the name hash isn't thread safe");
ad570c680bc ( Ben Peart 2018-09-12 16:18:56 +0000 2056)
mem_pool_init(&istate->ce_mem_pool, 0);
ad570c680bc ( Ben Peart 2018-09-12 16:18:56 +0000 2057) if
(istate->version == 4)
ad570c680bc ( Ben Peart 2018-09-12 16:18:56 +0000
2058) previous_name = &previous_name_buf;
ad570c680bc ( Ben Peart 2018-09-12 16:18:56 +0000
2060) previous_name = NULL;
ad570c680bc ( Ben Peart 2018-09-12 16:18:56 +0000 2062)
ce_per_thread = DIV_ROUND_UP(istate->cache_nr, nr_threads);
ad570c680bc ( Ben Peart 2018-09-12 16:18:56 +0000 2063) data =
xcalloc(nr_threads, sizeof(struct load_cache_entries_thread_data));
ad570c680bc ( Ben Peart 2018-09-12 16:18:56 +0000 2070)
consumed = thread = 0;
ad570c680bc ( Ben Peart 2018-09-12 16:18:56 +0000 2071) for (i
= 0; ; i++) {
ad570c680bc ( Ben Peart 2018-09-12 16:18:56 +0000
2080) if (i % ce_per_thread == 0) {
ad570c680bc ( Ben Peart 2018-09-12 16:18:56 +0000
2081) struct load_cache_entries_thread_data *p =
&data[thread];
ad570c680bc ( Ben Peart 2018-09-12 16:18:56 +0000
2083) p->istate = istate;
ad570c680bc ( Ben Peart 2018-09-12 16:18:56 +0000
2084) p->offset = i;
ad570c680bc ( Ben Peart 2018-09-12 16:18:56 +0000
2085) p->nr = ce_per_thread < istate->cache_nr - i ?
ce_per_thread : istate->cache_nr - i;
13e1b78757a (Nguyễn Thái Ngọc Duy 2018-09-12 16:18:57 +0000
2086) p->mmap = mmap;
13e1b78757a (Nguyễn Thái Ngọc Duy 2018-09-12 16:18:57 +0000
2087) p->start_offset = src_offset;
13e1b78757a (Nguyễn Thái Ngọc Duy 2018-09-12 16:18:57 +0000
2090) if (istate->version == 4) {
ad570c680bc ( Ben Peart 2018-09-12 16:18:56 +0000 2091)
mem_pool_init(&p->ce_mem_pool,
ad570c680bc ( Ben Peart 2018-09-12 16:18:56 +0000 2092)
estimate_cache_size_from_compressed(p->nr));
13e1b78757a (Nguyễn Thái Ngọc Duy 2018-09-12 16:18:57 +0000
2095) if (previous_name->len) {
13e1b78757a (Nguyễn Thái Ngọc Duy 2018-09-12 16:18:57 +0000
2096) p->previous_ce = mem_pool__ce_alloc(p->ce_mem_pool,
previous_name->len);
13e1b78757a (Nguyễn Thái Ngọc Duy 2018-09-12 16:18:57 +0000
2097) p->previous_ce->ce_namelen = previous_name->len;
13e1b78757a (Nguyễn Thái Ngọc Duy 2018-09-12 16:18:57 +0000
2098) memcpy(p->previous_ce->name, previous_name->buf, previous_name->len);
ad570c680bc ( Ben Peart 2018-09-12 16:18:56 +0000 2101)
mem_pool_init(&p->ce_mem_pool,
ad570c680bc ( Ben Peart 2018-09-12 16:18:56 +0000 2102)
estimate_cache_size(mmap_size, p->nr));
ad570c680bc ( Ben Peart 2018-09-12 16:18:56 +0000
2105) if (pthread_create(&p->pthread, NULL,
load_cache_entries_thread, p))
ad570c680bc ( Ben Peart 2018-09-12 16:18:56 +0000
2106) die("unable to create
load_cache_entries_thread");
ad570c680bc ( Ben Peart 2018-09-12 16:18:56 +0000
2109) if (++thread == nr_threads)
ad570c680bc ( Ben Peart 2018-09-12 16:18:56 +0000
2110) break;
9928a34be6b ( Ben Peart 2018-09-12 16:18:59 +0000
2113) ondisk = (struct ondisk_cache_entry *)(mmap + src_offset);
ad570c680bc ( Ben Peart 2018-09-12 16:18:56 +0000
2116) flags = get_be16(&ondisk->flags);
ad570c680bc ( Ben Peart 2018-09-12 16:18:56 +0000
2118) if (flags & CE_EXTENDED) {
ad570c680bc ( Ben Peart 2018-09-12 16:18:56 +0000
2120) ondisk2 = (struct ondisk_cache_entry_extended
*)ondisk;
ad570c680bc ( Ben Peart 2018-09-12 16:18:56 +0000
2121) name = ondisk2->name;
ad570c680bc ( Ben Peart 2018-09-12 16:18:56 +0000
2123) name = ondisk->name;
13e1b78757a (Nguyễn Thái Ngọc Duy 2018-09-12 16:18:57 +0000
2125) if (istate->version != 4) {
ad570c680bc ( Ben Peart 2018-09-12 16:18:56 +0000
2129) len = flags & CE_NAMEMASK;
ad570c680bc ( Ben Peart 2018-09-12 16:18:56 +0000
2130) if (len == CE_NAMEMASK)
ad570c680bc ( Ben Peart 2018-09-12 16:18:56 +0000
2131) len = strlen(name);
ad570c680bc ( Ben Peart 2018-09-12 16:18:56 +0000
2132) src_offset += (flags & CE_EXTENDED) ?
ad570c680bc ( Ben Peart 2018-09-12 16:18:56 +0000 2133)
ondisk_cache_entry_extended_size(len) :
ad570c680bc ( Ben Peart 2018-09-12 16:18:56 +0000
2134) ondisk_cache_entry_size(len);
ad570c680bc ( Ben Peart 2018-09-12 16:18:56 +0000
2136) src_offset += (name - ((char *)ondisk)) +
expand_name_field(previous_name, name);
ad570c680bc ( Ben Peart 2018-09-12 16:18:56 +0000 2139) for (i
= 0; i < nr_threads; i++) {
ad570c680bc ( Ben Peart 2018-09-12 16:18:56 +0000
2140) struct load_cache_entries_thread_data *p = data + i;
ad570c680bc ( Ben Peart 2018-09-12 16:18:56 +0000
2141) if (pthread_join(p->pthread, NULL))
ad570c680bc ( Ben Peart 2018-09-12 16:18:56 +0000
2142) die("unable to join load_cache_entries_thread");
ad570c680bc ( Ben Peart 2018-09-12 16:18:56 +0000
2143) mem_pool_combine(istate->ce_mem_pool, p->ce_mem_pool);
ad570c680bc ( Ben Peart 2018-09-12 16:18:56 +0000
2144) consumed += p->consumed;
ad570c680bc ( Ben Peart 2018-09-12 16:18:56 +0000 2147)
free(data);
ad570c680bc ( Ben Peart 2018-09-12 16:18:56 +0000 2148)
strbuf_release(&previous_name_buf);
ad570c680bc ( Ben Peart 2018-09-12 16:18:56 +0000 2150) return
consumed;
ad570c680bc ( Ben Peart 2018-09-12 16:18:56 +0000
2217) nr_threads = cpus;
ad570c680bc ( Ben Peart 2018-09-12 16:18:56 +0000
2222) nr_threads = 3;
4ee45bacad0 ( Ben Peart 2018-09-12 16:18:55 +0000
2225) extension_offset = read_eoie_extension(mmap, mmap_size);
4ee45bacad0 ( Ben Peart 2018-09-12 16:18:55 +0000
2226) if (extension_offset) {
4ee45bacad0 ( Ben Peart 2018-09-12 16:18:55 +0000
2228) p.src_offset = extension_offset;
4ee45bacad0 ( Ben Peart 2018-09-12 16:18:55 +0000
2229) if (pthread_create(&p.pthread, NULL,
load_index_extensions, &p))
4ee45bacad0 ( Ben Peart 2018-09-12 16:18:55 +0000
2230) die(_("unable to create
load_index_extensions_thread"));
ad570c680bc ( Ben Peart 2018-09-12 16:18:56 +0000
2231) nr_threads--;
ad570c680bc ( Ben Peart 2018-09-12 16:18:56 +0000
2235) src_offset += load_cache_entries_threaded(nr_threads,
istate, mmap, mmap_size, src_offset);
4ee45bacad0 ( Ben Peart 2018-09-12 16:18:55 +0000
2248) die(_("unable to join load_index_extensions_thread"));
9928a34be6b ( Ben Peart 2018-09-12 16:18:59 +0000 3288)static
unsigned long read_eoie_extension(const char *mmap, size_t mmap_size)
8d44e69943e ( Ben Peart 2018-09-12 16:18:53 +0000 3306) if
(mmap_size < sizeof(struct cache_header) + EOIE_SIZE_WITH_HEADER +
the_hash_algo->rawsz)
8d44e69943e ( Ben Peart 2018-09-12 16:18:53 +0000
3307) return 0;
8d44e69943e ( Ben Peart 2018-09-12 16:18:53 +0000 3310) index
= eoie = mmap + mmap_size - EOIE_SIZE_WITH_HEADER - the_hash_algo->rawsz;
8d44e69943e ( Ben Peart 2018-09-12 16:18:53 +0000 3311) if
(CACHE_EXT(index) != CACHE_EXT_ENDOFINDEXENTRIES)
8d44e69943e ( Ben Peart 2018-09-12 16:18:53 +0000
3312) return 0;
8d44e69943e ( Ben Peart 2018-09-12 16:18:53 +0000 3313) index
+= sizeof(uint32_t);
8d44e69943e ( Ben Peart 2018-09-12 16:18:53 +0000 3316)
extsize = get_be32(index);
8d44e69943e ( Ben Peart 2018-09-12 16:18:53 +0000 3317) if
(extsize != EOIE_SIZE)
8d44e69943e ( Ben Peart 2018-09-12 16:18:53 +0000
3318) return 0;
8d44e69943e ( Ben Peart 2018-09-12 16:18:53 +0000 3319) index
+= sizeof(uint32_t);
8d44e69943e ( Ben Peart 2018-09-12 16:18:53 +0000 3325) offset
= get_be32(index);
8d44e69943e ( Ben Peart 2018-09-12 16:18:53 +0000 3326) if
(mmap + offset < mmap + sizeof(struct cache_header))
8d44e69943e ( Ben Peart 2018-09-12 16:18:53 +0000
3327) return 0;
8d44e69943e ( Ben Peart 2018-09-12 16:18:53 +0000 3328) if
(mmap + offset >= eoie)
8d44e69943e ( Ben Peart 2018-09-12 16:18:53 +0000
3329) return 0;
8d44e69943e ( Ben Peart 2018-09-12 16:18:53 +0000 3330) index
+= sizeof(uint32_t);
8d44e69943e ( Ben Peart 2018-09-12 16:18:53 +0000 3341)
src_offset = offset;
8d44e69943e ( Ben Peart 2018-09-12 16:18:53 +0000 3342)
the_hash_algo->init_fn(&c);
8d44e69943e ( Ben Peart 2018-09-12 16:18:53 +0000 3343) while
(src_offset < mmap_size - the_hash_algo->rawsz - EOIE_SIZE_WITH_HEADER) {
9928a34be6b ( Ben Peart 2018-09-12 16:18:59 +0000
3351) extsize = get_be32(mmap + src_offset + 4);
8d44e69943e ( Ben Peart 2018-09-12 16:18:53 +0000
3354) if (src_offset + 8 + extsize < src_offset)
8d44e69943e ( Ben Peart 2018-09-12 16:18:53 +0000
3355) return 0;
8d44e69943e ( Ben Peart 2018-09-12 16:18:53 +0000
3357) the_hash_algo->update_fn(&c, mmap + src_offset, 8);
8d44e69943e ( Ben Peart 2018-09-12 16:18:53 +0000
3359) src_offset += 8;
8d44e69943e ( Ben Peart 2018-09-12 16:18:53 +0000
3360) src_offset += extsize;
8d44e69943e ( Ben Peart 2018-09-12 16:18:53 +0000 3362)
the_hash_algo->final_fn(hash, &c);
8d44e69943e ( Ben Peart 2018-09-12 16:18:53 +0000 3363) if
(hashcmp(hash, (const unsigned char *)index))
8d44e69943e ( Ben Peart 2018-09-12 16:18:53 +0000
3364) return 0;
8d44e69943e ( Ben Peart 2018-09-12 16:18:53 +0000 3367) if
(src_offset != mmap_size - the_hash_algo->rawsz - EOIE_SIZE_WITH_HEADER)
8d44e69943e ( Ben Peart 2018-09-12 16:18:53 +0000
3368) return 0;
8d44e69943e ( Ben Peart 2018-09-12 16:18:53 +0000 3370) return
offset;
rebase-interactive.c
64a43cbd5da (Alban Gruin 2018-08-10 18:51:31 +0200
61) return error_errno(_("could not read '%s'."), todo_file);
64a43cbd5da (Alban Gruin 2018-08-10 18:51:31 +0200
65) strbuf_release(&buf);
64a43cbd5da (Alban Gruin 2018-08-10 18:51:31 +0200
66) return -1;
a9f5476fbca (Alban Gruin 2018-08-10 18:51:35 +0200
74) return error_errno(_("could not read '%s'."), todo_file);
a9f5476fbca (Alban Gruin 2018-08-10 18:51:35 +0200
78) strbuf_release(&buf);
a9f5476fbca (Alban Gruin 2018-08-10 18:51:35 +0200
79) return -1;
64a43cbd5da (Alban Gruin 2018-08-10 18:51:31 +0200
85) return -1;
refs.c
4a6067cda51 (Stefan Beller 2018-08-20 18:24:16 +0000
1405) return 0;
revision.c
2abf3503854 (Nguyễn Thái Ngọc Duy 2018-09-21 17:57:38 +0200
1529) if (ce_path_match(istate, ce, &revs->prune_data, NULL)) {
2abf3503854 (Nguyễn Thái Ngọc Duy 2018-09-21 17:57:38 +0200
1535) while ((i+1 < istate->cache_nr) &&
2abf3503854 (Nguyễn Thái Ngọc Duy 2018-09-21 17:57:38 +0200
1536) ce_same_name(ce, istate->cache[i+1]))
4943d288495 (Derrick Stolee 2018-09-21 10:39:36 -0700
2931) return;
4943d288495 (Derrick Stolee 2018-09-21 10:39:36 -0700
2934) return;
4943d288495 (Derrick Stolee 2018-09-21 10:39:36 -0700
2937) c->object.flags |= UNINTERESTING;
4943d288495 (Derrick Stolee 2018-09-21 10:39:36 -0700
2940) return;
4943d288495 (Derrick Stolee 2018-09-21 10:39:36 -0700
2943) mark_parents_uninteresting(c);
4943d288495 (Derrick Stolee 2018-09-21 10:39:36 -0700
2966) return;
4943d288495 (Derrick Stolee 2018-09-21 10:39:36 -0700
2969) return;
4943d288495 (Derrick Stolee 2018-09-21 10:39:36 -0700
2974) return;
4943d288495 (Derrick Stolee 2018-09-21 10:39:36 -0700 3018) case
REV_SORT_BY_COMMIT_DATE:
4943d288495 (Derrick Stolee 2018-09-21 10:39:36 -0700
3019) info->topo_queue.compare = compare_commits_by_commit_date;
4943d288495 (Derrick Stolee 2018-09-21 10:39:36 -0700
3020) break;
4943d288495 (Derrick Stolee 2018-09-21 10:39:36 -0700 3021) case
REV_SORT_BY_AUTHOR_DATE:
4943d288495 (Derrick Stolee 2018-09-21 10:39:36 -0700 3022)
init_author_date_slab(&info->author_date);
4943d288495 (Derrick Stolee 2018-09-21 10:39:36 -0700
3023) info->topo_queue.compare = compare_commits_by_author_date;
4943d288495 (Derrick Stolee 2018-09-21 10:39:36 -0700
3024) info->topo_queue.cb_data = &info->author_date;
4943d288495 (Derrick Stolee 2018-09-21 10:39:36 -0700
3025) break;
4943d288495 (Derrick Stolee 2018-09-21 10:39:36 -0700
3038) continue;
4943d288495 (Derrick Stolee 2018-09-21 10:39:36 -0700 3048)
record_author_date(&info->author_date, c);
6c04ff30019 (Derrick Stolee 2018-09-21 10:39:32 -0700
3086) if (!revs->ignore_missing_links)
6c04ff30019 (Derrick Stolee 2018-09-21 10:39:32 -0700
3087) die("Failed to traverse parents of commit %s",
4943d288495 (Derrick Stolee 2018-09-21 10:39:36 -0700 3088)
oid_to_hex(&commit->object.oid));
4943d288495 (Derrick Stolee 2018-09-21 10:39:36 -0700
3096) continue;
sequencer.c
65850686cf0 (Alban Gruin 2018-08-28 14:10:40 +0200
2276) return;
65850686cf0 (Alban Gruin 2018-08-28 14:10:40 +0200
2373) write_file(rebase_path_quiet(), "%s\n", quiet);
2c58483a598 (Alban Gruin 2018-08-10 18:51:33 +0200
3371) return error(_("could not checkout %s"), commit);
4df66c40b08 (Alban Gruin 2018-08-10 18:51:34 +0200
3385) return error(_("%s: not a valid OID"), orig_head);
b97e1873649 (Alban Gruin 2018-08-28 14:10:36 +0200
4750) return -1;
b97e1873649 (Alban Gruin 2018-08-28 14:10:36 +0200
4753) return -1;
b97e1873649 (Alban Gruin 2018-08-28 14:10:36 +0200
4759) return error_errno(_("could not read '%s'."), todo_file);
b97e1873649 (Alban Gruin 2018-08-28 14:10:36 +0200
4762) todo_list_release(&todo_list);
b97e1873649 (Alban Gruin 2018-08-28 14:10:36 +0200
4763) return error(_("unusable todo list: '%s'"), todo_file);
b97e1873649 (Alban Gruin 2018-08-28 14:10:36 +0200
4782) todo_list_release(&todo_list);
b97e1873649 (Alban Gruin 2018-08-28 14:10:36 +0200
4783) return -1;
b97e1873649 (Alban Gruin 2018-08-28 14:10:36 +0200
4787) return error(_("could not copy '%s' to '%s'."), todo_file,
b97e1873649 (Alban Gruin 2018-08-28 14:10:36 +0200
4791) return error(_("could not transform the todo list"));
b97e1873649 (Alban Gruin 2018-08-28 14:10:36 +0200
4820) return error(_("could not transform the todo list"));
b97e1873649 (Alban Gruin 2018-08-28 14:10:36 +0200
4823) return error(_("could not skip unnecessary pick commands"));
b97e1873649 (Alban Gruin 2018-08-28 14:10:36 +0200
4829) return -1;
strbuf.c
f95736288a3 (Pratik Karki 2018-08-08 20:51:16 +0545
127) --sb->len;
wt-status.c
f3bd35fa0dd (Stephen P. Smith 2018-09-05 17:53:29 -0700
671) s->committable = 1;
jch..pu:
builtin/bisect--helper.c
0b1f0fd910c (Pranit Bauva 2017-10-27 15:06:37 +0000
43) free((void *) terms->term_good);
3d3237b0e6b (Pranit Bauva 2017-10-27 15:06:37 +0000
162) if (get_oid_commit(commit, &oid))
3d3237b0e6b (Pranit Bauva 2017-10-27 15:06:37 +0000
163) return error(_("'%s' is not a valid commit"),
commit);
3d3237b0e6b (Pranit Bauva 2017-10-27 15:06:37 +0000
164) strbuf_addstr(&branch, commit);
3d3237b0e6b (Pranit Bauva 2017-10-27 15:06:37 +0000
172) error(_("Could not check out original HEAD '%s'.
Try "
3d3237b0e6b (Pranit Bauva 2017-10-27 15:06:37 +0000
174) strbuf_release(&branch);
3d3237b0e6b (Pranit Bauva 2017-10-27 15:06:37 +0000
175) argv_array_clear(&argv);
3d3237b0e6b (Pranit Bauva 2017-10-27 15:06:37 +0000
176) return -1;
0b1f0fd910c (Pranit Bauva 2017-10-27 15:06:37 +0000
215) error(_("Bad bisect_write argument: %s"), state);
0b1f0fd910c (Pranit Bauva 2017-10-27 15:06:37 +0000
216) goto fail;
0b1f0fd910c (Pranit Bauva 2017-10-27 15:06:37 +0000
220) error(_("couldn't get the oid of the rev '%s'"), rev);
0b1f0fd910c (Pranit Bauva 2017-10-27 15:06:37 +0000
221) goto fail;
0b1f0fd910c (Pranit Bauva 2017-10-27 15:06:37 +0000
226) goto fail;
0b1f0fd910c (Pranit Bauva 2017-10-27 15:06:37 +0000
230) error_errno(_("couldn't open the file '%s'"),
git_path_bisect_log());
0b1f0fd910c (Pranit Bauva 2017-10-27 15:06:37 +0000
231) goto fail;
0b1f0fd910c (Pranit Bauva 2017-10-27 15:06:37 +0000 242)fail:
0b1f0fd910c (Pranit Bauva 2017-10-27 15:06:37 +0000 243) retval
= -1;
a919f328ba3 (Pranit Bauva 2017-10-27 15:06:37 +0000
323) yesno = git_prompt(_("Are you sure [Y/n]? "), PROMPT_ECHO);
a919f328ba3 (Pranit Bauva 2017-10-27 15:06:37 +0000
324) if (starts_with(yesno, "N") || starts_with(yesno, "n"))
a919f328ba3 (Pranit Bauva 2017-10-27 15:06:37 +0000
327) goto finish;
a919f328ba3 (Pranit Bauva 2017-10-27 15:06:37 +0000
336) error(_("You need to start by \"git bisect start\". You "
a919f328ba3 (Pranit Bauva 2017-10-27 15:06:37 +0000
341) goto fail;
a919f328ba3 (Pranit Bauva 2017-10-27 15:06:37 +0000 345)fail:
35f7ca528ae (Pranit Bauva 2017-10-27 15:06:37 +0000
387) return error(_("--bisect-term requires exactly one
argument"));
35f7ca528ae (Pranit Bauva 2017-10-27 15:06:37 +0000
400) error(_("BUG: invalid argument %s for 'git
bisect terms'.\n"
5dfeec316b8 (Pranit Bauva 2017-10-27 15:06:37 +0000
416) return -1;
5dfeec316b8 (Pranit Bauva 2017-10-27 15:06:37 +0000
419) goto fail;
5dfeec316b8 (Pranit Bauva 2017-10-27 15:06:37 +0000
423) goto fail;
5dfeec316b8 (Pranit Bauva 2017-10-27 15:06:37 +0000 427)fail:
5dfeec316b8 (Pranit Bauva 2017-10-27 15:06:37 +0000 428) retval
= -1;
5dfeec316b8 (Pranit Bauva 2017-10-27 15:06:37 +0000
466) no_checkout = 1;
5dfeec316b8 (Pranit Bauva 2017-10-27 15:06:37 +0000
488) !one_of(arg, "--term-good", "--term-bad", NULL)) {
5dfeec316b8 (Pranit Bauva 2017-10-27 15:06:37 +0000
489) return error(_("unrecognised option: '%s'"), arg);
5dfeec316b8 (Pranit Bauva 2017-10-27 15:06:37 +0000
523) if (get_oid("HEAD", &head_oid))
5dfeec316b8 (Pranit Bauva 2017-10-27 15:06:37 +0000
524) return error(_("Bad HEAD - I need a HEAD"));
5dfeec316b8 (Pranit Bauva 2017-10-27 15:06:37 +0000
539) error(_("checking out '%s' failed. Try
'git "
5dfeec316b8 (Pranit Bauva 2017-10-27 15:06:37 +0000
559) return error(_("won't bisect on
cg-seek'ed tree"));
5dfeec316b8 (Pranit Bauva 2017-10-27 15:06:37 +0000
562) return error(_("Bad HEAD - strange symbolic ref"));
5dfeec316b8 (Pranit Bauva 2017-10-27 15:06:37 +0000
570) return -1;
5dfeec316b8 (Pranit Bauva 2017-10-27 15:06:37 +0000
588) goto fail;
5dfeec316b8 (Pranit Bauva 2017-10-27 15:06:37 +0000
598) goto fail;
5dfeec316b8 (Pranit Bauva 2017-10-27 15:06:37 +0000
606) goto fail;
3d3237b0e6b (Pranit Bauva 2017-10-27 15:06:37 +0000
686) return error(_("--bisect-reset requires either
no argument or a commit"));
0b1f0fd910c (Pranit Bauva 2017-10-27 15:06:37 +0000
690) return error(_("--bisect-write requires either 4
or 5 arguments"));
20edf353b72 (Pranit Bauva 2017-10-27 15:06:37 +0000
697) return error(_("--check-and-set-terms requires 3
arguments"));
a919f328ba3 (Pranit Bauva 2017-10-27 15:06:37 +0000
703) return error(_("--bisect-next-check requires 2
or 3 arguments"));
builtin/blame.c
74e8221b523 (Linus Torvalds 2018-07-07 15:02:35 -0700 922) case
DATE_HUMAN:
74e8221b523 (Linus Torvalds 2018-07-07 15:02:35 -0700
924) blame_date_width = sizeof("Thu Oct 19 16:00");
74e8221b523 (Linus Torvalds 2018-07-07 15:02:35 -0700
925) break;
builtin/gc.c
3029970275b (Jonathan Nieder 2018-07-16 23:57:40 -0700
461) ret = error_errno(_("cannot stat '%s'"), gc_log_path);
3029970275b (Jonathan Nieder 2018-07-16 23:57:40 -0700
462) goto done;
3029970275b (Jonathan Nieder 2018-07-16 23:57:40 -0700
470) ret = error_errno(_("cannot read '%s'"), gc_log_path);
3029970275b (Jonathan Nieder 2018-07-16 23:57:40 -0700
585) exit(128);
builtin/submodule--helper.c
df255b8cac7 (Brandon Williams 2018-08-08 15:33:22 -0700
914) usage(_("git submodule--helper gitdir <name>"));
date.c
74e8221b523 (Linus Torvalds 2018-07-07 15:02:35 -0700
113) die("Timestamp too large for this system: %"PRItime, time);
74e8221b523 (Linus Torvalds 2018-07-07 15:02:35 -0700
216) if (tm->tm_mon == human_tm->tm_mon) {
74e8221b523 (Linus Torvalds 2018-07-07 15:02:35 -0700
217) if (tm->tm_mday > human_tm->tm_mday) {
74e8221b523 (Linus Torvalds 2018-07-07 15:02:35 -0700
219) } else if (tm->tm_mday == human_tm->tm_mday) {
74e8221b523 (Linus Torvalds 2018-07-07 15:02:35 -0700
220) hide.date = hide.wday = 1;
74e8221b523 (Linus Torvalds 2018-07-07 15:02:35 -0700
221) } else if (tm->tm_mday + 5 > human_tm->tm_mday) {
74e8221b523 (Linus Torvalds 2018-07-07 15:02:35 -0700
223) hide.date = 1;
74e8221b523 (Linus Torvalds 2018-07-07 15:02:35 -0700
231) gettimeofday(&now, NULL);
74e8221b523 (Linus Torvalds 2018-07-07 15:02:35 -0700
232) show_date_relative(time, tz, &now, buf);
74e8221b523 (Linus Torvalds 2018-07-07 15:02:35 -0700
233) return;
74e8221b523 (Linus Torvalds 2018-07-07 15:02:35 -0700
246) hide.seconds = 1;
74e8221b523 (Linus Torvalds 2018-07-07 15:02:35 -0700
247) hide.tz |= !hide.date;
74e8221b523 (Linus Torvalds 2018-07-07 15:02:35 -0700
248) hide.wday = hide.time = !hide.year;
74e8221b523 (Linus Torvalds 2018-07-07 15:02:35 -0700
262) strbuf_rtrim(buf);
74e8221b523 (Linus Torvalds 2018-07-07 15:02:35 -0700
287) gettimeofday(&now, NULL);
74e8221b523 (Linus Torvalds 2018-07-07 15:02:35 -0700
290) human_tz = local_time_tzoffset(now.tv_sec, &human_tm);
74e8221b523 (Linus Torvalds 2018-07-07 15:02:35 -0700 886)static int
auto_date_style(void)
74e8221b523 (Linus Torvalds 2018-07-07 15:02:35 -0700 888) return
(isatty(1) || pager_in_use()) ? DATE_HUMAN : DATE_NORMAL;
74e8221b523 (Linus Torvalds 2018-07-07 15:02:35 -0700
909) return DATE_HUMAN;
74e8221b523 (Linus Torvalds 2018-07-07 15:02:35 -0700
911) return auto_date_style();
sha1-array.c
7007a318a68 (Stefan Beller 2018-09-11 16:49:44 -0700
90) oidcpy(&array->oid[dst], &array->oid[src]);
string-list.c
6fecf7cd01a (Stefan Beller 2018-09-11 16:49:43 -0700
86) BUG("tried to remove an item from empty string list");
6fecf7cd01a (Stefan Beller 2018-09-11 16:49:43 -0700
89) free(list->items[list->nr - 1].string);
6fecf7cd01a (Stefan Beller 2018-09-11 16:49:43 -0700
92) free(list->items[list->nr - 1].util);
submodule-config.c
cc3e30d6aea (Antonio Ospite 2018-09-17 16:09:32 +0200
716) return CONFIG_INVALID_KEY;
89039393db8 (Antonio Ospite 2018-09-17 16:09:33 +0200
731) warning(_("Could not update .gitmodules entry %s"), key);
submodule.c
bab609b4dc1 (Stefan Beller 2018-09-11 16:49:50 -0700
1369) string_list_pop(&spf->retry, 0);
bab609b4dc1 (Stefan Beller 2018-09-11 16:49:50 -0700
1370) goto retry_next;
bab609b4dc1 (Stefan Beller 2018-09-11 16:49:50 -0700
1429) warning(_("Could not get submodule repository for
submodule '%s' in repository '%s'"),
bab609b4dc1 (Stefan Beller 2018-09-11 16:49:50 -0700
1430) sub->path, spf->r->worktree);
df255b8cac7 (Brandon Williams 2018-08-08 15:33:22 -0700
1884) die(_("could not create directory '%s'"), new_gitdir.buf);
wrapper.c
7e621449185 (Pranit Bauva 2017-10-27 15:06:37 +0000
701) die_errno(_("could not stat %s"), filename);
wt-status.c
1c623c066c2 (Junio C Hamano 2018-09-07 15:32:24 -0700
1949) if (s->state.rebase_in_progress ||
1c623c066c2 (Junio C Hamano 2018-09-07 15:32:24 -0700 1950)
s->state.rebase_interactive_in_progress)
1c623c066c2 (Junio C Hamano 2018-09-07 15:32:24 -0700
1951) branch_name = s->state.onto;
1c623c066c2 (Junio C Hamano 2018-09-07 15:32:24 -0700
1952) else if (s->state.detached_from)
1c623c066c2 (Junio C Hamano 2018-09-07 15:32:24 -0700
1953) branch_name = s->state.detached_from;
^ permalink raw reply [relevance 4%]
* Re: [PATCH 2/8] Add a place for (not) sharing stuff between worktrees
2018-09-25 16:55 5% ` Duy Nguyen
@ 2018-09-25 17:56 4% ` Stefan Beller
0 siblings, 0 replies; 200+ results
From: Stefan Beller @ 2018-09-25 17:56 UTC (permalink / raw)
To: Duy Nguyen; +Cc: git, Elijah Newren, Jeff King
On Tue, Sep 25, 2018 at 9:55 AM Duy Nguyen <pclouds@gmail.com> wrote:
> > And with that said, I wonder if the "local" part should be feature agnostic,
> > or if we want to be "local" for worktrees, "local" for remotes, "local"
> > for submodules (i.e. our own refs vs submodule refs).
>
> You lost me here.
Yeah, me too after rereading. :P
I think the "local" part always implies that there is a part that is
not local and depending on the feature you call it remote or other
worktree.
When writing this comment I briefly wondered if we want to combine
the local aspects of the various features.
However the "local" part really depends on the feature
(e.g. a ref on a different worktree is still local from the here/remote
perspective or from the superproject/submodule perspective),
so I think I was misguided.
> > > think as long as the word "worktree" is in there, people would notice
> > > the difference.
> >
> > That makes sense. But is refs/worktree shared or local? It's not quite
> > obvious to me, as I could have refs/worktree/<worktree-name>/master
> > instead when it is shared, so I tend to favor refs/local-worktree/ a bit
> > more, but that is more typing. :/
>
> OK I think mixing the two patches will different purposes messes you
> (or me) up ;-)
possible.
>
> refs/worktrees/xxx (and refs/main/xxx) are about visibility from other
> worktrees. Or like Eric put it, they are simply aliases. These refs
> are not shared because if they are, you can already see them without
> new "ref mount points" like this.
>
> refs/worktree (previously refs/local) is also per-worktree but it's
> specifically because you can't have per-worktree inside "refs/" (the
> only exception so far is refs/bisect which is hard coded). You can
> have refs outside "refs/" (like HEAD or FETCH_HEAD) and they will not
> be shared, but they cannot be iterated while those inside refs/ can
> be. This is more about deciding what to share and I believe is really
> worktree-specific and only matters to _current_ worktree.
>
> Since refs/worktree is per-worktree, you can also view them from a
> different worktree via refs/worktrees/. E.g. if you have
> refs/worktree/foo then another worktree can see it via
> refs/worktrees/xxx/refs/worktree/foo (besides pseudo refs like
> refs/worktrees/xxx/HEAD)
Ah. now I seem to understand, thanks for explaining.
^ permalink raw reply [relevance 4%]
* Re: [PATCH 2/8] Add a place for (not) sharing stuff between worktrees
2018-09-25 16:24 5% ` Stefan Beller
@ 2018-09-25 16:55 5% ` Duy Nguyen
2018-09-25 17:56 4% ` Stefan Beller
0 siblings, 1 reply; 200+ results
From: Duy Nguyen @ 2018-09-25 16:55 UTC (permalink / raw)
To: Stefan Beller; +Cc: Git Mailing List, Elijah Newren, Jeff King
On Tue, Sep 25, 2018 at 6:24 PM Stefan Beller <sbeller@google.com> wrote:
> > > That sounds dangerous to me. There is already a concept of
> > > local and remote-tracking branches. So I would think that local
> > > may soon become an overused word, (just like "index" today or
> > > "recursive" to a lesser extend).
> > >
> > > Could this special area be more explicit?
> > > (refs/worktree-local/ ? or after peeking at the docs below
> > > refs/un-common/ ?)
> >
> > refs/un-common sounds really "uncommon" :D. If refs/local is bad, I
> > guess we could go with either refs/worktree-local, refs/worktree,
> > refs/private, refs/per-worktree... My vote is on refs/worktree. I
>
> refs/worktree sounds good to me (I do not object), but I am not
> overly enthused either, as when I think further worktrees and
> submodules are both features with a very similar nature in that
> they touch a lot of core concepts in Git, but seem to be a niche
> feature for the masses for now.
I think the similarity is partly because submodule feature also has to
manage worktrees. My view is at some point, this "git worktree" would
be good enough that it can handle submodules as well (for the worktree
part only of course)
> For example I could think of submodules following this addressing
> mode as well: submodule/<path>/master sounds similar to the
> originally proposed worktree/<name>/<branch> convention.
> For now it is not quite clear to me why you would want to have
> access to the submodule refs in the superproject, but maybe
> the use case will come later.
Yeah. In theory we could "mount" the submodule ref store to a
superproject's ref store. I think it may be needed just for the same
reason it's needed for worktree: error reporting. If you peek into a
submodule and say "HEAD has an error", the user will get confused
whether it's superproject's HEAD or a submodule's HEAD.
> And with that said, I wonder if the "local" part should be feature agnostic,
> or if we want to be "local" for worktrees, "local" for remotes, "local"
> for submodules (i.e. our own refs vs submodule refs).
You lost me here.
>
> > think as long as the word "worktree" is in there, people would notice
> > the difference.
>
> That makes sense. But is refs/worktree shared or local? It's not quite
> obvious to me, as I could have refs/worktree/<worktree-name>/master
> instead when it is shared, so I tend to favor refs/local-worktree/ a bit
> more, but that is more typing. :/
OK I think mixing the two patches will different purposes messes you
(or me) up ;-)
refs/worktrees/xxx (and refs/main/xxx) are about visibility from other
worktrees. Or like Eric put it, they are simply aliases. These refs
are not shared because if they are, you can already see them without
new "ref mount points" like this.
refs/worktree (previously refs/local) is also per-worktree but it's
specifically because you can't have per-worktree inside "refs/" (the
only exception so far is refs/bisect which is hard coded). You can
have refs outside "refs/" (like HEAD or FETCH_HEAD) and they will not
be shared, but they cannot be iterated while those inside refs/ can
be. This is more about deciding what to share and I believe is really
worktree-specific and only matters to _current_ worktree.
Since refs/worktree is per-worktree, you can also view them from a
different worktree via refs/worktrees/. E.g. if you have
refs/worktree/foo then another worktree can see it via
refs/worktrees/xxx/refs/worktree/foo (besides pseudo refs like
refs/worktrees/xxx/HEAD)
> As we grow the worktree feature, do we ever expect the need to
> reference the current worktree?
>
> For example when there is a ref "test" that could be unique per
> repo and in the common area, so refs/heads/test would describe
> it and "test" would get there in DWIM mode.
>
> But then I could also delete the common ref and recreate a "test"
> ref in worktree A, in worktree B however DWIMming "test" could still
> refer to A's "test" as it is unique (so far) in the repository.
> And maybe I would want to check if test exists locally, so I'd
> want to ask for "self/test" (with "self" == "B" as that is my cwd).
You probably lost me again. In theory we must be able to detect
ambiguity and stop DWIMing. If you want to be ambiguity-free, you
specify full ref name, starting with "refs/" which should function
like "self/" because worktree design so far is always about the
current worktree's view.
--
Duy
^ permalink raw reply [relevance 5%]
* Re: [PATCH 2/8] Add a place for (not) sharing stuff between worktrees
@ 2018-09-25 16:24 5% ` Stefan Beller
2018-09-25 16:55 5% ` Duy Nguyen
0 siblings, 1 reply; 200+ results
From: Stefan Beller @ 2018-09-25 16:24 UTC (permalink / raw)
To: Duy Nguyen; +Cc: git, Elijah Newren, Jeff King
On Tue, Sep 25, 2018 at 8:36 AM Duy Nguyen <pclouds@gmail.com> wrote:
>
> On Tue, Sep 25, 2018 at 4:35 AM Stefan Beller <sbeller@google.com> wrote:
> >
> > On Sat, Sep 22, 2018 at 11:05 AM Nguyễn Thái Ngọc Duy <pclouds@gmail.com> wrote:
> > >
> > > When multiple worktrees are used, we need rules to determine if
> > > something belongs to one worktree or all of them. Instead of keeping
> > > adding rules when new stuff comes, have a generic rule:
> > >
> > > - Inside $GIT_DIR, which is per-worktree by default, add
> > > $GIT_DIR/common which is always shared. New features that want to
> > > share stuff should put stuff under this directory.
> >
> > So that /common is a directory and you have to use it specifically
> > in new code? That would be easy to overlook when coming up
> > with $GIT_DIR/foo for implementing the git-foo.
>
> There's no easy way out. I have to do _something_ if you want to share
> $GIT_DIR/foo to all worktrees. Either we have to update path.c and add
> "foo" which is not even an option for external commands, or we put
> "foo" in a common place, e.g. $GIT_DIR/common/foo.
>
> > > - Inside refs/, which is shared by default except refs/bisect, add
> > > refs/local/ which is per-worktree. We may eventually move
> > > refs/bisect to this new location and remove the exception in refs
> > > code.
> >
> > That sounds dangerous to me. There is already a concept of
> > local and remote-tracking branches. So I would think that local
> > may soon become an overused word, (just like "index" today or
> > "recursive" to a lesser extend).
> >
> > Could this special area be more explicit?
> > (refs/worktree-local/ ? or after peeking at the docs below
> > refs/un-common/ ?)
>
> refs/un-common sounds really "uncommon" :D. If refs/local is bad, I
> guess we could go with either refs/worktree-local, refs/worktree,
> refs/private, refs/per-worktree... My vote is on refs/worktree. I
refs/worktree sounds good to me (I do not object), but I am not
overly enthused either, as when I think further worktrees and
submodules are both features with a very similar nature in that
they touch a lot of core concepts in Git, but seem to be a niche
feature for the masses for now.
For example I could think of submodules following this addressing
mode as well: submodule/<path>/master sounds similar to the
originally proposed worktree/<name>/<branch> convention.
For now it is not quite clear to me why you would want to have
access to the submodule refs in the superproject, but maybe
the use case will come later.
And with that said, I wonder if the "local" part should be feature agnostic,
or if we want to be "local" for worktrees, "local" for remotes, "local"
for submodules (i.e. our own refs vs submodule refs).
> think as long as the word "worktree" is in there, people would notice
> the difference.
That makes sense. But is refs/worktree shared or local? It's not quite
obvious to me, as I could have refs/worktree/<worktree-name>/master
instead when it is shared, so I tend to favor refs/local-worktree/ a bit
more, but that is more typing. :/
==
As we grow the worktree feature, do we ever expect the need to
reference the current worktree?
For example when there is a ref "test" that could be unique per
repo and in the common area, so refs/heads/test would describe
it and "test" would get there in DWIM mode.
But then I could also delete the common ref and recreate a "test"
ref in worktree A, in worktree B however DWIMming "test" could still
refer to A's "test" as it is unique (so far) in the repository.
And maybe I would want to check if test exists locally, so I'd
want to ask for "self/test" (with "self" == "B" as that is my cwd).
Stefan
^ permalink raw reply [relevance 5%]
* Re: [PATCH] submodule.c: Make get_superproject_working_tree() work when supermodule has unmerged changes of the submodule reference
@ 2018-09-25 1:24 7% ` Stefan Beller
0 siblings, 0 replies; 200+ results
From: Stefan Beller @ 2018-09-25 1:24 UTC (permalink / raw)
To: sammck; +Cc: git, smckelvie
On Mon, Sep 24, 2018 at 2:25 PM Sam McKelvie <sammck@gmail.com> wrote:
Thanks for writing a patch!
I wonder if we can shorten the subject and make it a bit more concise,
how about
submodule: Alllow staged changes for get_superproject_working_tree
or
rev-parse: allow staged submodule in --show-superproject-working-tree
> Previously, "fatal: BUG: returned path string doesn't match cwd?" is displayed and
> git-rev-parse aborted.
Usually it is hard to read, continuing from the commit title to the body
of the commit message, as the title may be displayed differently or
somewhere else, so it would be good to restate it or rather state the
command that is broken, which we are trying to fix.
Invoking 'git rev-parse --show-superproject-working-tree' exits with
fatal: BUG ...
instead of displaying the superproject working tree when ....
> The problem is due to the fact that when a merge of the submodule reference is in progress,
> "git --stage —full-name <submodule-relative-path>” returns three seperate entries for the
"ls-files" is missing in that git invocation?
> submodule (one for each stage) rather than a single entry; e.g.,
>
> $ git ls-files --stage --full-name submodule-child-test
> 160000 dbbd2766fa330fa741ea59bb38689fcc2d283ac5 1 submodule-child-test
> 160000 f174d1dbfe863a59692c3bdae730a36f2a788c51 2 submodule-child-test
> 160000 e6178f3a58b958543952e12824aa2106d560f21d 3 submodule-child-test
>
> The code in get_superproject_working_tree() expected exactly one entry to be returned;
> this patch makes it use the first entry if multiple entries are returned.
The code looks good.
I wonder if we want to add a test for it, see t/t1500-rev-parse.sh
(at the very end 'showing the superproject correctly').
>
> Signed-off-by: Sam McKelvie <smckelvie@xevo.com>
Side note: the commit message seems to be indented,
but the patch applies fine, which makes me curious:
How did you produce&send the patch?
(Usually a patch would not apply as white spaces are
mangled in the code for example in some email setups)
(If I had to guess I could imagine that it is the output of
git-show as that indents the commit message usually,
see git-format-patch for a better tool)
Thanks for writing the patch!
Stefan
^ permalink raw reply [relevance 7%]
* Re: [PATCH v5 1/9] submodule: add a print_config_from_gitmodules() helper
@ 2018-09-24 23:06 4% ` Stefan Beller
0 siblings, 0 replies; 200+ results
From: Stefan Beller @ 2018-09-24 23:06 UTC (permalink / raw)
To: Antonio Ospite
Cc: Junio C Hamano, git, Brandon Williams, Daniel Graña,
Jonathan Nieder, Richard Hartmann,
Ævar Arnfjörð Bjarmason
> > +int print_config_from_gitmodules(const char *key)
>
> I am thinking about adding a "struct repository" argument to this
> function
Sounds like a good idea.
^ permalink raw reply [relevance 4%]
* Re: [PATCH v5 9/9] submodule: support reading .gitmodules when it's not in the working tree
@ 2018-09-24 21:00 7% ` Stefan Beller
2018-09-27 14:44 7% ` Antonio Ospite
0 siblings, 1 reply; 200+ results
From: Stefan Beller @ 2018-09-24 21:00 UTC (permalink / raw)
To: Antonio Ospite
Cc: SZEDER Gábor, Junio C Hamano, git, Brandon Williams,
Daniel Graña, Jonathan Nieder, Richard Hartmann,
Ævar Arnfjörð Bjarmason
On Mon, Sep 24, 2018 at 3:20 AM Antonio Ospite <ao2@ao2.it> wrote:
> > The third call, however, looks at the nested submodule at
> > 'submodule/sub', which doesn't contain a '.gitmodules' file. So this
> > function goes on with the second condition and calls
> > get_oid(GITMODULES_INDEX, &oid), which then appears to find the blob
> > in the _superproject's_ index.
> >
>
> You are correct.
>
> This is a limitation of the object store in git, there is no equivalent
> of get_oid() to get the oid from a specific repository and this affects
> config_with_options too when the config source is a blob.
Not yet, as there is a big push to pass-through an object-store object
or similar recently and rely less on global variables.
I am not sure I get to this code, though.
> This does not affect commands called via "git -C submodule_dir cmd"
> because in that case the chdir happens before the_repository is set up,
> for instance "git-submodule $SOMETHING --recursive" commands seem to
> change the working directory before the recursion.
For this it may be worth looking into the option
--super-prefix=<path>
Currently for internal use only. Set a prefix which gives a
path from above a repository down to its root. One use is
to give submodules context about the superproject that
invoked it.
the whole motion of moving to in-process deprecates this clunky
API to pass around strings to subprocesses.
> The test suite passes even after removing repo_read_gitmodules()
> entirely from builtin/grep.c, but I am still not confident that I get
> all the implication of why that call was originally added in commit
> f9ee2fcdfa (grep: recurse in-process using 'struct repository',
> 2017-08-02).
If you checkout that commit and remove the call to repo_read_gitmodules
and then call git-grep in a superproject with nested submodules, you
get a segfault.
On master (and deleting out that line) you do not get the segfault,
I think praise goes to ff6f1f564c4 (submodule-config: lazy-load a
repository's .gitmodules file, 2017-08-03) which happened shortly
after f9ee2fcdfa.
It showcased that it worked by converting ls-files, but left out grep.
So I think based on ff6f1f564c4 it is safe to remove all calls to
repo_read_gitmodules.
> Anyways, even if we removed the call we would prevent the problem from
> happening in the test suite, but not in the real world, in case non-leaf
> submodules without .gitmodules in their working tree.
Quite frankly I think grep was just overlooked in review of
https://public-inbox.org/git/20170803182000.179328-14-bmwill@google.com/
Stefan
^ permalink raw reply [relevance 7%]
* Re: "git rev-parse --show-superproject-working-tree" fails with "fatal: BUG: returned path string doesn't match cwd?" if supermodule has unmerged changes of the submodule reference
@ 2018-09-24 19:15 4% ` Stefan Beller
0 siblings, 0 replies; 200+ results
From: Stefan Beller @ 2018-09-24 19:15 UTC (permalink / raw)
To: sammck; +Cc: git
On Mon, Sep 24, 2018 at 10:59 AM Sam McKelvie <sammck@gmail.com> wrote:
>
> I experienced this problem using git 2.17.1; however, from inspection of the next branch, function get_superproject_working_tree() in submodule.c has not changed in 2 years.
>
> I believe the problem is related to the fact that when a merge of the submodule reference is in progress, "git --stage —full-name <submodule-relative-path>” returns three seperate entries for the submodule (one for each stage) rather than a single entry; e.g.,
>
> $ git ls-files --stage --full-name submodule-child-test
> 160000 dbbd2766fa330fa741ea59bb38689fcc2d283ac5 1 submodule-child-test
> 160000 f174d1dbfe863a59692c3bdae730a36f2a788c51 2 submodule-child-test
> 160000 e6178f3a58b958543952e12824aa2106d560f21d 3 submodule-child-test
>
> The code in get_superproject_working_tree() uses the “-z” option on ls-files, so it expects null-byte termination between entries. However, the computation of super_sub_len:
>
> super_sub_len = sb.buf + sb.len - super_sub - 1;
>
> will only work when there is exactly one entry returned. If this line is changed to:
>
> super_sub_len = strlen(super_sub);
>
> then only the first returned entry is used, and the bug is resolved.
>
> strlen() should be safe to use here because strbuf_read ensures the result buffer is null-terminated.
This is good analysis of the issue. Thanks for writing it up!
Would you also mind to send a patch fixing the problem?
I agree that using strlen should work. I do not recall why I
did not use it at the time of writing it.
Thanks,
Stefan
^ permalink raw reply [relevance 4%]
* [PATCH 8/8] builtin/fetch: check for submodule updates for non branch fetches
2018-09-21 22:35 9% [PATCHv3 0/8] fetch: make sure submodule oids are fetched Stefan Beller
` (6 preceding siblings ...)
2018-09-21 22:35 23% ` [PATCH 7/8] fetch: retry fetching submodules if needed objects were not fetched Stefan Beller
@ 2018-09-21 22:35 21% ` Stefan Beller
7 siblings, 0 replies; 200+ results
From: Stefan Beller @ 2018-09-21 22:35 UTC (permalink / raw)
To: git; +Cc: Stefan Beller
Gerrit, the code review tool, has a different workflow than our mailing
list based approach. Usually users upload changes to a Gerrit server and
continuous integration and testing happens by bots. Sometimes however a
user wants to checkout a change locally and look at it locally. For this
use case, Gerrit offers a command line snippet to copy and paste to your
terminal, which looks like
git fetch https://<host>/gerrit refs/changes/<id> &&
git checkout FETCH_HEAD
For Gerrit changes that contain changing submodule gitlinks, it would be
easy to extend both the fetch and checkout with the '--recurse-submodules'
flag, such that this command line snippet would produce the state of a
change locally.
However the functionality added in the previous patch, which would
ensure that we fetch the objects in the submodule that the gitlink pointed
at, only works for remote tracking branches so far, not for FETCH_HEAD.
Make sure that fetching a superproject to its FETCH_HEAD, also respects
the existence checks for objects in the submodule recursion.
Signed-off-by: Stefan Beller <sbeller@google.com>
---
builtin/fetch.c | 5 ++++-
t/t5526-fetch-submodules.sh | 2 +-
2 files changed, 5 insertions(+), 2 deletions(-)
diff --git a/builtin/fetch.c b/builtin/fetch.c
index e3b03ad3bd3..f2d9e548bf0 100644
--- a/builtin/fetch.c
+++ b/builtin/fetch.c
@@ -894,11 +894,14 @@ static int store_updated_refs(const char *raw_url, const char *remote_name,
rc |= update_local_ref(ref, what, rm, ¬e,
summary_width);
free(ref);
- } else
+ } else {
+ check_for_new_submodule_commits(&rm->old_oid);
format_display(¬e, '*',
*kind ? kind : "branch", NULL,
*what ? what : "HEAD",
"FETCH_HEAD", summary_width);
+ }
+
if (note.len) {
if (verbosity >= 0 && !shown_url) {
fprintf(stderr, _("From %.*s\n"),
diff --git a/t/t5526-fetch-submodules.sh b/t/t5526-fetch-submodules.sh
index af12c50e7dd..a509eabb044 100755
--- a/t/t5526-fetch-submodules.sh
+++ b/t/t5526-fetch-submodules.sh
@@ -615,7 +615,7 @@ test_expect_success "fetch new commits on-demand when they are not reachable" '
git update-ref refs/changes/2 $D &&
(
cd downstream &&
- git fetch --recurse-submodules --recurse-submodules-default on-demand origin refs/changes/2:refs/heads/my_branch &&
+ git fetch --recurse-submodules origin refs/changes/2 &&
git -C submodule cat-file -t $C &&
git checkout --recurse-submodules FETCH_HEAD
)
--
2.19.0.444.g18242da7ef-goog
^ permalink raw reply related [relevance 21%]
* [PATCH 7/8] fetch: retry fetching submodules if needed objects were not fetched
2018-09-21 22:35 9% [PATCHv3 0/8] fetch: make sure submodule oids are fetched Stefan Beller
` (5 preceding siblings ...)
2018-09-21 22:35 26% ` [PATCH 6/8] submodule: fetch in submodules git directory instead of in worktree Stefan Beller
@ 2018-09-21 22:35 23% ` Stefan Beller
2018-09-21 22:35 21% ` [PATCH 8/8] builtin/fetch: check for submodule updates for non branch fetches Stefan Beller
7 siblings, 0 replies; 200+ results
From: Stefan Beller @ 2018-09-21 22:35 UTC (permalink / raw)
To: git; +Cc: Stefan Beller
Currently when git-fetch is asked to recurse into submodules, it dispatches
a plain "git-fetch -C <submodule-dir>" (with some submodule related options
such as prefix and recusing strategy, but) without any information of the
remote or the tip that should be fetched.
This works surprisingly well in some workflows (such as using submodules
as a third party library), while not so well in other scenarios, such
as in a Gerrit topic-based workflow, that can tie together changes
(potentially across repositories) on the server side. One of the parts
of such a Gerrit workflow is to download a change when wanting to examine
it, and you'd want to have its submodule changes that are in the same
topic downloaded as well. However these submodule changes reside in their
own repository in their own ref (refs/changes/<int>).
Retry fetching a submodule by object name if the object id that the
superproject points to, cannot be found.
This retrying does not happen when the "git fetch" done at the
superproject is not storing the fetched results in remote
tracking branches (i.e. instead just recording them to
FETCH_HEAD) in this step. A later patch will fix this.
builtin/fetch used to only inspect submodules when they were fetched
"on-demand", as in either on/off case it was clear whether the submodule
needs to be fetched. However to know whether we need to try fetching the
object ids, we need to identify the object names, which is done in this
function check_for_new_submodule_commits(), so we'll also run that code
in case the submodule recursion is set to "on".
Signed-off-by: Stefan Beller <sbeller@google.com>
---
builtin/fetch.c | 9 +-
submodule.c | 182 ++++++++++++++++++++++++++++++------
t/t5526-fetch-submodules.sh | 16 ++++
3 files changed, 174 insertions(+), 33 deletions(-)
diff --git a/builtin/fetch.c b/builtin/fetch.c
index 0696abfc2a1..e3b03ad3bd3 100644
--- a/builtin/fetch.c
+++ b/builtin/fetch.c
@@ -707,8 +707,7 @@ static int update_local_ref(struct ref *ref,
what = _("[new ref]");
}
- if ((recurse_submodules != RECURSE_SUBMODULES_OFF) &&
- (recurse_submodules != RECURSE_SUBMODULES_ON))
+ if (recurse_submodules != RECURSE_SUBMODULES_OFF)
check_for_new_submodule_commits(&ref->new_oid);
r = s_update_ref(msg, ref, 0);
format_display(display, r ? '!' : '*', what,
@@ -723,8 +722,7 @@ static int update_local_ref(struct ref *ref,
strbuf_add_unique_abbrev(&quickref, ¤t->object.oid, DEFAULT_ABBREV);
strbuf_addstr(&quickref, "..");
strbuf_add_unique_abbrev(&quickref, &ref->new_oid, DEFAULT_ABBREV);
- if ((recurse_submodules != RECURSE_SUBMODULES_OFF) &&
- (recurse_submodules != RECURSE_SUBMODULES_ON))
+ if (recurse_submodules != RECURSE_SUBMODULES_OFF)
check_for_new_submodule_commits(&ref->new_oid);
r = s_update_ref("fast-forward", ref, 1);
format_display(display, r ? '!' : ' ', quickref.buf,
@@ -738,8 +736,7 @@ static int update_local_ref(struct ref *ref,
strbuf_add_unique_abbrev(&quickref, ¤t->object.oid, DEFAULT_ABBREV);
strbuf_addstr(&quickref, "...");
strbuf_add_unique_abbrev(&quickref, &ref->new_oid, DEFAULT_ABBREV);
- if ((recurse_submodules != RECURSE_SUBMODULES_OFF) &&
- (recurse_submodules != RECURSE_SUBMODULES_ON))
+ if (recurse_submodules != RECURSE_SUBMODULES_OFF)
check_for_new_submodule_commits(&ref->new_oid);
r = s_update_ref("forced-update", ref, 1);
format_display(display, r ? '!' : '+', quickref.buf,
diff --git a/submodule.c b/submodule.c
index 8829e8d4eff..ba98a94d4db 100644
--- a/submodule.c
+++ b/submodule.c
@@ -1128,8 +1128,12 @@ struct submodule_parallel_fetch {
int result;
struct string_list changed_submodule_names;
+ struct get_next_submodule_task **retry;
+ int retry_nr, retry_alloc;
};
-#define SPF_INIT {0, ARGV_ARRAY_INIT, NULL, NULL, 0, 0, 0, 0, STRING_LIST_INIT_DUP }
+#define SPF_INIT {0, ARGV_ARRAY_INIT, NULL, NULL, 0, 0, 0, 0, \
+ STRING_LIST_INIT_DUP, \
+ NULL, 0, 0}
static void calculate_changed_submodule_paths(
struct submodule_parallel_fetch *spf)
@@ -1234,6 +1238,56 @@ static int get_fetch_recurse_config(const struct submodule *submodule,
return spf->default_option;
}
+struct get_next_submodule_task {
+ struct repository *repo;
+ const struct submodule *sub;
+ unsigned free_sub : 1; /* Do we need to free the submodule? */
+ struct oid_array *commits;
+};
+
+static const struct submodule *get_default_submodule(const char *path)
+{
+ struct submodule *ret = NULL;
+ const char *name = default_name_or_path(path);
+
+ if (!name)
+ return NULL;
+
+ ret = xmalloc(sizeof(*ret));
+ memset(ret, 0, sizeof(*ret));
+ ret->path = name;
+ ret->name = name;
+
+ return (const struct submodule *) ret;
+}
+
+static struct get_next_submodule_task *get_next_submodule_task_create(
+ struct repository *r, const char *path)
+{
+ struct get_next_submodule_task *task = xmalloc(sizeof(*task));
+ memset(task, 0, sizeof(*task));
+
+ task->sub = submodule_from_path(r, &null_oid, path);
+ if (!task->sub) {
+ task->sub = get_default_submodule(path);
+ task->free_sub = 1;
+ }
+
+ return task;
+}
+
+static void get_next_submodule_task_release(struct get_next_submodule_task *p)
+{
+ if (p->free_sub)
+ free((void*)p->sub);
+ p->free_sub = 0;
+ p->sub = NULL;
+
+ if (p->repo)
+ repo_clear(p->repo);
+ FREE_AND_NULL(p->repo);
+}
+
static struct repository *get_submodule_repo_for(struct repository *r,
const char *path)
{
@@ -1256,39 +1310,35 @@ static struct repository *get_submodule_repo_for(struct repository *r,
static int get_next_submodule(struct child_process *cp,
struct strbuf *err, void *data, void **task_cb)
{
- int ret = 0;
struct submodule_parallel_fetch *spf = data;
for (; spf->count < spf->r->index->cache_nr; spf->count++) {
- struct strbuf submodule_prefix = STRBUF_INIT;
+ int recurse_config;
const struct cache_entry *ce = spf->r->index->cache[spf->count];
const char *default_argv;
- const struct submodule *submodule;
- struct repository *repo;
- struct submodule default_submodule = SUBMODULE_INIT;
+ struct get_next_submodule_task *task;
if (!S_ISGITLINK(ce->ce_mode))
continue;
- submodule = submodule_from_path(spf->r, &null_oid, ce->name);
- if (!submodule) {
- const char *name = default_name_or_path(ce->name);
- if (name) {
- default_submodule.path = name;
- default_submodule.name = name;
- submodule = &default_submodule;
- }
+ task = get_next_submodule_task_create(spf->r, ce->name);
+
+ if (!task->sub) {
+ free(task);
+ continue;
}
- switch (get_fetch_recurse_config(submodule, spf))
+ recurse_config = get_fetch_recurse_config(task->sub, spf);
+
+ switch (recurse_config)
{
default:
case RECURSE_SUBMODULES_DEFAULT:
case RECURSE_SUBMODULES_ON_DEMAND:
- if (!submodule ||
+ if (!task->sub ||
!string_list_lookup(
&spf->changed_submodule_names,
- submodule->name))
+ task->sub->name))
continue;
default_argv = "on-demand";
break;
@@ -1299,12 +1349,13 @@ static int get_next_submodule(struct child_process *cp,
continue;
}
- strbuf_addf(&submodule_prefix, "%s%s/", spf->prefix, ce->name);
- repo = get_submodule_repo_for(spf->r, ce->name);
- if (repo) {
+ task->repo = get_submodule_repo_for(spf->r,
+ task->sub->path);
+ if (task->repo) {
+ struct strbuf submodule_prefix = STRBUF_INIT;
child_process_init(cp);
prepare_submodule_repo_env_in_gitdir(&cp->env_array);
- cp->dir = xstrdup(repo->gitdir);
+ cp->dir = task->repo->gitdir;
cp->git_cmd = 1;
if (!spf->quiet)
strbuf_addf(err, "Fetching submodule %s%s\n",
@@ -1313,18 +1364,51 @@ static int get_next_submodule(struct child_process *cp,
argv_array_pushv(&cp->args, spf->args.argv);
argv_array_push(&cp->args, default_argv);
argv_array_push(&cp->args, "--submodule-prefix");
+
+ strbuf_addf(&submodule_prefix, "%s%s/",
+ spf->prefix,
+ task->sub->path);
argv_array_push(&cp->args, submodule_prefix.buf);
- repo_clear(repo);
- free(repo);
- ret = 1;
- }
- strbuf_release(&submodule_prefix);
- if (ret) {
spf->count++;
+ *task_cb = task;
+
+ strbuf_release(&submodule_prefix);
return 1;
+ } else {
+ get_next_submodule_task_release(task);
+ free(task);
}
}
+
+ if (spf->retry_nr) {
+ struct get_next_submodule_task *task = spf->retry[spf->retry_nr - 1];
+ struct strbuf submodule_prefix = STRBUF_INIT;
+ spf->retry_nr--;
+
+ strbuf_addf(&submodule_prefix, "%s%s/", spf->prefix, task->sub->path);
+
+ child_process_init(cp);
+ prepare_submodule_repo_env_in_gitdir(&cp->env_array);
+ cp->git_cmd = 1;
+ cp->dir = task->repo->gitdir;
+
+ argv_array_init(&cp->args);
+ argv_array_pushv(&cp->args, spf->args.argv);
+ argv_array_push(&cp->args, "on-demand");
+ argv_array_push(&cp->args, "--submodule-prefix");
+ argv_array_push(&cp->args, submodule_prefix.buf);
+
+ /* NEEDSWORK: have get_default_remote from s--h */
+ argv_array_push(&cp->args, "origin");
+ oid_array_for_each_unique(task->commits,
+ append_oid_to_argv, &cp->args);
+
+ *task_cb = task;
+ strbuf_release(&submodule_prefix);
+ return 1;
+ }
+
return 0;
}
@@ -1332,20 +1416,64 @@ static int fetch_start_failure(struct strbuf *err,
void *cb, void *task_cb)
{
struct submodule_parallel_fetch *spf = cb;
+ struct get_next_submodule_task *task = task_cb;
spf->result = 1;
+ get_next_submodule_task_release(task);
return 0;
}
+static int commit_exists_in_sub(const struct object_id *oid, void *data)
+{
+ struct repository *subrepo = data;
+
+ enum object_type type = oid_object_info(subrepo, oid, NULL);
+
+ return type != OBJ_COMMIT;
+}
+
static int fetch_finish(int retvalue, struct strbuf *err,
void *cb, void *task_cb)
{
struct submodule_parallel_fetch *spf = cb;
+ struct get_next_submodule_task *task = task_cb;
+ const struct submodule *sub;
+
+ struct string_list_item *it;
+ struct oid_array *commits;
if (retvalue)
spf->result = 1;
+ if (!task)
+ return 0;
+
+ sub = task->sub;
+ if (!sub)
+ goto out;
+
+ it = string_list_lookup(&spf->changed_submodule_names, sub->name);
+ if (!it)
+ goto out;
+
+ commits = it->util;
+ oid_array_filter(commits,
+ commit_exists_in_sub,
+ task->repo);
+
+ trace_printf("checking for submodule: needs %d more commits", commits->nr);
+ if (commits->nr) {
+ task->commits = commits;
+ ALLOC_GROW(spf->retry, spf->retry_nr + 1, spf->retry_alloc);
+ spf->retry[spf->retry_nr] = task;
+ spf->retry_nr++;
+ return 0;
+ }
+
+out:
+ get_next_submodule_task_release(task);
+
return 0;
}
diff --git a/t/t5526-fetch-submodules.sh b/t/t5526-fetch-submodules.sh
index 42692219a1a..af12c50e7dd 100755
--- a/t/t5526-fetch-submodules.sh
+++ b/t/t5526-fetch-submodules.sh
@@ -605,4 +605,20 @@ test_expect_success "fetch new commits when submodule got renamed" '
test_cmp expect actual
'
+test_expect_success "fetch new commits on-demand when they are not reachable" '
+ git checkout --detach &&
+ C=$(git -C submodule commit-tree -m "new change outside refs/heads" HEAD^{tree}) &&
+ git -C submodule update-ref refs/changes/1 $C &&
+ git update-index --cacheinfo 160000 $C submodule &&
+ git commit -m "updated submodule outside of refs/heads" &&
+ D=$(git rev-parse HEAD) &&
+ git update-ref refs/changes/2 $D &&
+ (
+ cd downstream &&
+ git fetch --recurse-submodules --recurse-submodules-default on-demand origin refs/changes/2:refs/heads/my_branch &&
+ git -C submodule cat-file -t $C &&
+ git checkout --recurse-submodules FETCH_HEAD
+ )
+'
+
test_done
--
2.19.0.444.g18242da7ef-goog
^ permalink raw reply related [relevance 23%]
* [PATCH 6/8] submodule: fetch in submodules git directory instead of in worktree
2018-09-21 22:35 9% [PATCHv3 0/8] fetch: make sure submodule oids are fetched Stefan Beller
` (4 preceding siblings ...)
2018-09-21 22:35 17% ` [PATCH 5/8] submodule.c: do not copy around submodule list Stefan Beller
@ 2018-09-21 22:35 26% ` Stefan Beller
2018-09-21 22:35 23% ` [PATCH 7/8] fetch: retry fetching submodules if needed objects were not fetched Stefan Beller
2018-09-21 22:35 21% ` [PATCH 8/8] builtin/fetch: check for submodule updates for non branch fetches Stefan Beller
7 siblings, 0 replies; 200+ results
From: Stefan Beller @ 2018-09-21 22:35 UTC (permalink / raw)
To: git; +Cc: Stefan Beller
This patch started as a refactoring to make 'get_next_submodule' more
readable, but upon doing so, I realized that "git fetch" of the submodule
actually doesn't need to be run in the submodules worktree. So let's run
it in its git dir instead.
That should pave the way towards fetching submodules that are currently
not checked out.
This patch leaks the cp->dir in get_next_submodule, as any further
callback in run_processes_parallel doesn't have access to the child
process any more. In an early iteration of this patch, the function
get_submodule_repo_for directly returned the string containing the
git directory, which would be a better design choice for this patch.
However the next patch both fixes the memory leak of cp->dir and also has
a use case for using the full repository handle of the submodule, so
it makes sense to introduce the get_submodule_repo_for here already.
Signed-off-by: Stefan Beller <sbeller@google.com>
---
submodule.c | 47 +++++++++++++++++++++++++++----------
t/t5526-fetch-submodules.sh | 7 +++++-
2 files changed, 40 insertions(+), 14 deletions(-)
diff --git a/submodule.c b/submodule.c
index 209680ec076..8829e8d4eff 100644
--- a/submodule.c
+++ b/submodule.c
@@ -482,6 +482,12 @@ void prepare_submodule_repo_env(struct argv_array *out)
DEFAULT_GIT_DIR_ENVIRONMENT);
}
+static void prepare_submodule_repo_env_in_gitdir(struct argv_array *out)
+{
+ prepare_submodule_repo_env_no_git_dir(out);
+ argv_array_pushf(out, "%s=.", GIT_DIR_ENVIRONMENT);
+}
+
/* Helper function to display the submodule header line prior to the full
* summary output. If it can locate the submodule objects directory it will
* attempt to lookup both the left and right commits and put them into the
@@ -1228,6 +1234,25 @@ static int get_fetch_recurse_config(const struct submodule *submodule,
return spf->default_option;
}
+static struct repository *get_submodule_repo_for(struct repository *r,
+ const char *path)
+{
+ struct repository *ret = xmalloc(sizeof(*ret));
+
+ if (repo_submodule_init(ret, r, path)) {
+ /* no entry in .gitmodules? */
+ struct strbuf gitdir = STRBUF_INIT;
+ strbuf_repo_worktree_path(&gitdir, r, "%s/.git", path);
+ if (repo_init(ret, gitdir.buf, NULL)) {
+ strbuf_release(&gitdir);
+ return NULL;
+ }
+ strbuf_release(&gitdir);
+ }
+
+ return ret;
+}
+
static int get_next_submodule(struct child_process *cp,
struct strbuf *err, void *data, void **task_cb)
{
@@ -1235,12 +1260,11 @@ static int get_next_submodule(struct child_process *cp,
struct submodule_parallel_fetch *spf = data;
for (; spf->count < spf->r->index->cache_nr; spf->count++) {
- struct strbuf submodule_path = STRBUF_INIT;
- struct strbuf submodule_git_dir = STRBUF_INIT;
struct strbuf submodule_prefix = STRBUF_INIT;
const struct cache_entry *ce = spf->r->index->cache[spf->count];
- const char *git_dir, *default_argv;
+ const char *default_argv;
const struct submodule *submodule;
+ struct repository *repo;
struct submodule default_submodule = SUBMODULE_INIT;
if (!S_ISGITLINK(ce->ce_mode))
@@ -1275,16 +1299,12 @@ static int get_next_submodule(struct child_process *cp,
continue;
}
- strbuf_repo_worktree_path(&submodule_path, spf->r, "%s", ce->name);
- strbuf_addf(&submodule_git_dir, "%s/.git", submodule_path.buf);
strbuf_addf(&submodule_prefix, "%s%s/", spf->prefix, ce->name);
- git_dir = read_gitfile(submodule_git_dir.buf);
- if (!git_dir)
- git_dir = submodule_git_dir.buf;
- if (is_directory(git_dir)) {
+ repo = get_submodule_repo_for(spf->r, ce->name);
+ if (repo) {
child_process_init(cp);
- cp->dir = strbuf_detach(&submodule_path, NULL);
- prepare_submodule_repo_env(&cp->env_array);
+ prepare_submodule_repo_env_in_gitdir(&cp->env_array);
+ cp->dir = xstrdup(repo->gitdir);
cp->git_cmd = 1;
if (!spf->quiet)
strbuf_addf(err, "Fetching submodule %s%s\n",
@@ -1294,10 +1314,11 @@ static int get_next_submodule(struct child_process *cp,
argv_array_push(&cp->args, default_argv);
argv_array_push(&cp->args, "--submodule-prefix");
argv_array_push(&cp->args, submodule_prefix.buf);
+
+ repo_clear(repo);
+ free(repo);
ret = 1;
}
- strbuf_release(&submodule_path);
- strbuf_release(&submodule_git_dir);
strbuf_release(&submodule_prefix);
if (ret) {
spf->count++;
diff --git a/t/t5526-fetch-submodules.sh b/t/t5526-fetch-submodules.sh
index 6c2f9b2ba26..42692219a1a 100755
--- a/t/t5526-fetch-submodules.sh
+++ b/t/t5526-fetch-submodules.sh
@@ -566,7 +566,12 @@ test_expect_success 'fetching submodule into a broken repository' '
test_must_fail git -C dst status &&
test_must_fail git -C dst diff &&
- test_must_fail git -C dst fetch --recurse-submodules
+
+ # git-fetch cannot find the git directory of the submodule,
+ # so it will do nothing, successfully, as it cannot distinguish between
+ # this broken submodule and a submodule that was just set active but
+ # not cloned yet
+ git -C dst fetch --recurse-submodules
'
test_expect_success "fetch new commits when submodule got renamed" '
--
2.19.0.444.g18242da7ef-goog
^ permalink raw reply related [relevance 26%]
* [PATCH 5/8] submodule.c: do not copy around submodule list
2018-09-21 22:35 9% [PATCHv3 0/8] fetch: make sure submodule oids are fetched Stefan Beller
` (3 preceding siblings ...)
2018-09-21 22:35 16% ` [PATCH 4/8] submodule: move global changed_submodule_names into fetch submodule struct Stefan Beller
@ 2018-09-21 22:35 17% ` Stefan Beller
2018-09-21 22:35 26% ` [PATCH 6/8] submodule: fetch in submodules git directory instead of in worktree Stefan Beller
` (2 subsequent siblings)
7 siblings, 0 replies; 200+ results
From: Stefan Beller @ 2018-09-21 22:35 UTC (permalink / raw)
To: git; +Cc: Stefan Beller
'calculate_changed_submodule_paths' uses a local list to compute the
changed submodules, and then produces the result by copying appropriate
items into the result list.
Instead use the result list directly and prune items afterwards
using string_list_remove_empty_items.
By doing so we'll have access to the util pointer for longer that
contains the commits that we need to fetch, which will be
useful in a later patch.
Signed-off-by: Stefan Beller <sbeller@google.com>
---
submodule.c | 19 ++++++++++---------
1 file changed, 10 insertions(+), 9 deletions(-)
diff --git a/submodule.c b/submodule.c
index 7b4d136849b..209680ec076 100644
--- a/submodule.c
+++ b/submodule.c
@@ -1129,8 +1129,7 @@ static void calculate_changed_submodule_paths(
struct submodule_parallel_fetch *spf)
{
struct argv_array argv = ARGV_ARRAY_INIT;
- struct string_list changed_submodules = STRING_LIST_INIT_DUP;
- const struct string_list_item *name;
+ struct string_list_item *name;
/* No need to check if there are no submodules configured */
if (!submodule_from_path(the_repository, NULL, NULL))
@@ -1147,9 +1146,9 @@ static void calculate_changed_submodule_paths(
* Collect all submodules (whether checked out or not) for which new
* commits have been recorded upstream in "changed_submodule_names".
*/
- collect_changed_submodules(&changed_submodules, &argv);
+ collect_changed_submodules(&spf->changed_submodule_names, &argv);
- for_each_string_list_item(name, &changed_submodules) {
+ for_each_string_list_item(name, &spf->changed_submodule_names) {
struct oid_array *commits = name->util;
const struct submodule *submodule;
const char *path = NULL;
@@ -1163,12 +1162,14 @@ static void calculate_changed_submodule_paths(
if (!path)
continue;
- if (!submodule_has_commits(path, commits))
- string_list_append(&spf->changed_submodule_names,
- name->string);
+ if (submodule_has_commits(path, commits)) {
+ oid_array_clear(commits);
+ *name->string = '\0';
+ }
}
- free_submodules_oids(&changed_submodules);
+ string_list_remove_empty_items(&spf->changed_submodule_names, 1);
+
argv_array_clear(&argv);
oid_array_clear(&ref_tips_before_fetch);
oid_array_clear(&ref_tips_after_fetch);
@@ -1364,7 +1365,7 @@ int fetch_populated_submodules(struct repository *r,
argv_array_clear(&spf.args);
out:
- string_list_clear(&spf.changed_submodule_names, 1);
+ free_submodules_oids(&spf.changed_submodule_names);
return spf.result;
}
--
2.19.0.444.g18242da7ef-goog
^ permalink raw reply related [relevance 17%]
* [PATCH 4/8] submodule: move global changed_submodule_names into fetch submodule struct
2018-09-21 22:35 9% [PATCHv3 0/8] fetch: make sure submodule oids are fetched Stefan Beller
` (2 preceding siblings ...)
2018-09-21 22:35 17% ` [PATCH 3/8] submodule.c: sort changed_submodule_names before searching it Stefan Beller
@ 2018-09-21 22:35 16% ` Stefan Beller
2018-09-21 22:35 17% ` [PATCH 5/8] submodule.c: do not copy around submodule list Stefan Beller
` (3 subsequent siblings)
7 siblings, 0 replies; 200+ results
From: Stefan Beller @ 2018-09-21 22:35 UTC (permalink / raw)
To: git; +Cc: Stefan Beller
The `changed_submodule_names` are only used for fetching, so let's make it
part of the struct that is passed around for fetching submodules.
Signed-off-by: Stefan Beller <sbeller@google.com>
---
submodule.c | 42 +++++++++++++++++++++++-------------------
1 file changed, 23 insertions(+), 19 deletions(-)
diff --git a/submodule.c b/submodule.c
index 2f5c19d0aac..7b4d136849b 100644
--- a/submodule.c
+++ b/submodule.c
@@ -25,7 +25,7 @@
#include "commit-reach.h"
static int config_update_recurse_submodules = RECURSE_SUBMODULES_OFF;
-static struct string_list changed_submodule_names = STRING_LIST_INIT_DUP;
+
static int initialized_fetch_ref_tips;
static struct oid_array ref_tips_before_fetch;
static struct oid_array ref_tips_after_fetch;
@@ -1111,7 +1111,22 @@ void check_for_new_submodule_commits(struct object_id *oid)
oid_array_append(&ref_tips_after_fetch, oid);
}
-static void calculate_changed_submodule_paths(void)
+struct submodule_parallel_fetch {
+ int count;
+ struct argv_array args;
+ struct repository *r;
+ const char *prefix;
+ int command_line_option;
+ int default_option;
+ int quiet;
+ int result;
+
+ struct string_list changed_submodule_names;
+};
+#define SPF_INIT {0, ARGV_ARRAY_INIT, NULL, NULL, 0, 0, 0, 0, STRING_LIST_INIT_DUP }
+
+static void calculate_changed_submodule_paths(
+ struct submodule_parallel_fetch *spf)
{
struct argv_array argv = ARGV_ARRAY_INIT;
struct string_list changed_submodules = STRING_LIST_INIT_DUP;
@@ -1149,7 +1164,8 @@ static void calculate_changed_submodule_paths(void)
continue;
if (!submodule_has_commits(path, commits))
- string_list_append(&changed_submodule_names, name->string);
+ string_list_append(&spf->changed_submodule_names,
+ name->string);
}
free_submodules_oids(&changed_submodules);
@@ -1186,18 +1202,6 @@ int submodule_touches_in_range(struct object_id *excl_oid,
return ret;
}
-struct submodule_parallel_fetch {
- int count;
- struct argv_array args;
- struct repository *r;
- const char *prefix;
- int command_line_option;
- int default_option;
- int quiet;
- int result;
-};
-#define SPF_INIT {0, ARGV_ARRAY_INIT, NULL, NULL, 0, 0, 0, 0}
-
static int get_fetch_recurse_config(const struct submodule *submodule,
struct submodule_parallel_fetch *spf)
{
@@ -1258,7 +1262,7 @@ static int get_next_submodule(struct child_process *cp,
case RECURSE_SUBMODULES_ON_DEMAND:
if (!submodule ||
!string_list_lookup(
- &changed_submodule_names,
+ &spf->changed_submodule_names,
submodule->name))
continue;
default_argv = "on-demand";
@@ -1350,8 +1354,8 @@ int fetch_populated_submodules(struct repository *r,
argv_array_push(&spf.args, "--recurse-submodules-default");
/* default value, "--submodule-prefix" and its value are added later */
- calculate_changed_submodule_paths();
- string_list_sort(&changed_submodule_names);
+ calculate_changed_submodule_paths(&spf);
+ string_list_sort(&spf.changed_submodule_names);
run_processes_parallel(max_parallel_jobs,
get_next_submodule,
fetch_start_failure,
@@ -1360,7 +1364,7 @@ int fetch_populated_submodules(struct repository *r,
argv_array_clear(&spf.args);
out:
- string_list_clear(&changed_submodule_names, 1);
+ string_list_clear(&spf.changed_submodule_names, 1);
return spf.result;
}
--
2.19.0.444.g18242da7ef-goog
^ permalink raw reply related [relevance 16%]
* [PATCH 3/8] submodule.c: sort changed_submodule_names before searching it
2018-09-21 22:35 9% [PATCHv3 0/8] fetch: make sure submodule oids are fetched Stefan Beller
2018-09-21 22:35 24% ` [PATCH 2/8] submodule.c: fix indentation Stefan Beller
@ 2018-09-21 22:35 17% ` Stefan Beller
2018-09-21 22:35 16% ` [PATCH 4/8] submodule: move global changed_submodule_names into fetch submodule struct Stefan Beller
` (4 subsequent siblings)
7 siblings, 0 replies; 200+ results
From: Stefan Beller @ 2018-09-21 22:35 UTC (permalink / raw)
To: git; +Cc: Stefan Beller
We can string_list_insert() to maintain sorted-ness of the
list as we find new items, or we can string_list_append() to
build an unsorted list and sort it at the end just once.
To pick which one is more appropriate, we notice the fact
that we discover new items more or less in the already
sorted order. That makes "append then sort" more
appropriate.
Signed-off-by: Stefan Beller <sbeller@google.com>
---
submodule.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/submodule.c b/submodule.c
index 67469a8f513..2f5c19d0aac 100644
--- a/submodule.c
+++ b/submodule.c
@@ -1257,7 +1257,7 @@ static int get_next_submodule(struct child_process *cp,
case RECURSE_SUBMODULES_DEFAULT:
case RECURSE_SUBMODULES_ON_DEMAND:
if (!submodule ||
- !unsorted_string_list_lookup(
+ !string_list_lookup(
&changed_submodule_names,
submodule->name))
continue;
@@ -1351,6 +1351,7 @@ int fetch_populated_submodules(struct repository *r,
/* default value, "--submodule-prefix" and its value are added later */
calculate_changed_submodule_paths();
+ string_list_sort(&changed_submodule_names);
run_processes_parallel(max_parallel_jobs,
get_next_submodule,
fetch_start_failure,
--
2.19.0.444.g18242da7ef-goog
^ permalink raw reply related [relevance 17%]
* [PATCH 2/8] submodule.c: fix indentation
2018-09-21 22:35 9% [PATCHv3 0/8] fetch: make sure submodule oids are fetched Stefan Beller
@ 2018-09-21 22:35 24% ` Stefan Beller
2018-09-21 22:35 17% ` [PATCH 3/8] submodule.c: sort changed_submodule_names before searching it Stefan Beller
` (5 subsequent siblings)
7 siblings, 0 replies; 200+ results
From: Stefan Beller @ 2018-09-21 22:35 UTC (permalink / raw)
To: git; +Cc: Stefan Beller
The submodule subsystem is really bad at staying within 80 characters.
Fix it while we are here.
Signed-off-by: Stefan Beller <sbeller@google.com>
---
submodule.c | 9 ++++++---
1 file changed, 6 insertions(+), 3 deletions(-)
diff --git a/submodule.c b/submodule.c
index ed05339b588..67469a8f513 100644
--- a/submodule.c
+++ b/submodule.c
@@ -1245,7 +1245,8 @@ static int get_next_submodule(struct child_process *cp,
if (!submodule) {
const char *name = default_name_or_path(ce->name);
if (name) {
- default_submodule.path = default_submodule.name = name;
+ default_submodule.path = name;
+ default_submodule.name = name;
submodule = &default_submodule;
}
}
@@ -1255,8 +1256,10 @@ static int get_next_submodule(struct child_process *cp,
default:
case RECURSE_SUBMODULES_DEFAULT:
case RECURSE_SUBMODULES_ON_DEMAND:
- if (!submodule || !unsorted_string_list_lookup(&changed_submodule_names,
- submodule->name))
+ if (!submodule ||
+ !unsorted_string_list_lookup(
+ &changed_submodule_names,
+ submodule->name))
continue;
default_argv = "on-demand";
break;
--
2.19.0.444.g18242da7ef-goog
^ permalink raw reply related [relevance 24%]
* [PATCHv3 0/8] fetch: make sure submodule oids are fetched
@ 2018-09-21 22:35 9% Stefan Beller
` (7 more replies)
0 siblings, 8 replies; 200+ results
From: Stefan Beller @ 2018-09-21 22:35 UTC (permalink / raw)
To: git; +Cc: Stefan Beller
v3:
* I discovered some issues with v2 after sending,
which is why I rewrote the later patches completely
and now we pass around a "task" struct that contains everything to know
about the things to work on and what needs free()ing afterwards.
* as it is no longer string list based, this drops adding string_list_{pop, last}
v2:
* extended commit messages,
* plugged a memory leak
* rewrote the patch "sha1-array: provide oid_array_filter" to be much more like
object_array_fiter
* fixed a typo pointed out by Ramsay.
The range diff is below.
Thanks,
Stefan
v1:
Currently when git-fetch is asked to recurse into submodules, it dispatches
a plain "git-fetch -C <submodule-dir>" (and some submodule related options
such as prefix and recusing strategy, but) without any information of the
remote or the tip that should be fetched.
This works surprisingly well in some workflows, not so well in others,
which this series aims to fix.
The first patches provide new basic functionality and do some refactoring;
the interesting part is in the two last patches.
This was discussed in
https://public-inbox.org/git/20180808221752.195419-1-sbeller@google.com/
and I think I addressed all feedback so far.
Stefan Beller (8):
sha1-array: provide oid_array_filter
submodule.c: fix indentation
submodule.c: sort changed_submodule_names before searching it
submodule: move global changed_submodule_names into fetch submodule
struct
submodule.c: do not copy around submodule list
submodule: fetch in submodules git directory instead of in worktree
fetch: retry fetching submodules if needed objects were not fetched
builtin/fetch: check for submodule updates for non branch fetches
builtin/fetch.c | 14 +-
sha1-array.c | 17 +++
sha1-array.h | 9 ++
submodule.c | 268 ++++++++++++++++++++++++++++--------
t/t5526-fetch-submodules.sh | 23 +++-
5 files changed, 268 insertions(+), 63 deletions(-)
--
2.19.0.444.g18242da7ef-goog
^ permalink raw reply [relevance 9%]
* Re: [PATCH] fetch: Ensure that fetch.recurseSubmodules overrides submodule.recurse.
@ 2018-09-21 19:22 7% ` Stefan Beller
0 siblings, 0 replies; 200+ results
From: Stefan Beller @ 2018-09-21 19:22 UTC (permalink / raw)
To: Marc Branchaud; +Cc: git
On Fri, Sep 21, 2018 at 12:00 PM Marc Branchaud <marcnarc@xiplink.com> wrote:
>
> Also document this fact.
>
> Signed-off-by: Marc Branchaud <marcnarc@xiplink.com>
> ---
>
> I ran into this bug when I had both fetch.recurseSubmodules=on-demand and
> submodule.recurse=true, and submodule.recurse was set *after*
> fetch.recurseSubmodules in my config.
>
> The fix ensures that fetch.recurseSubmodules always overrides
> submodule.recurse. If neither is set then fetch still behaves as if
> fetch.recurseSubmodules=on-demand (the documented default).
At least the second paragraph is valuable information in the commit
message, so maybe add it there? I am not sure if the first paragraph is
a good part for the commit message, but maybe helps for writing a test?
> + reference. This option overrides the more general submodule.recurse
> + option, for the `fetch` command.
>
> fetch.fsckObjects::
> If it is set to true, git-fetch-pack will check all fetched
> @@ -3465,7 +3466,8 @@ submodule.active::
> submodule.recurse::
> Specifies if commands recurse into submodules by default. This
> applies to all commands that have a `--recurse-submodules` option,
> - except `clone`.
> + except `clone`. Also, the `fetch` command's behaviour can be specified
> + independently with the fetch.recurseSubmodules option.
There is also push.recurseSubmodules, which should behave similarly?
The series that introduced submodule.recurse ends with 58f4203e7db
(builtin/fetch.c: respect 'submodule.recurse' option, 2017-05-31)
(sb/submodule-blanket-recursive)
seems to have overlooked this only for fetch/push, as the other
commands (checkout, read-tree, reset, grep) do not have their
own specific setting to recurse.
> @@ -88,6 +90,7 @@ static int git_fetch_config(const char *k, const char *v, void *cb)
> max_children = parse_submodule_fetchjobs(k, v);
> return 0;
> } else if (!strcmp(k, "fetch.recursesubmodules")) {
> + recurse_submodules_set_explicitly = 1;
the command line option also overried explicitely, but that
is ensured via the program flow (parse_config happens after
git_config to overlay options, which itself was pre-seeded
with fetch_config_from_gitmodules).
I briefly wondered if this overlaying approach would be better
(i.e. first do git_config with more generic option, and then
again with the more detailed option) as it would save one
global variable, but the downsides are terrible (way more
work to do, more code and such), so I think having a global
makes sense and gets the job done.
Ideally instead of a global we'd have this flag stored in
the repository struct, as eventually in the long run,
fetch_populated_submodules could happen in-process
instead of spawning fetch processes for each submodule
(and their nested submodules which may be configured
differently). But for now the global will do.
Thanks!
Stefan
^ permalink raw reply [relevance 7%]
* [PATCH 9/9] builtin/fetch: check for submodule updates for non branch fetches
2018-09-17 21:35 9% ` [PATCHv2 0/9] fetch: make sure submodule oids are fetched Stefan Beller
` (5 preceding siblings ...)
2018-09-17 21:35 18% ` [PATCH 8/9] fetch: retry fetching submodules if needed objects were not fetched Stefan Beller
@ 2018-09-17 21:35 21% ` Stefan Beller
6 siblings, 0 replies; 200+ results
From: Stefan Beller @ 2018-09-17 21:35 UTC (permalink / raw)
To: sbeller; +Cc: git
Gerrit, the code review tool, has a different workflow than our mailing
list based approach. Usually users upload changes to a Gerrit server and
continuous integration and testing happens by bots. Sometimes however a
user wants to checkout a change locally and look at it locally. For this
use case, Gerrit offers a command line snippet to copy and paste to your
terminal, which looks like
git fetch https://<host>/gerrit refs/changes/<id> &&
git checkout FETCH_HEAD
For Gerrit changes that contain changing submodule gitlinks, it would be
easy to extend both the fetch and checkout with the '--recurse-submodules'
flag, such that this command line snippet would produce the state of a
change locally.
However the functionality added in the previous patch, which would
ensure that we fetch the objects in the submodule that the gitlink pointed
at, only works for remote tracking branches so far, not for FETCH_HEAD.
Make sure that fetching a superproject to its FETCH_HEAD, also respects
the existence checks for objects in the submodule recursion.
Signed-off-by: Stefan Beller <sbeller@google.com>
---
builtin/fetch.c | 5 ++++-
t/t5526-fetch-submodules.sh | 2 +-
2 files changed, 5 insertions(+), 2 deletions(-)
diff --git a/builtin/fetch.c b/builtin/fetch.c
index 95c44bf6ffa..ea6ecd123e7 100644
--- a/builtin/fetch.c
+++ b/builtin/fetch.c
@@ -887,11 +887,14 @@ static int store_updated_refs(const char *raw_url, const char *remote_name,
rc |= update_local_ref(ref, what, rm, ¬e,
summary_width);
free(ref);
- } else
+ } else {
+ check_for_new_submodule_commits(&rm->old_oid);
format_display(¬e, '*',
*kind ? kind : "branch", NULL,
*what ? what : "HEAD",
"FETCH_HEAD", summary_width);
+ }
+
if (note.len) {
if (verbosity >= 0 && !shown_url) {
fprintf(stderr, _("From %.*s\n"),
diff --git a/t/t5526-fetch-submodules.sh b/t/t5526-fetch-submodules.sh
index af12c50e7dd..a509eabb044 100755
--- a/t/t5526-fetch-submodules.sh
+++ b/t/t5526-fetch-submodules.sh
@@ -615,7 +615,7 @@ test_expect_success "fetch new commits on-demand when they are not reachable" '
git update-ref refs/changes/2 $D &&
(
cd downstream &&
- git fetch --recurse-submodules --recurse-submodules-default on-demand origin refs/changes/2:refs/heads/my_branch &&
+ git fetch --recurse-submodules origin refs/changes/2 &&
git -C submodule cat-file -t $C &&
git checkout --recurse-submodules FETCH_HEAD
)
--
2.19.0.397.gdd90340f6a-goog
^ permalink raw reply related [relevance 21%]
* [PATCH 8/9] fetch: retry fetching submodules if needed objects were not fetched
2018-09-17 21:35 9% ` [PATCHv2 0/9] fetch: make sure submodule oids are fetched Stefan Beller
` (4 preceding siblings ...)
2018-09-17 21:35 26% ` [PATCH 7/9] submodule: fetch in submodules git directory instead of in worktree Stefan Beller
@ 2018-09-17 21:35 18% ` Stefan Beller
2018-09-17 21:35 21% ` [PATCH 9/9] builtin/fetch: check for submodule updates for non branch fetches Stefan Beller
6 siblings, 0 replies; 200+ results
From: Stefan Beller @ 2018-09-17 21:35 UTC (permalink / raw)
To: sbeller; +Cc: git
Currently when git-fetch is asked to recurse into submodules, it dispatches
a plain "git-fetch -C <submodule-dir>" (with some submodule related options
such as prefix and recusing strategy, but) without any information of the
remote or the tip that should be fetched.
This works surprisingly well in some workflows (such as using submodules
as a third party library), while not so well in other scenarios, such
as in a Gerrit topic-based workflow, that can tie together changes
(potentially across repositories) on the server side. One of the parts
of such a Gerrit workflow is to download a change when wanting to examine
it, and you'd want to have its submodule changes that are in the same
topic downloaded as well. However these submodule changes reside in their
own repository in their own ref (refs/changes/<int>).
Retry fetching a submodule by object name if the object id that the
superproject points to, cannot be found.
This retrying does not happen when the "git fetch" done at the
superproject is not storing the fetched results in remote
tracking branches (i.e. instead just recording them to
FETCH_HEAD) in this step. A later patch will fix this.
builtin/fetch used to only inspect submodules when they were fetched
"on-demand", as in either on/off case it was clear whether the submodule
needs to be fetched. However to know whether we need to try fetching the
object ids, we need to identify the object names, which is done in this
function check_for_new_submodule_commits(), so we'll also run that code
in case the submodule recursion is set to "on".
Signed-off-by: Stefan Beller <sbeller@google.com>
---
builtin/fetch.c | 9 ++--
submodule.c | 87 ++++++++++++++++++++++++++++++++++++-
t/t5526-fetch-submodules.sh | 16 +++++++
3 files changed, 104 insertions(+), 8 deletions(-)
diff --git a/builtin/fetch.c b/builtin/fetch.c
index 61bec5d213d..95c44bf6ffa 100644
--- a/builtin/fetch.c
+++ b/builtin/fetch.c
@@ -700,8 +700,7 @@ static int update_local_ref(struct ref *ref,
what = _("[new ref]");
}
- if ((recurse_submodules != RECURSE_SUBMODULES_OFF) &&
- (recurse_submodules != RECURSE_SUBMODULES_ON))
+ if (recurse_submodules != RECURSE_SUBMODULES_OFF)
check_for_new_submodule_commits(&ref->new_oid);
r = s_update_ref(msg, ref, 0);
format_display(display, r ? '!' : '*', what,
@@ -716,8 +715,7 @@ static int update_local_ref(struct ref *ref,
strbuf_add_unique_abbrev(&quickref, ¤t->object.oid, DEFAULT_ABBREV);
strbuf_addstr(&quickref, "..");
strbuf_add_unique_abbrev(&quickref, &ref->new_oid, DEFAULT_ABBREV);
- if ((recurse_submodules != RECURSE_SUBMODULES_OFF) &&
- (recurse_submodules != RECURSE_SUBMODULES_ON))
+ if (recurse_submodules != RECURSE_SUBMODULES_OFF)
check_for_new_submodule_commits(&ref->new_oid);
r = s_update_ref("fast-forward", ref, 1);
format_display(display, r ? '!' : ' ', quickref.buf,
@@ -731,8 +729,7 @@ static int update_local_ref(struct ref *ref,
strbuf_add_unique_abbrev(&quickref, ¤t->object.oid, DEFAULT_ABBREV);
strbuf_addstr(&quickref, "...");
strbuf_add_unique_abbrev(&quickref, &ref->new_oid, DEFAULT_ABBREV);
- if ((recurse_submodules != RECURSE_SUBMODULES_OFF) &&
- (recurse_submodules != RECURSE_SUBMODULES_ON))
+ if (recurse_submodules != RECURSE_SUBMODULES_OFF)
check_for_new_submodule_commits(&ref->new_oid);
r = s_update_ref("forced-update", ref, 1);
format_display(display, r ? '!' : '+', quickref.buf,
diff --git a/submodule.c b/submodule.c
index 88bce534268..7d59e56171f 100644
--- a/submodule.c
+++ b/submodule.c
@@ -1127,8 +1127,11 @@ struct submodule_parallel_fetch {
int result;
struct string_list changed_submodule_names;
+ struct string_list retry;
};
-#define SPF_INIT {0, ARGV_ARRAY_INIT, NULL, NULL, 0, 0, 0, 0, STRING_LIST_INIT_DUP }
+#define SPF_INIT {0, ARGV_ARRAY_INIT, NULL, NULL, 0, 0, 0, 0, \
+ STRING_LIST_INIT_DUP, \
+ STRING_LIST_INIT_NODUP}
static void calculate_changed_submodule_paths(
struct submodule_parallel_fetch *spf)
@@ -1259,8 +1262,10 @@ static int get_next_submodule(struct child_process *cp,
{
int ret = 0;
struct submodule_parallel_fetch *spf = data;
+ struct string_list_item *it;
for (; spf->count < spf->r->index->cache_nr; spf->count++) {
+ int recurse_config;
struct strbuf submodule_prefix = STRBUF_INIT;
const struct cache_entry *ce = spf->r->index->cache[spf->count];
const char *git_dir, *default_argv;
@@ -1280,7 +1285,9 @@ static int get_next_submodule(struct child_process *cp,
}
}
- switch (get_fetch_recurse_config(submodule, spf))
+ recurse_config = get_fetch_recurse_config(submodule, spf);
+
+ switch (recurse_config)
{
default:
case RECURSE_SUBMODULES_DEFAULT:
@@ -1321,9 +1328,50 @@ static int get_next_submodule(struct child_process *cp,
strbuf_release(&submodule_prefix);
if (ret) {
spf->count++;
+ if (submodule != &default_submodule)
+ /* discard const-ness: */
+ *task_cb = (void*)submodule;
return 1;
}
}
+
+retry_next:
+
+ if (spf->retry.nr) {
+ struct strbuf submodule_prefix = STRBUF_INIT;
+ const struct submodule *sub;
+
+ it = string_list_last(&spf->retry);
+ sub = submodule_from_name(spf->r, &null_oid,
+ it->string);
+
+ child_process_init(cp);
+ cp->dir = get_submodule_git_dir(spf->r, sub->path);
+ if (!cp->dir) {
+ string_list_pop(&spf->retry, 0);
+ goto retry_next;
+ }
+ prepare_submodule_repo_env_in_gitdir(&cp->env_array);
+ cp->git_cmd = 1;
+
+ strbuf_addf(&submodule_prefix, "%s%s/", spf->prefix, sub->path);
+ argv_array_init(&cp->args);
+ argv_array_pushv(&cp->args, spf->args.argv);
+ argv_array_push(&cp->args, "on-demand");
+ argv_array_push(&cp->args, "--submodule-prefix");
+ argv_array_push(&cp->args, submodule_prefix.buf);
+
+ /* NEEDSWORK: have get_default_remote from s--h */
+ argv_array_push(&cp->args, "origin");
+ oid_array_for_each_unique(it->util,
+ append_oid_to_argv, &cp->args);
+
+ *task_cb = NULL; /* make sure we do not recurse forever */
+ strbuf_release(&submodule_prefix);
+ string_list_pop(&spf->retry, 0);
+ return 1;
+ }
+
return 0;
}
@@ -1337,14 +1385,49 @@ static int fetch_start_failure(struct strbuf *err,
return 0;
}
+static int commit_exists_in_sub(const struct object_id *oid, void *data)
+{
+ struct repository *subrepo = data;
+
+ enum object_type type = oid_object_info(subrepo, oid, NULL);
+
+ return type != OBJ_COMMIT;
+}
+
static int fetch_finish(int retvalue, struct strbuf *err,
void *cb, void *task_cb)
{
struct submodule_parallel_fetch *spf = cb;
+ struct submodule *sub = task_cb;
+ struct repository subrepo;
if (retvalue)
spf->result = 1;
+ if (!sub)
+ return 0;
+
+ if (repo_submodule_init(&subrepo, spf->r, sub->path) < 0)
+ warning(_("Could not get submodule repository for submodule '%s' in repository '%s'"),
+ sub->path, spf->r->worktree);
+ else {
+ struct string_list_item *it;
+ struct oid_array *commits;
+
+ it = string_list_lookup(&spf->changed_submodule_names, sub->name);
+ if (!it)
+ return 0;
+
+ commits = it->util;
+ oid_array_filter(commits,
+ commit_exists_in_sub,
+ &subrepo);
+
+ if (commits->nr)
+ string_list_append(&spf->retry, sub->name)
+ ->util = commits;
+ }
+
return 0;
}
diff --git a/t/t5526-fetch-submodules.sh b/t/t5526-fetch-submodules.sh
index 42692219a1a..af12c50e7dd 100755
--- a/t/t5526-fetch-submodules.sh
+++ b/t/t5526-fetch-submodules.sh
@@ -605,4 +605,20 @@ test_expect_success "fetch new commits when submodule got renamed" '
test_cmp expect actual
'
+test_expect_success "fetch new commits on-demand when they are not reachable" '
+ git checkout --detach &&
+ C=$(git -C submodule commit-tree -m "new change outside refs/heads" HEAD^{tree}) &&
+ git -C submodule update-ref refs/changes/1 $C &&
+ git update-index --cacheinfo 160000 $C submodule &&
+ git commit -m "updated submodule outside of refs/heads" &&
+ D=$(git rev-parse HEAD) &&
+ git update-ref refs/changes/2 $D &&
+ (
+ cd downstream &&
+ git fetch --recurse-submodules --recurse-submodules-default on-demand origin refs/changes/2:refs/heads/my_branch &&
+ git -C submodule cat-file -t $C &&
+ git checkout --recurse-submodules FETCH_HEAD
+ )
+'
+
test_done
--
2.19.0.397.gdd90340f6a-goog
^ permalink raw reply related [relevance 18%]
* [PATCH 7/9] submodule: fetch in submodules git directory instead of in worktree
2018-09-17 21:35 9% ` [PATCHv2 0/9] fetch: make sure submodule oids are fetched Stefan Beller
` (3 preceding siblings ...)
2018-09-17 21:35 17% ` [PATCH 6/9] submodule.c: do not copy around submodule list Stefan Beller
@ 2018-09-17 21:35 26% ` Stefan Beller
2018-09-17 21:35 18% ` [PATCH 8/9] fetch: retry fetching submodules if needed objects were not fetched Stefan Beller
2018-09-17 21:35 21% ` [PATCH 9/9] builtin/fetch: check for submodule updates for non branch fetches Stefan Beller
6 siblings, 0 replies; 200+ results
From: Stefan Beller @ 2018-09-17 21:35 UTC (permalink / raw)
To: sbeller; +Cc: git
This patch started as a refactoring to make 'get_next_submodule' more
readable, but upon doing so, I realized that "git fetch" of the submodule
actually doesn't need to be run in the submodules worktree. So let's run
it in its git dir instead.
That should pave the way towards fetching submodules that are currently
not checked out.
Signed-off-by: Stefan Beller <sbeller@google.com>
---
submodule.c | 45 +++++++++++++++++++++++++++----------
t/t5526-fetch-submodules.sh | 7 +++++-
2 files changed, 39 insertions(+), 13 deletions(-)
diff --git a/submodule.c b/submodule.c
index 00a9a3c6b12..88bce534268 100644
--- a/submodule.c
+++ b/submodule.c
@@ -481,6 +481,12 @@ void prepare_submodule_repo_env(struct argv_array *out)
DEFAULT_GIT_DIR_ENVIRONMENT);
}
+static void prepare_submodule_repo_env_in_gitdir(struct argv_array *out)
+{
+ prepare_submodule_repo_env_no_git_dir(out);
+ argv_array_pushf(out, "%s=.", GIT_DIR_ENVIRONMENT);
+}
+
/* Helper function to display the submodule header line prior to the full
* summary output. If it can locate the submodule objects directory it will
* attempt to lookup both the left and right commits and put them into the
@@ -1227,6 +1233,27 @@ static int get_fetch_recurse_config(const struct submodule *submodule,
return spf->default_option;
}
+static char *get_submodule_git_dir(struct repository *r, const char *path)
+{
+ struct repository subrepo;
+ const char *ret;
+
+ if (repo_submodule_init(&subrepo, r, path)) {
+ /* no entry in .gitmodules? */
+ struct strbuf gitdir = STRBUF_INIT;
+ strbuf_repo_worktree_path(&gitdir, r, "%s/.git", path);
+ if (repo_init(&subrepo, gitdir.buf, NULL)) {
+ strbuf_release(&gitdir);
+ return NULL;
+ }
+ }
+
+ ret = xstrdup(subrepo.gitdir);
+ repo_clear(&subrepo);
+
+ return ret;
+}
+
static int get_next_submodule(struct child_process *cp,
struct strbuf *err, void *data, void **task_cb)
{
@@ -1234,8 +1261,6 @@ static int get_next_submodule(struct child_process *cp,
struct submodule_parallel_fetch *spf = data;
for (; spf->count < spf->r->index->cache_nr; spf->count++) {
- struct strbuf submodule_path = STRBUF_INIT;
- struct strbuf submodule_git_dir = STRBUF_INIT;
struct strbuf submodule_prefix = STRBUF_INIT;
const struct cache_entry *ce = spf->r->index->cache[spf->count];
const char *git_dir, *default_argv;
@@ -1274,16 +1299,12 @@ static int get_next_submodule(struct child_process *cp,
continue;
}
- strbuf_repo_worktree_path(&submodule_path, spf->r, "%s", ce->name);
- strbuf_addf(&submodule_git_dir, "%s/.git", submodule_path.buf);
strbuf_addf(&submodule_prefix, "%s%s/", spf->prefix, ce->name);
- git_dir = read_gitfile(submodule_git_dir.buf);
- if (!git_dir)
- git_dir = submodule_git_dir.buf;
- if (is_directory(git_dir)) {
+ git_dir = get_submodule_git_dir(spf->r, ce->name);
+ if (git_dir) {
child_process_init(cp);
- cp->dir = strbuf_detach(&submodule_path, NULL);
- prepare_submodule_repo_env(&cp->env_array);
+ prepare_submodule_repo_env_in_gitdir(&cp->env_array);
+ cp->dir = git_dir;
cp->git_cmd = 1;
if (!spf->quiet)
strbuf_addf(err, "Fetching submodule %s%s\n",
@@ -1293,10 +1314,10 @@ static int get_next_submodule(struct child_process *cp,
argv_array_push(&cp->args, default_argv);
argv_array_push(&cp->args, "--submodule-prefix");
argv_array_push(&cp->args, submodule_prefix.buf);
+
+ free(git_dir);
ret = 1;
}
- strbuf_release(&submodule_path);
- strbuf_release(&submodule_git_dir);
strbuf_release(&submodule_prefix);
if (ret) {
spf->count++;
diff --git a/t/t5526-fetch-submodules.sh b/t/t5526-fetch-submodules.sh
index 6c2f9b2ba26..42692219a1a 100755
--- a/t/t5526-fetch-submodules.sh
+++ b/t/t5526-fetch-submodules.sh
@@ -566,7 +566,12 @@ test_expect_success 'fetching submodule into a broken repository' '
test_must_fail git -C dst status &&
test_must_fail git -C dst diff &&
- test_must_fail git -C dst fetch --recurse-submodules
+
+ # git-fetch cannot find the git directory of the submodule,
+ # so it will do nothing, successfully, as it cannot distinguish between
+ # this broken submodule and a submodule that was just set active but
+ # not cloned yet
+ git -C dst fetch --recurse-submodules
'
test_expect_success "fetch new commits when submodule got renamed" '
--
2.19.0.397.gdd90340f6a-goog
^ permalink raw reply related [relevance 26%]
* [PATCH 6/9] submodule.c: do not copy around submodule list
2018-09-17 21:35 9% ` [PATCHv2 0/9] fetch: make sure submodule oids are fetched Stefan Beller
` (2 preceding siblings ...)
2018-09-17 21:35 16% ` [PATCH 5/9] submodule: move global changed_submodule_names into fetch submodule struct Stefan Beller
@ 2018-09-17 21:35 17% ` Stefan Beller
2018-09-17 21:35 26% ` [PATCH 7/9] submodule: fetch in submodules git directory instead of in worktree Stefan Beller
` (2 subsequent siblings)
6 siblings, 0 replies; 200+ results
From: Stefan Beller @ 2018-09-17 21:35 UTC (permalink / raw)
To: sbeller; +Cc: git
'calculate_changed_submodule_paths' uses a local list to compute the
changed submodules, and then produces the result by copying appropriate
items into the result list.
Instead use the result list directly and prune items afterwards
using string_list_remove_empty_items.
By doing so we'll have access to the util pointer for longer that
contains the commits that we need to fetch, which will be
useful in a later patch.
Signed-off-by: Stefan Beller <sbeller@google.com>
---
submodule.c | 19 ++++++++++---------
1 file changed, 10 insertions(+), 9 deletions(-)
diff --git a/submodule.c b/submodule.c
index 3520dd76bdf..00a9a3c6b12 100644
--- a/submodule.c
+++ b/submodule.c
@@ -1128,8 +1128,7 @@ static void calculate_changed_submodule_paths(
struct submodule_parallel_fetch *spf)
{
struct argv_array argv = ARGV_ARRAY_INIT;
- struct string_list changed_submodules = STRING_LIST_INIT_DUP;
- const struct string_list_item *name;
+ struct string_list_item *name;
/* No need to check if there are no submodules configured */
if (!submodule_from_path(the_repository, NULL, NULL))
@@ -1146,9 +1145,9 @@ static void calculate_changed_submodule_paths(
* Collect all submodules (whether checked out or not) for which new
* commits have been recorded upstream in "changed_submodule_names".
*/
- collect_changed_submodules(&changed_submodules, &argv);
+ collect_changed_submodules(&spf->changed_submodule_names, &argv);
- for_each_string_list_item(name, &changed_submodules) {
+ for_each_string_list_item(name, &spf->changed_submodule_names) {
struct oid_array *commits = name->util;
const struct submodule *submodule;
const char *path = NULL;
@@ -1162,12 +1161,14 @@ static void calculate_changed_submodule_paths(
if (!path)
continue;
- if (!submodule_has_commits(path, commits))
- string_list_append(&spf->changed_submodule_names,
- name->string);
+ if (submodule_has_commits(path, commits)) {
+ oid_array_clear(commits);
+ *name->string = '\0';
+ }
}
- free_submodules_oids(&changed_submodules);
+ string_list_remove_empty_items(&spf->changed_submodule_names, 1);
+
argv_array_clear(&argv);
oid_array_clear(&ref_tips_before_fetch);
oid_array_clear(&ref_tips_after_fetch);
@@ -1363,7 +1364,7 @@ int fetch_populated_submodules(struct repository *r,
argv_array_clear(&spf.args);
out:
- string_list_clear(&spf.changed_submodule_names, 1);
+ free_submodules_oids(&spf.changed_submodule_names);
return spf.result;
}
--
2.19.0.397.gdd90340f6a-goog
^ permalink raw reply related [relevance 17%]
* [PATCH 5/9] submodule: move global changed_submodule_names into fetch submodule struct
2018-09-17 21:35 9% ` [PATCHv2 0/9] fetch: make sure submodule oids are fetched Stefan Beller
2018-09-17 21:35 24% ` [PATCH 3/9] submodule.c: fix indentation Stefan Beller
2018-09-17 21:35 17% ` [PATCH 4/9] submodule.c: sort changed_submodule_names before searching it Stefan Beller
@ 2018-09-17 21:35 16% ` Stefan Beller
2018-09-17 21:35 17% ` [PATCH 6/9] submodule.c: do not copy around submodule list Stefan Beller
` (3 subsequent siblings)
6 siblings, 0 replies; 200+ results
From: Stefan Beller @ 2018-09-17 21:35 UTC (permalink / raw)
To: sbeller; +Cc: git
The `changed_submodule_names` are only used for fetching, so let's make it
part of the struct that is passed around for fetching submodules.
Signed-off-by: Stefan Beller <sbeller@google.com>
---
submodule.c | 42 +++++++++++++++++++++++-------------------
1 file changed, 23 insertions(+), 19 deletions(-)
diff --git a/submodule.c b/submodule.c
index c6eff7699f3..3520dd76bdf 100644
--- a/submodule.c
+++ b/submodule.c
@@ -24,7 +24,7 @@
#include "object-store.h"
static int config_update_recurse_submodules = RECURSE_SUBMODULES_OFF;
-static struct string_list changed_submodule_names = STRING_LIST_INIT_DUP;
+
static int initialized_fetch_ref_tips;
static struct oid_array ref_tips_before_fetch;
static struct oid_array ref_tips_after_fetch;
@@ -1110,7 +1110,22 @@ void check_for_new_submodule_commits(struct object_id *oid)
oid_array_append(&ref_tips_after_fetch, oid);
}
-static void calculate_changed_submodule_paths(void)
+struct submodule_parallel_fetch {
+ int count;
+ struct argv_array args;
+ struct repository *r;
+ const char *prefix;
+ int command_line_option;
+ int default_option;
+ int quiet;
+ int result;
+
+ struct string_list changed_submodule_names;
+};
+#define SPF_INIT {0, ARGV_ARRAY_INIT, NULL, NULL, 0, 0, 0, 0, STRING_LIST_INIT_DUP }
+
+static void calculate_changed_submodule_paths(
+ struct submodule_parallel_fetch *spf)
{
struct argv_array argv = ARGV_ARRAY_INIT;
struct string_list changed_submodules = STRING_LIST_INIT_DUP;
@@ -1148,7 +1163,8 @@ static void calculate_changed_submodule_paths(void)
continue;
if (!submodule_has_commits(path, commits))
- string_list_append(&changed_submodule_names, name->string);
+ string_list_append(&spf->changed_submodule_names,
+ name->string);
}
free_submodules_oids(&changed_submodules);
@@ -1185,18 +1201,6 @@ int submodule_touches_in_range(struct object_id *excl_oid,
return ret;
}
-struct submodule_parallel_fetch {
- int count;
- struct argv_array args;
- struct repository *r;
- const char *prefix;
- int command_line_option;
- int default_option;
- int quiet;
- int result;
-};
-#define SPF_INIT {0, ARGV_ARRAY_INIT, NULL, NULL, 0, 0, 0, 0}
-
static int get_fetch_recurse_config(const struct submodule *submodule,
struct submodule_parallel_fetch *spf)
{
@@ -1257,7 +1261,7 @@ static int get_next_submodule(struct child_process *cp,
case RECURSE_SUBMODULES_ON_DEMAND:
if (!submodule ||
!string_list_lookup(
- &changed_submodule_names,
+ &spf->changed_submodule_names,
submodule->name))
continue;
default_argv = "on-demand";
@@ -1349,8 +1353,8 @@ int fetch_populated_submodules(struct repository *r,
argv_array_push(&spf.args, "--recurse-submodules-default");
/* default value, "--submodule-prefix" and its value are added later */
- calculate_changed_submodule_paths();
- string_list_sort(&changed_submodule_names);
+ calculate_changed_submodule_paths(&spf);
+ string_list_sort(&spf.changed_submodule_names);
run_processes_parallel(max_parallel_jobs,
get_next_submodule,
fetch_start_failure,
@@ -1359,7 +1363,7 @@ int fetch_populated_submodules(struct repository *r,
argv_array_clear(&spf.args);
out:
- string_list_clear(&changed_submodule_names, 1);
+ string_list_clear(&spf.changed_submodule_names, 1);
return spf.result;
}
--
2.19.0.397.gdd90340f6a-goog
^ permalink raw reply related [relevance 16%]
* [PATCH 4/9] submodule.c: sort changed_submodule_names before searching it
2018-09-17 21:35 9% ` [PATCHv2 0/9] fetch: make sure submodule oids are fetched Stefan Beller
2018-09-17 21:35 24% ` [PATCH 3/9] submodule.c: fix indentation Stefan Beller
@ 2018-09-17 21:35 17% ` Stefan Beller
2018-09-17 21:35 16% ` [PATCH 5/9] submodule: move global changed_submodule_names into fetch submodule struct Stefan Beller
` (4 subsequent siblings)
6 siblings, 0 replies; 200+ results
From: Stefan Beller @ 2018-09-17 21:35 UTC (permalink / raw)
To: sbeller; +Cc: git
We can string_list_insert() to maintain sorted-ness of the
list as we find new items, or we can string_list_append() to
build an unsorted list and sort it at the end just once.
To pick which one is more appropriate, we notice the fact
that we discover new items more or less in the already
sorted order. That makes "append then sort" more
appropriate.
Signed-off-by: Stefan Beller <sbeller@google.com>
---
submodule.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/submodule.c b/submodule.c
index d29dfa3d1f5..c6eff7699f3 100644
--- a/submodule.c
+++ b/submodule.c
@@ -1256,7 +1256,7 @@ static int get_next_submodule(struct child_process *cp,
case RECURSE_SUBMODULES_DEFAULT:
case RECURSE_SUBMODULES_ON_DEMAND:
if (!submodule ||
- !unsorted_string_list_lookup(
+ !string_list_lookup(
&changed_submodule_names,
submodule->name))
continue;
@@ -1350,6 +1350,7 @@ int fetch_populated_submodules(struct repository *r,
/* default value, "--submodule-prefix" and its value are added later */
calculate_changed_submodule_paths();
+ string_list_sort(&changed_submodule_names);
run_processes_parallel(max_parallel_jobs,
get_next_submodule,
fetch_start_failure,
--
2.19.0.397.gdd90340f6a-goog
^ permalink raw reply related [relevance 17%]
* [PATCH 3/9] submodule.c: fix indentation
2018-09-17 21:35 9% ` [PATCHv2 0/9] fetch: make sure submodule oids are fetched Stefan Beller
@ 2018-09-17 21:35 24% ` Stefan Beller
2018-09-17 21:35 17% ` [PATCH 4/9] submodule.c: sort changed_submodule_names before searching it Stefan Beller
` (5 subsequent siblings)
6 siblings, 0 replies; 200+ results
From: Stefan Beller @ 2018-09-17 21:35 UTC (permalink / raw)
To: sbeller; +Cc: git
The submodule subsystem is really bad at staying within 80 characters.
Fix it while we are here.
Signed-off-by: Stefan Beller <sbeller@google.com>
---
submodule.c | 9 ++++++---
1 file changed, 6 insertions(+), 3 deletions(-)
diff --git a/submodule.c b/submodule.c
index a2b266fbfae..d29dfa3d1f5 100644
--- a/submodule.c
+++ b/submodule.c
@@ -1244,7 +1244,8 @@ static int get_next_submodule(struct child_process *cp,
if (!submodule) {
const char *name = default_name_or_path(ce->name);
if (name) {
- default_submodule.path = default_submodule.name = name;
+ default_submodule.path = name;
+ default_submodule.name = name;
submodule = &default_submodule;
}
}
@@ -1254,8 +1255,10 @@ static int get_next_submodule(struct child_process *cp,
default:
case RECURSE_SUBMODULES_DEFAULT:
case RECURSE_SUBMODULES_ON_DEMAND:
- if (!submodule || !unsorted_string_list_lookup(&changed_submodule_names,
- submodule->name))
+ if (!submodule ||
+ !unsorted_string_list_lookup(
+ &changed_submodule_names,
+ submodule->name))
continue;
default_argv = "on-demand";
break;
--
2.19.0.397.gdd90340f6a-goog
^ permalink raw reply related [relevance 24%]
* [PATCHv2 0/9] fetch: make sure submodule oids are fetched
2018-09-11 23:49 10% [PATCH 0/9] fetch: make sure submodule oids are fetched Stefan Beller
` (6 preceding siblings ...)
2018-09-11 23:49 23% ` [PATCH 9/9] builtin/fetch: check for submodule updates for non branch fetches Stefan Beller
@ 2018-09-17 21:35 9% ` Stefan Beller
2018-09-17 21:35 24% ` [PATCH 3/9] submodule.c: fix indentation Stefan Beller
` (6 more replies)
7 siblings, 7 replies; 200+ results
From: Stefan Beller @ 2018-09-17 21:35 UTC (permalink / raw)
To: sbeller; +Cc: git
v2:
* extended commit messages,
* plugged a memory leak
* rewrote the patch "sha1-array: provide oid_array_filter" to be much more like
object_array_fiter
* fixed a typo pointed out by Ramsay.
The range diff is below.
Thanks,
Stefan
v1:
Currently when git-fetch is asked to recurse into submodules, it dispatches
a plain "git-fetch -C <submodule-dir>" (and some submodule related options
such as prefix and recusing strategy, but) without any information of the
remote or the tip that should be fetched.
This works surprisingly well in some workflows, not so well in others,
which this series aims to fix.
The first patches provide new basic functionality and do some refactoring;
the interesting part is in the two last patches.
This was discussed in
https://public-inbox.org/git/20180808221752.195419-1-sbeller@google.com/
and I think I addressed all feedback so far.
Stefan Beller (9):
string-list: add string_list_{pop, last} functions
sha1-array: provide oid_array_filter
submodule.c: fix indentation
submodule.c: sort changed_submodule_names before searching it
submodule: move global changed_submodule_names into fetch submodule
struct
submodule.c: do not copy around submodule list
submodule: fetch in submodules git directory instead of in worktree
fetch: retry fetching submodules if needed objects were not fetched
builtin/fetch: check for submodule updates for non branch fetches
builtin/fetch.c | 14 +--
sha1-array.c | 17 ++++
sha1-array.h | 9 ++
string-list.c | 14 +++
string-list.h | 15 +++
submodule.c | 191 ++++++++++++++++++++++++++++--------
t/t5526-fetch-submodules.sh | 23 ++++-
7 files changed, 236 insertions(+), 47 deletions(-)
git range-diff origin/sb/submodule-recursive-fetch-gets-the-tip...HEAD
1: 6fecf7cd01a ! 1: 54d31d90734 string-list: add string_list_{pop, last} functions
@@ -12,8 +12,8 @@
- string_list_pop() removes the string_list_item at the end of
the string list.
- - string_list_push() is not added, but string_list_append() can be
- used in its place.
+ - there is no string_list_push(); string_list_append() can be used
+ in its place.
You can use them in this pattern:
@@ -26,7 +26,6 @@
Helped-by: Junio C Hamano <gitster@pobox.com>
Signed-off-by: Stefan Beller <sbeller@google.com>
- Signed-off-by: Junio C Hamano <gitster@pobox.com>
diff --git a/string-list.c b/string-list.c
--- a/string-list.c
@@ -66,6 +65,10 @@
+ */
+void string_list_pop(struct string_list *list, int free_util);
+
++/*
++ * Returns the last item of the list. As it returns the raw access, do not
++ * modify the list while holding onto the returned pointer.
++ */
+static inline struct string_list_item *string_list_last(struct string_list *list)
+{
+ return &list->items[list->nr - 1];
2: 7007a318a68 < -: ----------- sha1-array: provide oid_array_filter
-: ----------- > 2: a2bd6ef2bf0 sha1-array: provide oid_array_filter
3: 807429234ac ! 3: 0300c27cbd7 submodule.c: fix indentation
@@ -6,7 +6,6 @@
Fix it while we are here.
Signed-off-by: Stefan Beller <sbeller@google.com>
- Signed-off-by: Junio C Hamano <gitster@pobox.com>
diff --git a/submodule.c b/submodule.c
--- a/submodule.c
4: f6fa5273af9 ! 4: 80cf0221bbe submodule.c: sort changed_submodule_names before searching it
@@ -12,7 +12,6 @@
appropriate.
Signed-off-by: Stefan Beller <sbeller@google.com>
- Signed-off-by: Junio C Hamano <gitster@pobox.com>
diff --git a/submodule.c b/submodule.c
--- a/submodule.c
5: adf7a2fd203 ! 5: 7ddb448b748 submodule: move global changed_submodule_names into fetch submodule struct
@@ -6,7 +6,6 @@
part of the struct that is passed around for fetching submodules.
Signed-off-by: Stefan Beller <sbeller@google.com>
- Signed-off-by: Junio C Hamano <gitster@pobox.com>
diff --git a/submodule.c b/submodule.c
--- a/submodule.c
6: 56c9398589a ! 6: 7975a7f1e3b submodule.c: do not copy around submodule list
@@ -9,12 +9,11 @@
Instead use the result list directly and prune items afterwards
using string_list_remove_empty_items.
- By doin so we'll have access to the util pointer for longer that
+ By doing so we'll have access to the util pointer for longer that
contains the commits that we need to fetch, which will be
useful in a later patch.
Signed-off-by: Stefan Beller <sbeller@google.com>
- Signed-off-by: Junio C Hamano <gitster@pobox.com>
diff --git a/submodule.c b/submodule.c
--- a/submodule.c
7: 9f70a5f32c9 ! 7: 29bc2868f26 submodule: fetch in submodules git directory instead of in worktree
@@ -3,14 +3,14 @@
submodule: fetch in submodules git directory instead of in worktree
This patch started as a refactoring to make 'get_next_submodule' more
- readable, but upon doing so, I realized that git-fetch actually doesn't
- need to be run in the worktree. So let's run it in the git dir instead.
+ readable, but upon doing so, I realized that "git fetch" of the submodule
+ actually doesn't need to be run in the submodules worktree. So let's run
+ it in its git dir instead.
That should pave the way towards fetching submodules that are currently
not checked out.
Signed-off-by: Stefan Beller <sbeller@google.com>
- Signed-off-by: Junio C Hamano <gitster@pobox.com>
diff --git a/submodule.c b/submodule.c
--- a/submodule.c
@@ -32,7 +32,7 @@
return spf->default_option;
}
-+static const char *get_submodule_git_dir(struct repository *r, const char *path)
++static char *get_submodule_git_dir(struct repository *r, const char *path)
+{
+ struct repository subrepo;
+ const char *ret;
@@ -87,7 +87,11 @@
if (!spf->quiet)
strbuf_addf(err, "Fetching submodule %s%s\n",
@@
+ argv_array_push(&cp->args, default_argv);
+ argv_array_push(&cp->args, "--submodule-prefix");
argv_array_push(&cp->args, submodule_prefix.buf);
++
++ free(git_dir);
ret = 1;
}
- strbuf_release(&submodule_path);
8: bab609b4dc1 ! 8: f837c4a0789 fetch: retry fetching submodules if sha1 were not fetched
@@ -1,9 +1,9 @@
Author: Stefan Beller <sbeller@google.com>
- fetch: retry fetching submodules if sha1 were not fetched
+ fetch: retry fetching submodules if needed objects were not fetched
Currently when git-fetch is asked to recurse into submodules, it dispatches
- a plain "git-fetch -C <submodule-dir>" (and some submodule related options
+ a plain "git-fetch -C <submodule-dir>" (with some submodule related options
such as prefix and recusing strategy, but) without any information of the
remote or the tip that should be fetched.
@@ -16,16 +16,22 @@
topic downloaded as well. However these submodule changes reside in their
own repository in their own ref (refs/changes/<int>).
- Retry fetching a submodule if the object id that the superproject points
- to, cannot be found.
+ Retry fetching a submodule by object name if the object id that the
+ superproject points to, cannot be found.
- This doesn't support fetching to FETCH_HEAD yet, but only into a local
- branch. To make fetching into FETCH_HEAD work, we need some refactoring
- in builtin/fetch.c to adjust the calls to 'check_for_new_submodule_commits'
- that is coming in the next patch.
+ This retrying does not happen when the "git fetch" done at the
+ superproject is not storing the fetched results in remote
+ tracking branches (i.e. instead just recording them to
+ FETCH_HEAD) in this step. A later patch will fix this.
+
+ builtin/fetch used to only inspect submodules when they were fetched
+ "on-demand", as in either on/off case it was clear whether the submodule
+ needs to be fetched. However to know whether we need to try fetching the
+ object ids, we need to identify the object names, which is done in this
+ function check_for_new_submodule_commits(), so we'll also run that code
+ in case the submodule recursion is set to "on".
Signed-off-by: Stefan Beller <sbeller@google.com>
- Signed-off-by: Junio C Hamano <gitster@pobox.com>
diff --git a/builtin/fetch.c b/builtin/fetch.c
--- a/builtin/fetch.c
9: c16d21313f6 ! 9: 71f2bb035b1 builtin/fetch: check for submodule updates for non branch fetches
@@ -2,11 +2,29 @@
builtin/fetch: check for submodule updates for non branch fetches
- For Gerrit users that use submodules the invocation of fetch without a
- branch is their main use case.
+ Gerrit, the code review tool, has a different workflow than our mailing
+ list based approach. Usually users upload changes to a Gerrit server and
+ continuous integration and testing happens by bots. Sometimes however a
+ user wants to checkout a change locally and look at it locally. For this
+ use case, Gerrit offers a command line snippet to copy and paste to your
+ terminal, which looks like
+
+ git fetch https://<host>/gerrit refs/changes/<id> &&
+ git checkout FETCH_HEAD
+
+ For Gerrit changes that contain changing submodule gitlinks, it would be
+ easy to extend both the fetch and checkout with the '--recurse-submodules'
+ flag, such that this command line snippet would produce the state of a
+ change locally.
+
+ However the functionality added in the previous patch, which would
+ ensure that we fetch the objects in the submodule that the gitlink pointed
+ at, only works for remote tracking branches so far, not for FETCH_HEAD.
+
+ Make sure that fetching a superproject to its FETCH_HEAD, also respects
+ the existence checks for objects in the submodule recursion.
Signed-off-by: Stefan Beller <sbeller@google.com>
- Signed-off-by: Junio C Hamano <gitster@pobox.com>
diff --git a/builtin/fetch.c b/builtin/fetch.c
--- a/builtin/fetch.c
^ permalink raw reply [relevance 9%]
* Re: [PATCH 7/9] submodule: fetch in submodules git directory instead of in worktree
2018-09-12 18:36 7% ` Junio C Hamano
@ 2018-09-13 19:29 8% ` Stefan Beller
0 siblings, 0 replies; 200+ results
From: Stefan Beller @ 2018-09-13 19:29 UTC (permalink / raw)
To: Junio C Hamano; +Cc: git
On Wed, Sep 12, 2018 at 11:36 AM Junio C Hamano <gitster@pobox.com> wrote:
>
> Stefan Beller <sbeller@google.com> writes:
>
> > This patch started as a refactoring to make 'get_next_submodule' more
> > readable, but upon doing so, I realized that git-fetch actually doesn't
> > need to be run in the worktree. So let's run it in the git dir instead.
>
> It may be clear to the author but not clear to the reader of the
> above paragraph that "worktree", "fetch" and "git dir" all refer to
> the recursively invoked operation that updates the submodules
> repository. s/git-fetch/"git fetch" for the submodule/ should be
> sufficient to help the readers.
>
> > That should pave the way towards fetching submodules that are currently
> > not checked out.
>
> Very good.
>
> > +static void prepare_submodule_repo_env_in_gitdir(struct argv_array *out)
> > +{
> > + prepare_submodule_repo_env_no_git_dir(out);
> > + argv_array_pushf(out, "%s=.", GIT_DIR_ENVIRONMENT);
> > +}
> > +
> > /* Helper function to display the submodule header line prior to the full
> > * summary output. If it can locate the submodule objects directory it will
> > * attempt to lookup both the left and right commits and put them into the
> > @@ -1227,6 +1233,27 @@ static int get_fetch_recurse_config(const struct submodule *submodule,
> > return spf->default_option;
> > }
> >
> > +static const char *get_submodule_git_dir(struct repository *r, const char *path)
> > +{
> > + struct repository subrepo;
> > + const char *ret;
> > +
> > + if (repo_submodule_init(&subrepo, r, path)) {
> > + /* no entry in .gitmodules? */
> > + struct strbuf gitdir = STRBUF_INIT;
> > + strbuf_repo_worktree_path(&gitdir, r, "%s/.git", path);
> > + if (repo_init(&subrepo, gitdir.buf, NULL)) {
> > + strbuf_release(&gitdir);
> > + return NULL;
> > + }
>
> This is for the modern "absorbed" layout? Do we get a notice and
> encouragement to migrate from the historical layout, or there is no
> need to (e.g. the migration happens automatically in some other
> codepaths)?
No, the absorbed would also be handled by repo_submodule_init.
I wrote a patch once to migrate repo_submodule_init to take a
"struct *submodule" instead of a path as the third argument, which
would fall in line with this patch as well, I'll dig it up.
Historically git-fetch supported repositories that are not submodules
(but have a gitlink and a working tree in place) as well. That is covered
here. (see comment /* no entry in .gitmodules? */)
> > - strbuf_release(&submodule_path);
> > - strbuf_release(&submodule_git_dir);
>
> But if it is a leak, it is easily plugged by freeing git_dir here, I
> think.
Thanks.
^ permalink raw reply [relevance 8%]
* Re: [PATCH 9/9] builtin/fetch: check for submodule updates for non branch fetches
2018-09-11 23:49 23% ` [PATCH 9/9] builtin/fetch: check for submodule updates for non branch fetches Stefan Beller
@ 2018-09-12 19:20 6% ` Junio C Hamano
0 siblings, 0 replies; 200+ results
From: Junio C Hamano @ 2018-09-12 19:20 UTC (permalink / raw)
To: Stefan Beller; +Cc: git
Stefan Beller <sbeller@google.com> writes:
> For Gerrit users that use submodules the invocation of fetch without a
> branch is their main use case.
That's way under explains this commit. It is totally unclear how
that statement of fact relates to the problem this patch is trying
to address; it does not even make it clear what problem is being
addressed by the patch.
>
> Signed-off-by: Stefan Beller <sbeller@google.com>
> ---
> builtin/fetch.c | 5 ++++-
> t/t5526-fetch-submodules.sh | 2 +-
> 2 files changed, 5 insertions(+), 2 deletions(-)
>
> diff --git a/builtin/fetch.c b/builtin/fetch.c
> index 95c44bf6ffa..ea6ecd123e7 100644
> --- a/builtin/fetch.c
> +++ b/builtin/fetch.c
> @@ -887,11 +887,14 @@ static int store_updated_refs(const char *raw_url, const char *remote_name,
> rc |= update_local_ref(ref, what, rm, ¬e,
> summary_width);
> free(ref);
> - } else
> + } else {
> + check_for_new_submodule_commits(&rm->old_oid);
This happens when there is no "ref", which is set only when
rm->peer_ref exists, which is set only when we are using remote
tracking branch (or more generally storing the fetched rev somewhere
in our refs/ hierarchy), e.g. the rev is recorded only in FETCH_HEAD.
What does rm->old_oid have in such a case? Is this the tip of the
superproject history we just fetched?
When we keep record of what we saw in the previous attempt to fetch,
we can tell "we have seen their history up to this old commit
before, and now we fetched their history up to this new commit" and
the question "during that time, which submodules have been modified
in the history of the superproject" becomes answerable. When we are
not keeping the record of previous fetch, how would we answer that
question without going through the whole history?
The answer is that check-for-new does not even do the "old
branch tip was X and new branch tip is Y, so we can look
only at X..Y"; it only cares about the new branch tip of the
superproject, and excludes the existing tips of all branches
in the superproject (i.e. computing something akin to "Y
--not --all" instead of "X..Y").
So, I guess this is probably reasonable. But does the call to
"check-for-new submodule" need to be unconditional? In this
codepath, do we know when we are not doing a recursive fetch in a
superproject? If so, perhaps we can omit the cost of going through
all the refs to populate ref_tips_before_fetch array in such a case.
> format_display(¬e, '*',
> *kind ? kind : "branch", NULL,
> *what ? what : "HEAD",
> "FETCH_HEAD", summary_width);
> + }
> +
> if (note.len) {
> if (verbosity >= 0 && !shown_url) {
> fprintf(stderr, _("From %.*s\n"),
> diff --git a/t/t5526-fetch-submodules.sh b/t/t5526-fetch-submodules.sh
> index af12c50e7dd..a509eabb044 100755
> --- a/t/t5526-fetch-submodules.sh
> +++ b/t/t5526-fetch-submodules.sh
> @@ -615,7 +615,7 @@ test_expect_success "fetch new commits on-demand when they are not reachable" '
> git update-ref refs/changes/2 $D &&
> (
> cd downstream &&
> - git fetch --recurse-submodules --recurse-submodules-default on-demand origin refs/changes/2:refs/heads/my_branch &&
> + git fetch --recurse-submodules origin refs/changes/2 &&
> git -C submodule cat-file -t $C &&
> git checkout --recurse-submodules FETCH_HEAD
> )
^ permalink raw reply [relevance 6%]
* Re: [PATCH 0/1] contrib: Add script to show uncovered "new" lines
@ 2018-09-12 19:14 1% ` Derrick Stolee
0 siblings, 0 replies; 200+ results
From: Derrick Stolee @ 2018-09-12 19:14 UTC (permalink / raw)
To: Derrick Stolee via GitGitGadget, git; +Cc: peff, Junio C Hamano
On 9/12/2018 12:45 PM, Derrick Stolee via GitGitGadget wrote:
> For example, I ran this against the 'jch' branch (d3c0046)
> versus 'next' (dd90340)
As another example, I ran this against the 'pu' branch (4c416a53) versus
'jch' (d3c0046) and got the following output, submitted here without
commentary:
builtin/bisect--helper.c
0b1f0fd910c (Pranit Bauva 2017-10-27 15:06:37 +0000
43) free((void *) terms->term_good);
3d3237b0e6b (Pranit Bauva 2017-10-27 15:06:37 +0000
162) if (get_oid_commit(commit, &oid))
3d3237b0e6b (Pranit Bauva 2017-10-27 15:06:37 +0000
163) return error(_("'%s' is not a valid commit"),
commit);
3d3237b0e6b (Pranit Bauva 2017-10-27 15:06:37 +0000
164) strbuf_addstr(&branch, commit);
3d3237b0e6b (Pranit Bauva 2017-10-27 15:06:37 +0000
172) error(_("Could not check out original HEAD '%s'.
Try "
3d3237b0e6b (Pranit Bauva 2017-10-27 15:06:37 +0000
174) strbuf_release(&branch);
3d3237b0e6b (Pranit Bauva 2017-10-27 15:06:37 +0000
175) argv_array_clear(&argv);
3d3237b0e6b (Pranit Bauva 2017-10-27 15:06:37 +0000
176) return -1;
0b1f0fd910c (Pranit Bauva 2017-10-27 15:06:37 +0000
215) error(_("Bad bisect_write argument: %s"), state);
0b1f0fd910c (Pranit Bauva 2017-10-27 15:06:37 +0000
216) goto fail;
0b1f0fd910c (Pranit Bauva 2017-10-27 15:06:37 +0000
220) error(_("couldn't get the oid of the rev '%s'"), rev);
0b1f0fd910c (Pranit Bauva 2017-10-27 15:06:37 +0000
221) goto fail;
0b1f0fd910c (Pranit Bauva 2017-10-27 15:06:37 +0000
226) goto fail;
0b1f0fd910c (Pranit Bauva 2017-10-27 15:06:37 +0000
230) error_errno(_("couldn't open the file '%s'"),
git_path_bisect_log());
0b1f0fd910c (Pranit Bauva 2017-10-27 15:06:37 +0000
231) goto fail;
0b1f0fd910c (Pranit Bauva 2017-10-27 15:06:37 +0000 242)fail:
0b1f0fd910c (Pranit Bauva 2017-10-27 15:06:37 +0000 243) retval
= -1;
a919f328ba3 (Pranit Bauva 2017-10-27 15:06:37 +0000
323) yesno = git_prompt(_("Are you sure [Y/n]? "), PROMPT_ECHO);
a919f328ba3 (Pranit Bauva 2017-10-27 15:06:37 +0000
324) if (starts_with(yesno, "N") || starts_with(yesno, "n"))
a919f328ba3 (Pranit Bauva 2017-10-27 15:06:37 +0000
327) goto finish;
a919f328ba3 (Pranit Bauva 2017-10-27 15:06:37 +0000
336) error(_("You need to start by \"git bisect start\". You "
a919f328ba3 (Pranit Bauva 2017-10-27 15:06:37 +0000
341) goto fail;
a919f328ba3 (Pranit Bauva 2017-10-27 15:06:37 +0000 345)fail:
35f7ca528ae (Pranit Bauva 2017-10-27 15:06:37 +0000
387) return error(_("--bisect-term requires exactly one
argument"));
35f7ca528ae (Pranit Bauva 2017-10-27 15:06:37 +0000
400) error(_("BUG: invalid argument %s for 'git
bisect terms'.\n"
5dfeec316b8 (Pranit Bauva 2017-10-27 15:06:37 +0000
416) return -1;
5dfeec316b8 (Pranit Bauva 2017-10-27 15:06:37 +0000
419) goto fail;
5dfeec316b8 (Pranit Bauva 2017-10-27 15:06:37 +0000
423) goto fail;
5dfeec316b8 (Pranit Bauva 2017-10-27 15:06:37 +0000 427)fail:
5dfeec316b8 (Pranit Bauva 2017-10-27 15:06:37 +0000 428) retval
= -1;
5dfeec316b8 (Pranit Bauva 2017-10-27 15:06:37 +0000
466) no_checkout = 1;
5dfeec316b8 (Pranit Bauva 2017-10-27 15:06:37 +0000
488) !one_of(arg, "--term-good", "--term-bad", NULL)) {
5dfeec316b8 (Pranit Bauva 2017-10-27 15:06:37 +0000
489) return error(_("unrecognised option: '%s'"), arg);
5dfeec316b8 (Pranit Bauva 2017-10-27 15:06:37 +0000
523) if (get_oid("HEAD", &head_oid))
5dfeec316b8 (Pranit Bauva 2017-10-27 15:06:37 +0000
524) return error(_("Bad HEAD - I need a HEAD"));
5dfeec316b8 (Pranit Bauva 2017-10-27 15:06:37 +0000
539) error(_("checking out '%s' failed. Try
'git "
5dfeec316b8 (Pranit Bauva 2017-10-27 15:06:37 +0000
559) return error(_("won't bisect on
cg-seek'ed tree"));
5dfeec316b8 (Pranit Bauva 2017-10-27 15:06:37 +0000
562) return error(_("Bad HEAD - strange symbolic ref"));
5dfeec316b8 (Pranit Bauva 2017-10-27 15:06:37 +0000
570) return -1;
5dfeec316b8 (Pranit Bauva 2017-10-27 15:06:37 +0000
588) goto fail;
5dfeec316b8 (Pranit Bauva 2017-10-27 15:06:37 +0000
598) goto fail;
5dfeec316b8 (Pranit Bauva 2017-10-27 15:06:37 +0000
606) goto fail;
3d3237b0e6b (Pranit Bauva 2017-10-27 15:06:37 +0000
686) return error(_("--bisect-reset requires either
no argument or a commit"));
0b1f0fd910c (Pranit Bauva 2017-10-27 15:06:37 +0000
690) return error(_("--bisect-write requires either 4
or 5 arguments"));
20edf353b72 (Pranit Bauva 2017-10-27 15:06:37 +0000
697) return error(_("--check-and-set-terms requires 3
arguments"));
a919f328ba3 (Pranit Bauva 2017-10-27 15:06:37 +0000
703) return error(_("--bisect-next-check requires 2
or 3 arguments"));
builtin/blame.c
74e8221b523 (Linus Torvalds 2018-07-07 15:02:35 -0700 922) case
DATE_HUMAN:
74e8221b523 (Linus Torvalds 2018-07-07 15:02:35 -0700
924) blame_date_width = sizeof("Thu Oct 19 16:00");
74e8221b523 (Linus Torvalds 2018-07-07 15:02:35 -0700
925) break;
builtin/gc.c
3029970275b (Jonathan Nieder 2018-07-16 23:57:40 -0700
461) ret = error_errno(_("cannot stat '%s'"), gc_log_path);
3029970275b (Jonathan Nieder 2018-07-16 23:57:40 -0700
470) ret = error_errno(_("cannot read '%s'"), gc_log_path);
fec2ed21871 (Jonathan Nieder 2018-07-16 23:54:16 -0700
495) die(FAILED_RUN, pack_refs_cmd.argv[0]);
fec2ed21871 (Jonathan Nieder 2018-07-16 23:54:16 -0700
498) die(FAILED_RUN, reflog.argv[0]);
3029970275b (Jonathan Nieder 2018-07-16 23:57:40 -0700
585) exit(128);
fec2ed21871 (Jonathan Nieder 2018-07-16 23:54:16 -0700
637) die(FAILED_RUN, repack.argv[0]);
fec2ed21871 (Jonathan Nieder 2018-07-16 23:54:16 -0700
647) die(FAILED_RUN, prune.argv[0]);
fec2ed21871 (Jonathan Nieder 2018-07-16 23:54:16 -0700
654) die(FAILED_RUN, prune_worktrees.argv[0]);
fec2ed21871 (Jonathan Nieder 2018-07-16 23:54:16 -0700
658) die(FAILED_RUN, rerere.argv[0]);
builtin/rebase--interactive.c
93420467efe (Alban Gruin 2018-08-28 14:10:42 +0200
24) return error(_("no HEAD?"));
93420467efe (Alban Gruin 2018-08-28 14:10:42 +0200
51) return error_errno(_("could not create temporary %s"),
path_state_dir());
93420467efe (Alban Gruin 2018-08-28 14:10:42 +0200
57) return error_errno(_("could not mark as interactive"));
93420467efe (Alban Gruin 2018-08-28 14:10:42 +0200
77) return -1;
93420467efe (Alban Gruin 2018-08-28 14:10:42 +0200
81) return -1;
93420467efe (Alban Gruin 2018-08-28 14:10:42 +0200
87) free(revisions);
93420467efe (Alban Gruin 2018-08-28 14:10:42 +0200
88) free(shortrevisions);
93420467efe (Alban Gruin 2018-08-28 14:10:42 +0200
90) return -1;
93420467efe (Alban Gruin 2018-08-28 14:10:42 +0200
98) free(revisions);
93420467efe (Alban Gruin 2018-08-28 14:10:42 +0200
99) free(shortrevisions);
93420467efe (Alban Gruin 2018-08-28 14:10:42 +0200
101) return error_errno(_("could not open %s"),
rebase_path_todo());
93420467efe (Alban Gruin 2018-08-28 14:10:42 +0200
106) argv_array_push(&make_script_args, restrict_revision);
93420467efe (Alban Gruin 2018-08-28 14:10:42 +0200
114) error(_("could not generate todo list"));
93420467efe (Alban Gruin 2018-08-28 14:10:42 +0200 205)
usage_with_options(builtin_rebase_interactive_usage, options);
93420467efe (Alban Gruin 2018-08-28 14:10:42 +0200
219) warning(_("--[no-]rebase-cousins has no effect without "
adb4f8f6b72 (Alban Gruin 2018-08-28 14:10:43 +0200
225) die(_("a base commit must be provided with
--upstream or --onto"));
b3fe2e1f8cb (Alban Gruin 2018-08-28 14:10:45 +0200 259) case
REARRANGE_SQUASH:
b3fe2e1f8cb (Alban Gruin 2018-08-28 14:10:45 +0200
260) ret = rearrange_squash();
b3fe2e1f8cb (Alban Gruin 2018-08-28 14:10:45 +0200
261) break;
b3fe2e1f8cb (Alban Gruin 2018-08-28 14:10:45 +0200 262) case
ADD_EXEC:
b3fe2e1f8cb (Alban Gruin 2018-08-28 14:10:45 +0200
263) ret = sequencer_add_exec_commands(cmd);
b3fe2e1f8cb (Alban Gruin 2018-08-28 14:10:45 +0200
264) break;
adb4f8f6b72 (Alban Gruin 2018-08-28 14:10:43 +0200 265) default:
adb4f8f6b72 (Alban Gruin 2018-08-28 14:10:43 +0200
266) BUG("invalid command '%d'", command);
builtin/rebase.c
55071ea248e (Pratik Karki 2018-08-07 01:16:09 +0545 61)
strbuf_trim(&out);
55071ea248e (Pratik Karki 2018-08-07 01:16:09 +0545 62) ret =
!strcmp("true", out.buf);
55071ea248e (Pratik Karki 2018-08-07 01:16:09 +0545 63)
strbuf_release(&out);
002ee2fe682 (Pratik Karki 2018-09-04 14:59:57 -0700 114) case
REBASE_AM:
002ee2fe682 (Pratik Karki 2018-09-04 14:59:57 -0700
115) die(_("%s requires an interactive rebase"), option);
f95736288a3 (Pratik Karki 2018-08-08 20:51:16 +0545
148) return error_errno(_("could not read '%s'"), path);
f95736288a3 (Pratik Karki 2018-08-08 20:51:16 +0545
162) return -1;
f95736288a3 (Pratik Karki 2018-08-08 20:51:16 +0545
167) return error(_("could not get 'onto': '%s'"), buf.buf);
f95736288a3 (Pratik Karki 2018-08-08 20:51:16 +0545
178) return -1;
f95736288a3 (Pratik Karki 2018-08-08 20:51:16 +0545 179) } else
if (read_one(state_dir_path("head", opts), &buf))
f95736288a3 (Pratik Karki 2018-08-08 20:51:16 +0545
180) return -1;
f95736288a3 (Pratik Karki 2018-08-08 20:51:16 +0545
182) return error(_("invalid orig-head: '%s'"), buf.buf);
f95736288a3 (Pratik Karki 2018-08-08 20:51:16 +0545
186) return -1;
f95736288a3 (Pratik Karki 2018-08-08 20:51:16 +0545
188) opts->flags &= ~REBASE_NO_QUIET;
73d51ed0a59 (Pratik Karki 2018-09-04 14:59:50 -0700
196) opts->signoff = 1;
73d51ed0a59 (Pratik Karki 2018-09-04 14:59:50 -0700
197) opts->flags |= REBASE_FORCE;
ead98c111b8 (Pratik Karki 2018-09-04 14:59:52 -0700
204) return -1;
28a02c5a790 (Pratik Karki 2018-09-04 15:00:00 -0700
219) return -1;
399a505296a (Pratik Karki 2018-09-04 15:00:11 -0700
227) return -1;
399a505296a (Pratik Karki 2018-09-04 15:00:11 -0700
235) return -1;
7debdaa4ad1 (Pratik Karki 2018-09-04 15:00:02 -0700
255) return error(_("Could not read '%s'"), path);
7debdaa4ad1 (Pratik Karki 2018-09-04 15:00:02 -0700
271) res = error(_("Cannot store %s"), autostash.buf);
7debdaa4ad1 (Pratik Karki 2018-09-04 15:00:02 -0700
275) return res;
b2263c13613 (Johannes Schindelin 2018-08-29 07:31:17 -0700
373) argv_array_pushf(&child.args,
b2263c13613 (Johannes Schindelin 2018-08-29 07:31:17 -0700
375) oid_to_hex(&opts->restrict_revision->object.oid));
ac7f467fef8 (Pratik Karki 2018-08-07 01:16:11 +0545 479) case
REBASE_INTERACTIVE:
ac7f467fef8 (Pratik Karki 2018-08-07 01:16:11 +0545
480) backend = "git-rebase--interactive";
ac7f467fef8 (Pratik Karki 2018-08-07 01:16:11 +0545
481) backend_func = "git_rebase__interactive";
ac7f467fef8 (Pratik Karki 2018-08-07 01:16:11 +0545
482) break;
ac7f467fef8 (Pratik Karki 2018-08-07 01:16:11 +0545 491) default:
ac7f467fef8 (Pratik Karki 2018-08-07 01:16:11 +0545
492) BUG("Unhandled rebase type %d", opts->type);
ac7f467fef8 (Pratik Karki 2018-08-07 01:16:11 +0545
509) struct strbuf dir = STRBUF_INIT;
7debdaa4ad1 (Pratik Karki 2018-09-04 15:00:02 -0700
511) apply_autostash(opts);
ac7f467fef8 (Pratik Karki 2018-08-07 01:16:11 +0545
512) strbuf_addstr(&dir, opts->state_dir);
ac7f467fef8 (Pratik Karki 2018-08-07 01:16:11 +0545
513) remove_dir_recursively(&dir, 0);
ac7f467fef8 (Pratik Karki 2018-08-07 01:16:11 +0545
514) strbuf_release(&dir);
ac7f467fef8 (Pratik Karki 2018-08-07 01:16:11 +0545
515) die("Nothing to do");
d4c569f8f4c (Pratik Karki 2018-09-04 14:27:20 -0700
542) BUG("Not a fully qualified branch: '%s'", switch_to_branch);
ac7f467fef8 (Pratik Karki 2018-08-07 01:16:11 +0545
545) return -1;
ac7f467fef8 (Pratik Karki 2018-08-07 01:16:11 +0545
549) rollback_lock_file(&lock);
ac7f467fef8 (Pratik Karki 2018-08-07 01:16:11 +0545
550) return error(_("could not determine HEAD
revision"));
ac7f467fef8 (Pratik Karki 2018-08-07 01:16:11 +0545
567) rollback_lock_file(&lock);
ac7f467fef8 (Pratik Karki 2018-08-07 01:16:11 +0545
568) return error(_("could not read index"));
ac7f467fef8 (Pratik Karki 2018-08-07 01:16:11 +0545
572) error(_("failed to find tree of %s"), oid_to_hex(oid));
ac7f467fef8 (Pratik Karki 2018-08-07 01:16:11 +0545
573) rollback_lock_file(&lock);
ac7f467fef8 (Pratik Karki 2018-08-07 01:16:11 +0545
574) free((void *)desc.buffer);
ac7f467fef8 (Pratik Karki 2018-08-07 01:16:11 +0545
575) return -1;
ac7f467fef8 (Pratik Karki 2018-08-07 01:16:11 +0545
588) ret = error(_("could not write index"));
ac7f467fef8 (Pratik Karki 2018-08-07 01:16:11 +0545
592) return ret;
ac7f467fef8 (Pratik Karki 2018-08-07 01:16:11 +0545 608) } else
if (old_orig)
ac7f467fef8 (Pratik Karki 2018-08-07 01:16:11 +0545
609) delete_ref(NULL, "ORIG_HEAD", old_orig, 0);
bff014dac7d (Pratik Karki 2018-09-04 14:27:13 -0700
637) opts->flags &= !REBASE_DIFFSTAT;
9a48a615b47 (Pratik Karki 2018-09-04 14:27:16 -0700
671) return 1;
9a48a615b47 (Pratik Karki 2018-09-04 14:27:16 -0700
687) return 0;
55071ea248e (Pratik Karki 2018-08-07 01:16:09 +0545
894) const char *path = mkpath("%s/git-legacy-rebase",
55071ea248e (Pratik Karki 2018-08-07 01:16:09 +0545
897) if (sane_execvp(path, (char **)argv) < 0)
55071ea248e (Pratik Karki 2018-08-07 01:16:09 +0545
898) die_errno(_("could not exec %s"), path);
55071ea248e (Pratik Karki 2018-08-07 01:16:09 +0545
900) BUG("sane_execvp() returned???");
0eabf4b95ca (Pratik Karki 2018-08-08 20:51:22 +0545
916) die(_("It looks like 'git am' is in progress. Cannot
rebase."));
f28d40d3a99 (Pratik Karki 2018-09-04 14:27:07 -0700
953) usage_with_options(builtin_rebase_usage,
f95736288a3 (Pratik Karki 2018-08-08 20:51:16 +0545
973) die(_("Cannot read HEAD"));
f95736288a3 (Pratik Karki 2018-08-08 20:51:16 +0545
977) die(_("could not read index"));
f95736288a3 (Pratik Karki 2018-08-08 20:51:16 +0545
991) exit(1);
122420c2953 (Pratik Karki 2018-08-08 20:51:17 +0545
1003) die(_("could not discard worktree changes"));
122420c2953 (Pratik Karki 2018-08-08 20:51:17 +0545
1005) exit(1);
5e5d96197ca (Pratik Karki 2018-08-08 20:51:18 +0545
1016) exit(1);
5e5d96197ca (Pratik Karki 2018-08-08 20:51:18 +0545
1019) die(_("could not move back to %s"),
5a61494539b (Pratik Karki 2018-08-08 20:51:19 +0545
1029) die(_("could not remove '%s'"), options.state_dir);
f95736288a3 (Pratik Karki 2018-08-08 20:51:16 +0545 1042) default:
51e9ea6da76 (Pratik Karki 2018-08-08 20:51:20 +0545
1043) BUG("action: %d", action);
c54dacb50ea (Pratik Karki 2018-09-04 14:27:18 -0700
1048) const char *last_slash = strrchr(options.state_dir, '/');
c54dacb50ea (Pratik Karki 2018-09-04 14:27:18 -0700
1049) const char *state_dir_base =
c54dacb50ea (Pratik Karki 2018-09-04 14:27:18 -0700
1050) last_slash ? last_slash + 1 : options.state_dir;
c54dacb50ea (Pratik Karki 2018-09-04 14:27:18 -0700
1051) const char *cmd_live_rebase =
c54dacb50ea (Pratik Karki 2018-09-04 14:27:18 -0700
1053) strbuf_reset(&buf);
c54dacb50ea (Pratik Karki 2018-09-04 14:27:18 -0700
1054) strbuf_addf(&buf, "rm -fr \"%s\"", options.state_dir);
c54dacb50ea (Pratik Karki 2018-09-04 14:27:18 -0700
1055) die(_("It seems that there is already a %s directory, and\n"
53f9e5be94e (Pratik Karki 2018-09-04 14:59:56 -0700
1079) strbuf_addstr(&options.git_am_opt, " --ignore-date");
53f9e5be94e (Pratik Karki 2018-09-04 14:59:56 -0700
1080) options.flags |= REBASE_FORCE;
c7ee2134d42 (Pratik Karki 2018-09-04 15:00:01 -0700
1092) strbuf_addf(&options.git_am_opt, " -C%d", opt_c);
0073df2bd31 (Pratik Karki 2018-09-04 15:00:07 -0700
1124) else if (strcmp("no-rebase-cousins", rebase_merges))
0073df2bd31 (Pratik Karki 2018-09-04 15:00:07 -0700
1125) die(_("Unknown mode: %s"), rebase_merges);
399a505296a (Pratik Karki 2018-09-04 15:00:11 -0700
1146) case REBASE_AM:
399a505296a (Pratik Karki 2018-09-04 15:00:11 -0700
1147) die(_("--strategy requires --merge or
--interactive"));
399a505296a (Pratik Karki 2018-09-04 15:00:11 -0700
1156) default:
399a505296a (Pratik Karki 2018-09-04 15:00:11 -0700
1157) BUG("unhandled rebase type (%d)", options.type);
6dc73173f6c (Pratik Karki 2018-08-08 21:21:33 +0545
1165) strbuf_addstr(&options.git_format_patch_opt, " --progress");
ac7f467fef8 (Pratik Karki 2018-08-07 01:16:11 +0545 1173) case
REBASE_AM:
ac7f467fef8 (Pratik Karki 2018-08-07 01:16:11 +0545
1174) options.state_dir = apply_dir();
ac7f467fef8 (Pratik Karki 2018-08-07 01:16:11 +0545
1175) break;
ac7f467fef8 (Pratik Karki 2018-08-07 01:16:11 +0545
1252) die(_("invalid upstream '%s'"),
options.upstream_name);
bfa5147095f (Pratik Karki 2018-09-04 15:00:12 -0700
1258) die(_("Could not create new root commit"));
e65123a71d0 (Pratik Karki 2018-09-04 14:27:21 -0700
1308) die(_("fatal: no such branch/commit '%s'"),
ac7f467fef8 (Pratik Karki 2018-08-07 01:16:11 +0545
1316) die(_("No such ref: %s"), "HEAD");
ac7f467fef8 (Pratik Karki 2018-08-07 01:16:11 +0545
1328) die(_("Could not resolve HEAD to a revision"));
e65123a71d0 (Pratik Karki 2018-09-04 14:27:21 -0700
1330) BUG("unexpected number of arguments left to parse");
e0333e5c63f (Pratik Karki 2018-09-04 14:27:14 -0700
1341) die(_("could not read index"));
7debdaa4ad1 (Pratik Karki 2018-09-04 15:00:02 -0700
1368) die(_("Cannot autostash"));
7debdaa4ad1 (Pratik Karki 2018-09-04 15:00:02 -0700
1371) die(_("Unexpected stash response: '%s'"),
7debdaa4ad1 (Pratik Karki 2018-09-04 15:00:02 -0700
1377) die(_("Could not create directory for
'%s'"),
7debdaa4ad1 (Pratik Karki 2018-09-04 15:00:02 -0700
1383) die(_("could not reset --hard"));
e65123a71d0 (Pratik Karki 2018-09-04 14:27:21 -0700
1427) ret = !!error(_("could not parse
'%s'"),
e65123a71d0 (Pratik Karki 2018-09-04 14:27:21 -0700
1429) goto cleanup;
e65123a71d0 (Pratik Karki 2018-09-04 14:27:21 -0700
1438) ret = !!error(_("could not
switch to "
1ed9c14ff25 (Pratik Karki 2018-09-04 14:27:17 -0700
1448) resolve_ref_unsafe("HEAD", 0, NULL, &flag))
1ed9c14ff25 (Pratik Karki 2018-09-04 14:27:17 -0700
1449) puts(_("HEAD is up to date."));
9a48a615b47 (Pratik Karki 2018-09-04 14:27:16 -0700
1458) resolve_ref_unsafe("HEAD", 0, NULL, &flag))
9a48a615b47 (Pratik Karki 2018-09-04 14:27:16 -0700
1459) puts(_("HEAD is up to date, rebase forced."));
builtin/rev-list.c
0eee403f2f7 (Matthew DeVore 2018-09-04 11:05:47 -0700
227) die("unexpected missing %s object '%s'",
0eee403f2f7 (Matthew DeVore 2018-09-04 11:05:47 -0700
228) type_name(obj->type), oid_to_hex(&obj->oid));
builtin/stash.c
93871263d18 (Joel Teichroeb 2018-08-31 00:40:36 +0300
130) free_stash_info(info);
93871263d18 (Joel Teichroeb 2018-08-31 00:40:36 +0300
131) error(_("'%s' is not a stash-like commit"), revision);
93871263d18 (Joel Teichroeb 2018-08-31 00:40:36 +0300
132) exit(128);
93871263d18 (Joel Teichroeb 2018-08-31 00:40:36 +0300
165) free_stash_info(info);
93871263d18 (Joel Teichroeb 2018-08-31 00:40:36 +0300
166) fprintf_ln(stderr, _("No stash entries found."));
93871263d18 (Joel Teichroeb 2018-08-31 00:40:36 +0300
167) return -1;
93871263d18 (Joel Teichroeb 2018-08-31 00:40:36 +0300 201)
default: /* Invalid or ambiguous */
93871263d18 (Joel Teichroeb 2018-08-31 00:40:36 +0300
202) free_stash_info(info);
31f109a3618 (Joel Teichroeb 2018-08-31 00:40:37 +0300
229) return error(_("git stash clear with parameters is
unimplemented"));
93871263d18 (Joel Teichroeb 2018-08-31 00:40:36 +0300
244) return -1;
93871263d18 (Joel Teichroeb 2018-08-31 00:40:36 +0300
252) return -1;
93871263d18 (Joel Teichroeb 2018-08-31 00:40:36 +0300
265) return -1;
93871263d18 (Joel Teichroeb 2018-08-31 00:40:36 +0300
268) return error(_("unable to write new index file"));
93871263d18 (Joel Teichroeb 2018-08-31 00:40:36 +0300
374) remove_path(stash_index_path.buf);
93871263d18 (Joel Teichroeb 2018-08-31 00:40:36 +0300
375) return -1;
93871263d18 (Joel Teichroeb 2018-08-31 00:40:36 +0300
402) return -1;
93871263d18 (Joel Teichroeb 2018-08-31 00:40:36 +0300
405) return error(_("Cannot apply a stash in the middle of a
merge"));
93871263d18 (Joel Teichroeb 2018-08-31 00:40:36 +0300
415) strbuf_release(&out);
93871263d18 (Joel Teichroeb 2018-08-31 00:40:36 +0300
416) return -1;
93871263d18 (Joel Teichroeb 2018-08-31 00:40:36 +0300
422) return -1;
93871263d18 (Joel Teichroeb 2018-08-31 00:40:36 +0300
427) return -1;
93871263d18 (Joel Teichroeb 2018-08-31 00:40:36 +0300
434) return error(_("Could not restore untracked files from
stash"));
93871263d18 (Joel Teichroeb 2018-08-31 00:40:36 +0300
465) return -1;
93871263d18 (Joel Teichroeb 2018-08-31 00:40:36 +0300
470) strbuf_release(&out);
93871263d18 (Joel Teichroeb 2018-08-31 00:40:36 +0300
475) strbuf_release(&out);
93871263d18 (Joel Teichroeb 2018-08-31 00:40:36 +0300
476) return -1;
31f109a3618 (Joel Teichroeb 2018-08-31 00:40:37 +0300
551) return error(_("%s: Could not drop stash entry"),
b3513da4bd9 (Joel Teichroeb 2018-08-31 00:40:39 +0300
623) printf_ln(_("The stash entry is kept in case you need it
again."));
8ceb24b2c38 (Paul-Sebastian Ungureanu 2018-08-31 00:40:41
+0300 754) free_stash_info(&info);
129f0b0a009 (Paul-Sebastian Ungureanu 2018-08-31 00:40:48
+0300 755) usage_with_options(git_stash_show_usage, options);
0ac06fb81f2 (Paul-Sebastian Ungureanu 2018-08-31 00:40:43
+0300 808) fprintf_ln(stderr, _("\"git stash store\"
requires one <commit> argument"));
0ac06fb81f2 (Paul-Sebastian Ungureanu 2018-08-31 00:40:43
+0300 809) return -1;
f6f191b3f25 (Paul-Sebastian Ungureanu 2018-08-31 00:40:44
+0300 884) return 1;
f6f191b3f25 (Paul-Sebastian Ungureanu 2018-08-31 00:40:44
+0300 925) ret = -1;
f6f191b3f25 (Paul-Sebastian Ungureanu 2018-08-31 00:40:44
+0300 926) goto done;
f6f191b3f25 (Paul-Sebastian Ungureanu 2018-08-31 00:40:44
+0300 931) ret = -1;
f6f191b3f25 (Paul-Sebastian Ungureanu 2018-08-31 00:40:44
+0300 932) goto done;
f6f191b3f25 (Paul-Sebastian Ungureanu 2018-08-31 00:40:44
+0300 937) ret = -1;
f6f191b3f25 (Paul-Sebastian Ungureanu 2018-08-31 00:40:44
+0300 938) goto done;
f6f191b3f25 (Paul-Sebastian Ungureanu 2018-08-31 00:40:44
+0300 966) ret = -1;
f6f191b3f25 (Paul-Sebastian Ungureanu 2018-08-31 00:40:44
+0300 967) goto done;
f6f191b3f25 (Paul-Sebastian Ungureanu 2018-08-31 00:40:44
+0300 978) ret = -1;
f6f191b3f25 (Paul-Sebastian Ungureanu 2018-08-31 00:40:44
+0300 979) goto done;
f6f191b3f25 (Paul-Sebastian Ungureanu 2018-08-31 00:40:44
+0300 984) ret = -1;
f6f191b3f25 (Paul-Sebastian Ungureanu 2018-08-31 00:40:44
+0300 985) goto done;
f6f191b3f25 (Paul-Sebastian Ungureanu 2018-08-31 00:40:44
+0300 992) ret = -1;
f6f191b3f25 (Paul-Sebastian Ungureanu 2018-08-31 00:40:44
+0300 993) goto done;
f6f191b3f25 (Paul-Sebastian Ungureanu 2018-08-31 00:40:44
+0300 1018) ret = -1;
f6f191b3f25 (Paul-Sebastian Ungureanu 2018-08-31 00:40:44
+0300 1019) goto done;
f6f191b3f25 (Paul-Sebastian Ungureanu 2018-08-31 00:40:44
+0300 1031) ret = -1;
f6f191b3f25 (Paul-Sebastian Ungureanu 2018-08-31 00:40:44
+0300 1032) goto done;
f6f191b3f25 (Paul-Sebastian Ungureanu 2018-08-31 00:40:44
+0300 1037) ret = -1;
f6f191b3f25 (Paul-Sebastian Ungureanu 2018-08-31 00:40:44
+0300 1038) goto done;
f6f191b3f25 (Paul-Sebastian Ungureanu 2018-08-31 00:40:44
+0300 1049) ret = -1;
f6f191b3f25 (Paul-Sebastian Ungureanu 2018-08-31 00:40:44
+0300 1050) goto done;
f6f191b3f25 (Paul-Sebastian Ungureanu 2018-08-31 00:40:44
+0300 1055) ret = -1;
f6f191b3f25 (Paul-Sebastian Ungureanu 2018-08-31 00:40:44
+0300 1056) goto done;
8002b9e6264 (Paul-Sebastian Ungureanu 2018-08-31 00:40:46
+0300 1089) fprintf_ln(stderr, _("You do not
have the initial commit yet"));
8002b9e6264 (Paul-Sebastian Ungureanu 2018-08-31 00:40:46
+0300 1110) if (!quiet)
8002b9e6264 (Paul-Sebastian Ungureanu 2018-08-31 00:40:46
+0300 1111) fprintf_ln(stderr, _("Cannot save
the current index state"));
f6f191b3f25 (Paul-Sebastian Ungureanu 2018-08-31 00:40:44
+0300 1112) ret = -1;
f6f191b3f25 (Paul-Sebastian Ungureanu 2018-08-31 00:40:44
+0300 1113) *stash_msg = NULL;
f6f191b3f25 (Paul-Sebastian Ungureanu 2018-08-31 00:40:44
+0300 1114) goto done;
8002b9e6264 (Paul-Sebastian Ungureanu 2018-08-31 00:40:46
+0300 1119) if (!quiet)
8002b9e6264 (Paul-Sebastian Ungureanu 2018-08-31 00:40:46
+0300 1120) fprintf_ln(stderr, _("Cannot save the untracked files"));
f6f191b3f25 (Paul-Sebastian Ungureanu 2018-08-31 00:40:44
+0300 1121) ret = -1;
f6f191b3f25 (Paul-Sebastian Ungureanu 2018-08-31 00:40:44
+0300 1122) *stash_msg = NULL;
f6f191b3f25 (Paul-Sebastian Ungureanu 2018-08-31 00:40:44
+0300 1123) goto done;
8002b9e6264 (Paul-Sebastian Ungureanu 2018-08-31 00:40:46
+0300 1131) if (!quiet)
8002b9e6264 (Paul-Sebastian Ungureanu 2018-08-31 00:40:46
+0300 1132) fprintf_ln(stderr, _("Cannot save the current worktree
state"));
f6f191b3f25 (Paul-Sebastian Ungureanu 2018-08-31 00:40:44
+0300 1133) goto done;
8002b9e6264 (Paul-Sebastian Ungureanu 2018-08-31 00:40:46
+0300 1139) if (!quiet)
8002b9e6264 (Paul-Sebastian Ungureanu 2018-08-31 00:40:46
+0300 1140) fprintf_ln(stderr, _("Cannot save the current worktree
state"));
f6f191b3f25 (Paul-Sebastian Ungureanu 2018-08-31 00:40:44
+0300 1141) ret = -1;
f6f191b3f25 (Paul-Sebastian Ungureanu 2018-08-31 00:40:44
+0300 1142) *stash_msg = NULL;
f6f191b3f25 (Paul-Sebastian Ungureanu 2018-08-31 00:40:44
+0300 1143) goto done;
8002b9e6264 (Paul-Sebastian Ungureanu 2018-08-31 00:40:46
+0300 1166) if (!quiet)
8002b9e6264 (Paul-Sebastian Ungureanu 2018-08-31 00:40:46
+0300 1167) fprintf_ln(stderr, _("Cannot record
working tree state"));
f6f191b3f25 (Paul-Sebastian Ungureanu 2018-08-31 00:40:44
+0300 1168) ret = -1;
f6f191b3f25 (Paul-Sebastian Ungureanu 2018-08-31 00:40:44
+0300 1169) goto done;
48c061fa443 (Paul-Sebastian Ungureanu 2018-08-31 00:40:45
+0300 1251) return -1;
8002b9e6264 (Paul-Sebastian Ungureanu 2018-08-31 00:40:46
+0300 1260) if (!quiet)
8002b9e6264 (Paul-Sebastian Ungureanu 2018-08-31 00:40:46
+0300 1261) fprintf_ln(stderr, _("Cannot
initialize stash"));
48c061fa443 (Paul-Sebastian Ungureanu 2018-08-31 00:40:45
+0300 1262) return -1;
8002b9e6264 (Paul-Sebastian Ungureanu 2018-08-31 00:40:46
+0300 1272) if (!quiet)
8002b9e6264 (Paul-Sebastian Ungureanu 2018-08-31 00:40:46
+0300 1273) fprintf_ln(stderr, _("Cannot save
the current status"));
48c061fa443 (Paul-Sebastian Ungureanu 2018-08-31 00:40:45
+0300 1274) ret = -1;
48c061fa443 (Paul-Sebastian Ungureanu 2018-08-31 00:40:45
+0300 1275) goto done;
48c061fa443 (Paul-Sebastian Ungureanu 2018-08-31 00:40:45
+0300 1292) ret = -1;
48c061fa443 (Paul-Sebastian Ungureanu 2018-08-31 00:40:45
+0300 1311) ret = -1;
48c061fa443 (Paul-Sebastian Ungureanu 2018-08-31 00:40:45
+0300 1312) goto done;
48c061fa443 (Paul-Sebastian Ungureanu 2018-08-31 00:40:45
+0300 1321) ret = -1;
48c061fa443 (Paul-Sebastian Ungureanu 2018-08-31 00:40:45
+0300 1322) goto done;
48c061fa443 (Paul-Sebastian Ungureanu 2018-08-31 00:40:45
+0300 1330) ret = -1;
48c061fa443 (Paul-Sebastian Ungureanu 2018-08-31 00:40:45
+0300 1339) ret = -1;
48c061fa443 (Paul-Sebastian Ungureanu 2018-08-31 00:40:45
+0300 1350) ret = -1;
48c061fa443 (Paul-Sebastian Ungureanu 2018-08-31 00:40:45
+0300 1359) ret = -1;
48c061fa443 (Paul-Sebastian Ungureanu 2018-08-31 00:40:45
+0300 1360) goto done;
48c061fa443 (Paul-Sebastian Ungureanu 2018-08-31 00:40:45
+0300 1368) ret = -1;
48c061fa443 (Paul-Sebastian Ungureanu 2018-08-31 00:40:45
+0300 1394) ret = -1;
129f0b0a009 (Paul-Sebastian Ungureanu 2018-08-31 00:40:48
+0300 1524) usage_msg_opt(xstrfmt(_("unknown subcommand: %s"),
argv[0]),
129f0b0a009 (Paul-Sebastian Ungureanu 2018-08-31 00:40:48
+0300 1554) continue;
builtin/submodule--helper.c
9d34daefb74 (Antonio Ospite 2018-08-14 13:05:22 +0200 2174)
die("submodule--helper config takes 1 or 2 arguments: name [value]");
commit-graph.c
5cef295f283 (Derrick Stolee 2018-08-20 18:24:32 +0000
66) return 0;
20fd6d57996 (Derrick Stolee 2018-08-20 18:24:30 +0000
78) return 0;
date.c
74e8221b523 (Linus Torvalds 2018-07-07 15:02:35 -0700
113) die("Timestamp too large for this system: %"PRItime, time);
74e8221b523 (Linus Torvalds 2018-07-07 15:02:35 -0700
216) if (tm->tm_mon == human_tm->tm_mon) {
74e8221b523 (Linus Torvalds 2018-07-07 15:02:35 -0700
217) if (tm->tm_mday > human_tm->tm_mday) {
74e8221b523 (Linus Torvalds 2018-07-07 15:02:35 -0700
219) } else if (tm->tm_mday == human_tm->tm_mday) {
74e8221b523 (Linus Torvalds 2018-07-07 15:02:35 -0700
220) hide.date = hide.wday = 1;
74e8221b523 (Linus Torvalds 2018-07-07 15:02:35 -0700
221) } else if (tm->tm_mday + 5 > human_tm->tm_mday) {
74e8221b523 (Linus Torvalds 2018-07-07 15:02:35 -0700
223) hide.date = 1;
74e8221b523 (Linus Torvalds 2018-07-07 15:02:35 -0700
231) gettimeofday(&now, NULL);
74e8221b523 (Linus Torvalds 2018-07-07 15:02:35 -0700
232) show_date_relative(time, tz, &now, buf);
74e8221b523 (Linus Torvalds 2018-07-07 15:02:35 -0700
233) return;
74e8221b523 (Linus Torvalds 2018-07-07 15:02:35 -0700
246) hide.seconds = 1;
74e8221b523 (Linus Torvalds 2018-07-07 15:02:35 -0700
247) hide.tz |= !hide.date;
74e8221b523 (Linus Torvalds 2018-07-07 15:02:35 -0700
248) hide.wday = hide.time = !hide.year;
74e8221b523 (Linus Torvalds 2018-07-07 15:02:35 -0700
262) strbuf_rtrim(buf);
74e8221b523 (Linus Torvalds 2018-07-07 15:02:35 -0700
287) gettimeofday(&now, NULL);
74e8221b523 (Linus Torvalds 2018-07-07 15:02:35 -0700
290) human_tz = local_time_tzoffset(now.tv_sec, &human_tm);
74e8221b523 (Linus Torvalds 2018-07-07 15:02:35 -0700 886)static int
auto_date_style(void)
74e8221b523 (Linus Torvalds 2018-07-07 15:02:35 -0700 888) return
(isatty(1) || pager_in_use()) ? DATE_HUMAN : DATE_NORMAL;
74e8221b523 (Linus Torvalds 2018-07-07 15:02:35 -0700
909) return DATE_HUMAN;
74e8221b523 (Linus Torvalds 2018-07-07 15:02:35 -0700
911) return auto_date_style();
diff-lib.c
fa49029577c (Alex Vandiver 2018-01-02 19:04:56 -0800
205) continue;
list-objects-filter.c
dbebce653b5 (Matthew DeVore 2018-09-04 11:05:49 -0700
47) BUG("unknown filter_situation: %d", filter_situation);
7329b7c6cdc (Matthew DeVore 2018-09-04 11:05:50 -0700 100) default:
7329b7c6cdc (Matthew DeVore 2018-09-04 11:05:50 -0700
101) BUG("unknown filter_situation: %d", filter_situation);
dbebce653b5 (Matthew DeVore 2018-09-04 11:05:49 -0700
152) BUG("unknown filter_situation: %d", filter_situation);
dbebce653b5 (Matthew DeVore 2018-09-04 11:05:49 -0700
257) BUG("unknown filter_situation: %d", filter_situation);
dbebce653b5 (Matthew DeVore 2018-09-04 11:05:49 -0700
438) BUG("invalid list-objects filter choice: %d",
list-objects.c
f447a499dbb (Matthew DeVore 2018-08-13 11:14:28 -0700
197) ctx->show_object(obj, base->buf, ctx->show_data);
rebase-interactive.c
64a43cbd5da (Alban Gruin 2018-08-10 18:51:31 +0200
61) return error_errno(_("could not read '%s'."), todo_file);
64a43cbd5da (Alban Gruin 2018-08-10 18:51:31 +0200
65) strbuf_release(&buf);
64a43cbd5da (Alban Gruin 2018-08-10 18:51:31 +0200
66) return -1;
a9f5476fbca (Alban Gruin 2018-08-10 18:51:35 +0200
74) return error_errno(_("could not read '%s'."), todo_file);
a9f5476fbca (Alban Gruin 2018-08-10 18:51:35 +0200
78) strbuf_release(&buf);
a9f5476fbca (Alban Gruin 2018-08-10 18:51:35 +0200
79) return -1;
64a43cbd5da (Alban Gruin 2018-08-10 18:51:31 +0200
85) return -1;
refs.c
4a6067cda51 (Stefan Beller 2018-08-20 18:24:16 +0000
1405) return 0;
sequencer.c
65850686cf0 (Alban Gruin 2018-08-28 14:10:40 +0200
2276) return;
65850686cf0 (Alban Gruin 2018-08-28 14:10:40 +0200
2373) write_file(rebase_path_quiet(), "%s\n", quiet);
2c58483a598 (Alban Gruin 2018-08-10 18:51:33 +0200
3371) return error(_("could not checkout %s"), commit);
4df66c40b08 (Alban Gruin 2018-08-10 18:51:34 +0200
3385) return error(_("%s: not a valid OID"), orig_head);
b97e1873649 (Alban Gruin 2018-08-28 14:10:36 +0200
4750) return -1;
b97e1873649 (Alban Gruin 2018-08-28 14:10:36 +0200
4753) return -1;
b97e1873649 (Alban Gruin 2018-08-28 14:10:36 +0200
4759) return error_errno(_("could not read '%s'."), todo_file);
b97e1873649 (Alban Gruin 2018-08-28 14:10:36 +0200
4762) todo_list_release(&todo_list);
b97e1873649 (Alban Gruin 2018-08-28 14:10:36 +0200
4763) return error(_("unusable todo list: '%s'"), todo_file);
b97e1873649 (Alban Gruin 2018-08-28 14:10:36 +0200
4782) todo_list_release(&todo_list);
b97e1873649 (Alban Gruin 2018-08-28 14:10:36 +0200
4783) return -1;
b97e1873649 (Alban Gruin 2018-08-28 14:10:36 +0200
4787) return error(_("could not copy '%s' to '%s'."), todo_file,
b97e1873649 (Alban Gruin 2018-08-28 14:10:36 +0200
4791) return error(_("could not transform the todo list"));
b97e1873649 (Alban Gruin 2018-08-28 14:10:36 +0200
4820) return error(_("could not transform the todo list"));
b97e1873649 (Alban Gruin 2018-08-28 14:10:36 +0200
4823) return error(_("could not skip unnecessary pick commands"));
b97e1873649 (Alban Gruin 2018-08-28 14:10:36 +0200
4829) return -1;
strbuf.c
f95736288a3 (Pratik Karki 2018-08-08 20:51:16 +0545
127) --sb->len;
submodule-config.c
83a66534e63 (Antonio Ospite 2018-08-14 13:05:19 +0200
714) return CONFIG_INVALID_KEY;
150984f5339 (Antonio Ospite 2018-08-14 13:05:20 +0200
729) warning(_("Could not update .gitmodules entry %s"), key);
t/helper/test-dump-fsmonitor.c
6e1123ec573 (Alex Vandiver 2018-01-02 19:04:54 -0800
25) valid++;
t/helper/test-repository.c
b7758963424 (Derrick Stolee 2018-08-20 18:24:24 +0000
21) die("Couldn't init repo");
b7758963424 (Derrick Stolee 2018-08-20 18:24:24 +0000
47) die("Couldn't init repo");
wrapper.c
7e621449185 (Pranit Bauva 2017-10-27 15:06:37 +0000
701) die_errno(_("could not stat %s"), filename);
wt-status.c
f3bd35fa0dd (Stephen P. Smith 2018-09-05 17:53:29 -0700
671) s->committable = 1;
1c623c066c2 (Junio C Hamano 2018-09-07 15:32:24 -0700
1949) if (s->state.rebase_in_progress ||
1c623c066c2 (Junio C Hamano 2018-09-07 15:32:24 -0700 1950)
s->state.rebase_interactive_in_progress)
1c623c066c2 (Junio C Hamano 2018-09-07 15:32:24 -0700
1951) branch_name = s->state.onto;
1c623c066c2 (Junio C Hamano 2018-09-07 15:32:24 -0700
1952) else if (s->state.detached_from)
1c623c066c2 (Junio C Hamano 2018-09-07 15:32:24 -0700
1953) branch_name = s->state.detached_from;
^ permalink raw reply [relevance 1%]
* Re: [PATCH 4/9] submodule.c: sort changed_submodule_names before searching it
2018-09-12 18:18 6% ` Junio C Hamano
@ 2018-09-12 19:06 4% ` Stefan Beller
0 siblings, 0 replies; 200+ results
From: Stefan Beller @ 2018-09-12 19:06 UTC (permalink / raw)
To: Junio C Hamano; +Cc: git
On Wed, Sep 12, 2018 at 11:18 AM Junio C Hamano <gitster@pobox.com> wrote:
>
> Stefan Beller <sbeller@google.com> writes:
>
> > We can string_list_insert() to maintain sorted-ness of the
> > list as we find new items, or we can string_list_append() to
> > build an unsorted list and sort it at the end just once.
> >
> > To pick which one is more appropriate, we notice the fact
> > that we discover new items more or less in the already
> > sorted order. That makes "append then sort" more
> > appropriate.
>
> Sorry, but I still do not get the math you are implying in the
> second paragraph. Are you saying that append-then-sort is efficient
> when items being appended is already sorted? That depends on the
> sorting algorithm used, so the logic is incomplete unless you say
> "given that we use X for sorting,...", I think.
>
> Do we really discover new items in sorted order, by the way? In a
> single diff invocation made inside collect_changed_submodules() for
> one commit in the superproject's history, we will grab changed paths
> in the pathname order (i.e. sorted); if the superproject's tip commit
> touches the submodules at paths A and Z, we will discover these two
> paths in sorted order.
>
> But because we are walking the superproject's history to collect all
> paths that have been affected in that function, and repeatedly
> calling diff as we discover commit in the superproject's history, I
> am not sure how well the resulting set of paths would be sorted.
>
> The tip commit in superproject's history may have modified the
> submodule at path X, the parent of that commit may have touched the
> submodule at path M, and its parent may have touched the submodule
> at path A. Don't we end up grabbing these paths in that discoverd
> order, i.e. X, M and A?
That is true.
>
> I still think changing it from "insert as we find an item, keeping
> the list sorted" to "append all and then sort before we start
> looking things up from the result" makes sense, but I do not think
> the "we find things in sorted order" is either true, or it would
> affect the choice between the two. A justification to choose the
> latter I can think of that makes sense is that we don't have to pay
> cost to keep the list sorted while building it because we do not do
> any look-up while building the list.
ok.
Thanks,
Stefan
^ permalink raw reply [relevance 4%]
* Re: [PATCH 8/9] fetch: retry fetching submodules if sha1 were not fetched
2018-09-11 23:49 18% ` [PATCH 8/9] fetch: retry fetching submodules if sha1 were not fetched Stefan Beller
@ 2018-09-12 19:03 6% ` Junio C Hamano
0 siblings, 0 replies; 200+ results
From: Junio C Hamano @ 2018-09-12 19:03 UTC (permalink / raw)
To: Stefan Beller; +Cc: git
Stefan Beller <sbeller@google.com> writes:
> Retry fetching a submodule if the object id that the superproject points
> to, cannot be found.
By object name? By attempting to fetch all refs? Or by doing
something else? Fetching by the exact object name is the most
efficient approach, but the server side may not be prepared to
serve such a request, and that is why spelling it out here would
help the readers.
> This doesn't support fetching to FETCH_HEAD yet, but only into a local
> branch.
It is not clear if this sentence is talking about the fetch done at
the superproject level, or what happens in a submodule repository
when this "retrying" happens. Assuming that it is the former,
perhaps
This retrying does not happen when the "git fetch" done at the
superproject is not storing the fetched results in remote
tracking branches (i.e. instead just recording them to
FETCH_HEAD) in this step.
would help the readers understand what you are trying to say.
> To make fetching into FETCH_HEAD work, we need some refactoring
> in builtin/fetch.c to adjust the calls to 'check_for_new_submodule_commits'
> that is coming in the next patch.
>
> Signed-off-by: Stefan Beller <sbeller@google.com>
> ---
> builtin/fetch.c | 9 ++--
> submodule.c | 87 ++++++++++++++++++++++++++++++++++++-
> t/t5526-fetch-submodules.sh | 16 +++++++
> 3 files changed, 104 insertions(+), 8 deletions(-)
>
> diff --git a/builtin/fetch.c b/builtin/fetch.c
> index 61bec5d213d..95c44bf6ffa 100644
> --- a/builtin/fetch.c
> +++ b/builtin/fetch.c
> @@ -700,8 +700,7 @@ static int update_local_ref(struct ref *ref,
> what = _("[new ref]");
> }
>
> - if ((recurse_submodules != RECURSE_SUBMODULES_OFF) &&
> - (recurse_submodules != RECURSE_SUBMODULES_ON))
> + if (recurse_submodules != RECURSE_SUBMODULES_OFF)
> check_for_new_submodule_commits(&ref->new_oid);
> r = s_update_ref(msg, ref, 0);
> format_display(display, r ? '!' : '*', what,
> @@ -716,8 +715,7 @@ static int update_local_ref(struct ref *ref,
> strbuf_add_unique_abbrev(&quickref, ¤t->object.oid, DEFAULT_ABBREV);
> strbuf_addstr(&quickref, "..");
> strbuf_add_unique_abbrev(&quickref, &ref->new_oid, DEFAULT_ABBREV);
> - if ((recurse_submodules != RECURSE_SUBMODULES_OFF) &&
> - (recurse_submodules != RECURSE_SUBMODULES_ON))
> + if (recurse_submodules != RECURSE_SUBMODULES_OFF)
> check_for_new_submodule_commits(&ref->new_oid);
> r = s_update_ref("fast-forward", ref, 1);
> format_display(display, r ? '!' : ' ', quickref.buf,
> @@ -731,8 +729,7 @@ static int update_local_ref(struct ref *ref,
> strbuf_add_unique_abbrev(&quickref, ¤t->object.oid, DEFAULT_ABBREV);
> strbuf_addstr(&quickref, "...");
> strbuf_add_unique_abbrev(&quickref, &ref->new_oid, DEFAULT_ABBREV);
> - if ((recurse_submodules != RECURSE_SUBMODULES_OFF) &&
> - (recurse_submodules != RECURSE_SUBMODULES_ON))
> + if (recurse_submodules != RECURSE_SUBMODULES_OFF)
> check_for_new_submodule_commits(&ref->new_oid);
> r = s_update_ref("forced-update", ref, 1);
> format_display(display, r ? '!' : '+', quickref.buf,
All of these used to react to any value set to recurse-submodules
that is not off or on (i.e. on-demand, default, none), but now
unless the value is explicitly set to off, we call into the check.
It is not immediately clear how that change is linked to the
retrying. "When set to 'on', we did not check for new commits, but
now we do" can be read from the patch text but not the reasoning
behind it.
What was the reason why we didn't call "check-for-new" when recurse
is set to "on"? Is it because "we are going to recurse anyway, so
there is no need to check to decide if we need to fetch in
submodule"? And the reason why we now need to call when we are set
to recurse anyway is because check-for-new now learns much more than
just "do we need to cd there and run git-fetch? yes/no?"
The answers to these two questions would help readers in the log
message.
> diff --git a/submodule.c b/submodule.c
> index 1e6781504f0..a75146e89cf 100644
> --- a/submodule.c
> +++ b/submodule.c
> @@ -1127,8 +1127,11 @@ struct submodule_parallel_fetch {
> int result;
>
> struct string_list changed_submodule_names;
> + struct string_list retry;
> };
> -#define SPF_INIT {0, ARGV_ARRAY_INIT, NULL, NULL, 0, 0, 0, 0, STRING_LIST_INIT_DUP }
> +#define SPF_INIT {0, ARGV_ARRAY_INIT, NULL, NULL, 0, 0, 0, 0, \
> + STRING_LIST_INIT_DUP, \
> + STRING_LIST_INIT_NODUP}
>
> static void calculate_changed_submodule_paths(
> struct submodule_parallel_fetch *spf)
> @@ -1259,8 +1262,10 @@ static int get_next_submodule(struct child_process *cp,
> {
> int ret = 0;
> struct submodule_parallel_fetch *spf = data;
> + struct string_list_item *it;
>
> for (; spf->count < spf->r->index->cache_nr; spf->count++) {
> + int recurse_config;
> struct strbuf submodule_prefix = STRBUF_INIT;
> const struct cache_entry *ce = spf->r->index->cache[spf->count];
> const char *git_dir, *default_argv;
> @@ -1280,7 +1285,9 @@ static int get_next_submodule(struct child_process *cp,
> }
> }
>
> - switch (get_fetch_recurse_config(submodule, spf))
> + recurse_config = get_fetch_recurse_config(submodule, spf);
> +
> + switch (recurse_config)
> {
> default:
> case RECURSE_SUBMODULES_DEFAULT:
> @@ -1319,9 +1326,50 @@ static int get_next_submodule(struct child_process *cp,
> strbuf_release(&submodule_prefix);
> if (ret) {
> spf->count++;
> + if (submodule != &default_submodule)
> + /* discard const-ness: */
> + *task_cb = (void*)submodule;
> return 1;
> }
> }
> +
> +retry_next:
> +
> + if (spf->retry.nr) {
> + struct strbuf submodule_prefix = STRBUF_INIT;
> + const struct submodule *sub;
> +
> + it = string_list_last(&spf->retry);
> + sub = submodule_from_name(spf->r, &null_oid,
> + it->string);
> +
> + child_process_init(cp);
> + cp->dir = get_submodule_git_dir(spf->r, sub->path);
> + if (!cp->dir) {
> + string_list_pop(&spf->retry, 0);
> + goto retry_next;
> + }
> + prepare_submodule_repo_env_in_gitdir(&cp->env_array);
> + cp->git_cmd = 1;
> +
> + strbuf_addf(&submodule_prefix, "%s%s/", spf->prefix, sub->path);
> + argv_array_init(&cp->args);
> + argv_array_pushv(&cp->args, spf->args.argv);
> + argv_array_push(&cp->args, "on-demand");
> + argv_array_push(&cp->args, "--submodule-prefix");
> + argv_array_push(&cp->args, submodule_prefix.buf);
> +
> + /* NEEDSWORK: have get_default_remote from s--h */
> + argv_array_push(&cp->args, "origin");
> + oid_array_for_each_unique(it->util,
> + append_oid_to_argv, &cp->args);
> +
> + *task_cb = NULL; /* make sure we do not recurse forever */
> + strbuf_release(&submodule_prefix);
> + string_list_pop(&spf->retry, 0);
> + return 1;
> + }
> +
> return 0;
> }
>
> @@ -1335,14 +1383,49 @@ static int fetch_start_failure(struct strbuf *err,
> return 0;
> }
>
> +static int commit_exists_in_sub(const struct object_id *oid, void *data)
> +{
> + struct repository *subrepo = data;
> +
> + enum object_type type = oid_object_info(subrepo, oid, NULL);
> +
> + return type != OBJ_COMMIT;
> +}
Is this checking if the 'oid' exists as a comit in the submodule
repository? It smells to me that it is checking the opposite.
Shouldn't the function be named "commit_missing_from_submodule()" or
something like that?
> static int fetch_finish(int retvalue, struct strbuf *err,
> void *cb, void *task_cb)
> {
> struct submodule_parallel_fetch *spf = cb;
> + struct submodule *sub = task_cb;
> + struct repository subrepo;
>
> if (retvalue)
> spf->result = 1;
>
> + if (!sub)
> + return 0;
> +
> + if (repo_submodule_init(&subrepo, spf->r, sub->path) < 0)
> + warning(_("Could not get submodule repository for submodule '%s' in repository '%s'"),
> + sub->path, spf->r->worktree);
> + else {
> + struct string_list_item *it;
> + struct oid_array *commits;
> +
> + it = string_list_lookup(&spf->changed_submodule_names, sub->name);
> + if (!it)
> + return 0;
OK, so after a fetch, if we have missing commits, we append to the
spf->retry list, which will be looked at in the function we looked
at earlier.
> + commits = it->util;
> + oid_array_filter(commits,
> + commit_exists_in_sub,
> + &subrepo);
> +
> + if (commits->nr)
> + string_list_append(&spf->retry, sub->name)
> + ->util = commits;
> + }
> +
> return 0;
> }
>
> diff --git a/t/t5526-fetch-submodules.sh b/t/t5526-fetch-submodules.sh
> index 42692219a1a..af12c50e7dd 100755
> --- a/t/t5526-fetch-submodules.sh
> +++ b/t/t5526-fetch-submodules.sh
> @@ -605,4 +605,20 @@ test_expect_success "fetch new commits when submodule got renamed" '
> test_cmp expect actual
> '
>
> +test_expect_success "fetch new commits on-demand when they are not reachable" '
> + git checkout --detach &&
> + C=$(git -C submodule commit-tree -m "new change outside refs/heads" HEAD^{tree}) &&
> + git -C submodule update-ref refs/changes/1 $C &&
> + git update-index --cacheinfo 160000 $C submodule &&
> + git commit -m "updated submodule outside of refs/heads" &&
> + D=$(git rev-parse HEAD) &&
> + git update-ref refs/changes/2 $D &&
> + (
> + cd downstream &&
> + git fetch --recurse-submodules --recurse-submodules-default on-demand origin refs/changes/2:refs/heads/my_branch &&
> + git -C submodule cat-file -t $C &&
> + git checkout --recurse-submodules FETCH_HEAD
> + )
> +'
> +
> test_done
^ permalink raw reply [relevance 6%]
* Re: [PATCH 7/9] submodule: fetch in submodules git directory instead of in worktree
2018-09-11 23:49 26% ` [PATCH 7/9] submodule: fetch in submodules git directory instead of in worktree Stefan Beller
@ 2018-09-12 18:36 7% ` Junio C Hamano
2018-09-13 19:29 8% ` Stefan Beller
0 siblings, 1 reply; 200+ results
From: Junio C Hamano @ 2018-09-12 18:36 UTC (permalink / raw)
To: Stefan Beller; +Cc: git
Stefan Beller <sbeller@google.com> writes:
> This patch started as a refactoring to make 'get_next_submodule' more
> readable, but upon doing so, I realized that git-fetch actually doesn't
> need to be run in the worktree. So let's run it in the git dir instead.
It may be clear to the author but not clear to the reader of the
above paragraph that "worktree", "fetch" and "git dir" all refer to
the recursively invoked operation that updates the submodules
repository. s/git-fetch/"git fetch" for the submodule/ should be
sufficient to help the readers.
> That should pave the way towards fetching submodules that are currently
> not checked out.
Very good.
> +static void prepare_submodule_repo_env_in_gitdir(struct argv_array *out)
> +{
> + prepare_submodule_repo_env_no_git_dir(out);
> + argv_array_pushf(out, "%s=.", GIT_DIR_ENVIRONMENT);
> +}
> +
> /* Helper function to display the submodule header line prior to the full
> * summary output. If it can locate the submodule objects directory it will
> * attempt to lookup both the left and right commits and put them into the
> @@ -1227,6 +1233,27 @@ static int get_fetch_recurse_config(const struct submodule *submodule,
> return spf->default_option;
> }
>
> +static const char *get_submodule_git_dir(struct repository *r, const char *path)
> +{
> + struct repository subrepo;
> + const char *ret;
> +
> + if (repo_submodule_init(&subrepo, r, path)) {
> + /* no entry in .gitmodules? */
> + struct strbuf gitdir = STRBUF_INIT;
> + strbuf_repo_worktree_path(&gitdir, r, "%s/.git", path);
> + if (repo_init(&subrepo, gitdir.buf, NULL)) {
> + strbuf_release(&gitdir);
> + return NULL;
> + }
This is for the modern "absorbed" layout? Do we get a notice and
encouragement to migrate from the historical layout, or there is no
need to (e.g. the migration happens automatically in some other
codepaths)?
> + }
> +
> + ret = xstrdup(subrepo.gitdir);
> + repo_clear(&subrepo);
> +
> + return ret;
> +}
Returned value from this function is xstrdup()'ed so the caller
owns, not borrows. There is no need to return "const char *" from
this function. Also the caller needs to free it once done.
> static int get_next_submodule(struct child_process *cp,
> struct strbuf *err, void *data, void **task_cb)
> {
> @@ -1234,8 +1261,6 @@ static int get_next_submodule(struct child_process *cp,
> struct submodule_parallel_fetch *spf = data;
>
> for (; spf->count < spf->r->index->cache_nr; spf->count++) {
> - struct strbuf submodule_path = STRBUF_INIT;
> - struct strbuf submodule_git_dir = STRBUF_INIT;
> struct strbuf submodule_prefix = STRBUF_INIT;
> const struct cache_entry *ce = spf->r->index->cache[spf->count];
> const char *git_dir, *default_argv;
> @@ -1274,16 +1299,12 @@ static int get_next_submodule(struct child_process *cp,
> continue;
> }
>
> - strbuf_repo_worktree_path(&submodule_path, spf->r, "%s", ce->name);
> - strbuf_addf(&submodule_git_dir, "%s/.git", submodule_path.buf);
> strbuf_addf(&submodule_prefix, "%s%s/", spf->prefix, ce->name);
> - git_dir = read_gitfile(submodule_git_dir.buf);
> - if (!git_dir)
> - git_dir = submodule_git_dir.buf;
> - if (is_directory(git_dir)) {
In the old code, git_dir came from read_gitfile() which borrowed.
> + git_dir = get_submodule_git_dir(spf->r, ce->name);
In the new code, we own it, so we'd eventually need to get rid of
it. How does it happen?
> + if (git_dir) {
> child_process_init(cp);
> - cp->dir = strbuf_detach(&submodule_path, NULL);
> - prepare_submodule_repo_env(&cp->env_array);
> + prepare_submodule_repo_env_in_gitdir(&cp->env_array);
> + cp->dir = git_dir;
Does cp now own it and cp->dir gets freed once run_processes_parallel()
is done with this task? Or is cp->dir simply leaking? The old code
gave the result of strbuf_detach(), so even if cp->dir is leaking,
the leak is not new in this patch.
> cp->git_cmd = 1;
> if (!spf->quiet)
> strbuf_addf(err, "Fetching submodule %s%s\n",
> @@ -1295,8 +1316,6 @@ static int get_next_submodule(struct child_process *cp,
> argv_array_push(&cp->args, submodule_prefix.buf);
> ret = 1;
> }
> - strbuf_release(&submodule_path);
> - strbuf_release(&submodule_git_dir);
But if it is a leak, it is easily plugged by freeing git_dir here, I
think.
Thanks.
> strbuf_release(&submodule_prefix);
> if (ret) {
> spf->count++;
> diff --git a/t/t5526-fetch-submodules.sh b/t/t5526-fetch-submodules.sh
> index 6c2f9b2ba26..42692219a1a 100755
> --- a/t/t5526-fetch-submodules.sh
> +++ b/t/t5526-fetch-submodules.sh
> @@ -566,7 +566,12 @@ test_expect_success 'fetching submodule into a broken repository' '
>
> test_must_fail git -C dst status &&
> test_must_fail git -C dst diff &&
> - test_must_fail git -C dst fetch --recurse-submodules
> +
> + # git-fetch cannot find the git directory of the submodule,
> + # so it will do nothing, successfully, as it cannot distinguish between
> + # this broken submodule and a submodule that was just set active but
> + # not cloned yet
> + git -C dst fetch --recurse-submodules
> '
>
> test_expect_success "fetch new commits when submodule got renamed" '
^ permalink raw reply [relevance 7%]
* Re: [PATCH 4/9] submodule.c: sort changed_submodule_names before searching it
2018-09-11 23:49 17% ` [PATCH 4/9] submodule.c: sort changed_submodule_names before searching it Stefan Beller
@ 2018-09-12 18:18 6% ` Junio C Hamano
2018-09-12 19:06 4% ` Stefan Beller
0 siblings, 1 reply; 200+ results
From: Junio C Hamano @ 2018-09-12 18:18 UTC (permalink / raw)
To: Stefan Beller; +Cc: git
Stefan Beller <sbeller@google.com> writes:
> We can string_list_insert() to maintain sorted-ness of the
> list as we find new items, or we can string_list_append() to
> build an unsorted list and sort it at the end just once.
>
> To pick which one is more appropriate, we notice the fact
> that we discover new items more or less in the already
> sorted order. That makes "append then sort" more
> appropriate.
Sorry, but I still do not get the math you are implying in the
second paragraph. Are you saying that append-then-sort is efficient
when items being appended is already sorted? That depends on the
sorting algorithm used, so the logic is incomplete unless you say
"given that we use X for sorting,...", I think.
Do we really discover new items in sorted order, by the way? In a
single diff invocation made inside collect_changed_submodules() for
one commit in the superproject's history, we will grab changed paths
in the pathname order (i.e. sorted); if the superproject's tip commit
touches the submodules at paths A and Z, we will discover these two
paths in sorted order.
But because we are walking the superproject's history to collect all
paths that have been affected in that function, and repeatedly
calling diff as we discover commit in the superproject's history, I
am not sure how well the resulting set of paths would be sorted.
The tip commit in superproject's history may have modified the
submodule at path X, the parent of that commit may have touched the
submodule at path M, and its parent may have touched the submodule
at path A. Don't we end up grabbing these paths in that discoverd
order, i.e. X, M and A?
I still think changing it from "insert as we find an item, keeping
the list sorted" to "append all and then sort before we start
looking things up from the result" makes sense, but I do not think
the "we find things in sorted order" is either true, or it would
affect the choice between the two. A justification to choose the
latter I can think of that makes sense is that we don't have to pay
cost to keep the list sorted while building it because we do not do
any look-up while building the list.
> Signed-off-by: Stefan Beller <sbeller@google.com>
> ---
> submodule.c | 3 ++-
> 1 file changed, 2 insertions(+), 1 deletion(-)
>
> diff --git a/submodule.c b/submodule.c
> index d29dfa3d1f5..c6eff7699f3 100644
> --- a/submodule.c
> +++ b/submodule.c
> @@ -1256,7 +1256,7 @@ static int get_next_submodule(struct child_process *cp,
> case RECURSE_SUBMODULES_DEFAULT:
> case RECURSE_SUBMODULES_ON_DEMAND:
> if (!submodule ||
> - !unsorted_string_list_lookup(
> + !string_list_lookup(
> &changed_submodule_names,
> submodule->name))
> continue;
> @@ -1350,6 +1350,7 @@ int fetch_populated_submodules(struct repository *r,
> /* default value, "--submodule-prefix" and its value are added later */
>
> calculate_changed_submodule_paths();
> + string_list_sort(&changed_submodule_names);
> run_processes_parallel(max_parallel_jobs,
> get_next_submodule,
> fetch_start_failure,
^ permalink raw reply [relevance 6%]
* Re: [PATCH 3/9] submodule.c: fix indentation
2018-09-11 23:49 24% ` [PATCH 3/9] submodule.c: fix indentation Stefan Beller
@ 2018-09-12 18:02 4% ` Junio C Hamano
0 siblings, 0 replies; 200+ results
From: Junio C Hamano @ 2018-09-12 18:02 UTC (permalink / raw)
To: Stefan Beller; +Cc: git
Stefan Beller <sbeller@google.com> writes:
> The submodule subsystem is really bad at staying within 80 characters.
> Fix it while we are here.
Makes sense. Thanks.
>
> Signed-off-by: Stefan Beller <sbeller@google.com>
> ---
> submodule.c | 9 ++++++---
> 1 file changed, 6 insertions(+), 3 deletions(-)
>
> diff --git a/submodule.c b/submodule.c
> index a2b266fbfae..d29dfa3d1f5 100644
> --- a/submodule.c
> +++ b/submodule.c
> @@ -1244,7 +1244,8 @@ static int get_next_submodule(struct child_process *cp,
> if (!submodule) {
> const char *name = default_name_or_path(ce->name);
> if (name) {
> - default_submodule.path = default_submodule.name = name;
> + default_submodule.path = name;
> + default_submodule.name = name;
> submodule = &default_submodule;
> }
> }
> @@ -1254,8 +1255,10 @@ static int get_next_submodule(struct child_process *cp,
> default:
> case RECURSE_SUBMODULES_DEFAULT:
> case RECURSE_SUBMODULES_ON_DEMAND:
> - if (!submodule || !unsorted_string_list_lookup(&changed_submodule_names,
> - submodule->name))
> + if (!submodule ||
> + !unsorted_string_list_lookup(
> + &changed_submodule_names,
> + submodule->name))
> continue;
> default_argv = "on-demand";
> break;
^ permalink raw reply [relevance 4%]
* Re: [PATCH 6/9] submodule.c: do not copy around submodule list
2018-09-11 23:49 17% ` [PATCH 6/9] submodule.c: do not copy around submodule list Stefan Beller
@ 2018-09-12 2:25 5% ` Ramsay Jones
0 siblings, 0 replies; 200+ results
From: Ramsay Jones @ 2018-09-12 2:25 UTC (permalink / raw)
To: Stefan Beller, git
On 12/09/18 00:49, Stefan Beller wrote:
> 'calculate_changed_submodule_paths' uses a local list to compute the
> changed submodules, and then produces the result by copying appropriate
> items into the result list.
>
> Instead use the result list directly and prune items afterwards
> using string_list_remove_empty_items.
>
> By doin so we'll have access to the util pointer for longer that
s/doin/doing/
ATB,
Ramsay Jones
^ permalink raw reply [relevance 5%]
* [PATCH 9/9] builtin/fetch: check for submodule updates for non branch fetches
2018-09-11 23:49 10% [PATCH 0/9] fetch: make sure submodule oids are fetched Stefan Beller
` (5 preceding siblings ...)
2018-09-11 23:49 18% ` [PATCH 8/9] fetch: retry fetching submodules if sha1 were not fetched Stefan Beller
@ 2018-09-11 23:49 23% ` Stefan Beller
2018-09-12 19:20 6% ` Junio C Hamano
2018-09-17 21:35 9% ` [PATCHv2 0/9] fetch: make sure submodule oids are fetched Stefan Beller
7 siblings, 1 reply; 200+ results
From: Stefan Beller @ 2018-09-11 23:49 UTC (permalink / raw)
To: git; +Cc: Stefan Beller
For Gerrit users that use submodules the invocation of fetch without a
branch is their main use case.
Signed-off-by: Stefan Beller <sbeller@google.com>
---
builtin/fetch.c | 5 ++++-
t/t5526-fetch-submodules.sh | 2 +-
2 files changed, 5 insertions(+), 2 deletions(-)
diff --git a/builtin/fetch.c b/builtin/fetch.c
index 95c44bf6ffa..ea6ecd123e7 100644
--- a/builtin/fetch.c
+++ b/builtin/fetch.c
@@ -887,11 +887,14 @@ static int store_updated_refs(const char *raw_url, const char *remote_name,
rc |= update_local_ref(ref, what, rm, ¬e,
summary_width);
free(ref);
- } else
+ } else {
+ check_for_new_submodule_commits(&rm->old_oid);
format_display(¬e, '*',
*kind ? kind : "branch", NULL,
*what ? what : "HEAD",
"FETCH_HEAD", summary_width);
+ }
+
if (note.len) {
if (verbosity >= 0 && !shown_url) {
fprintf(stderr, _("From %.*s\n"),
diff --git a/t/t5526-fetch-submodules.sh b/t/t5526-fetch-submodules.sh
index af12c50e7dd..a509eabb044 100755
--- a/t/t5526-fetch-submodules.sh
+++ b/t/t5526-fetch-submodules.sh
@@ -615,7 +615,7 @@ test_expect_success "fetch new commits on-demand when they are not reachable" '
git update-ref refs/changes/2 $D &&
(
cd downstream &&
- git fetch --recurse-submodules --recurse-submodules-default on-demand origin refs/changes/2:refs/heads/my_branch &&
+ git fetch --recurse-submodules origin refs/changes/2 &&
git -C submodule cat-file -t $C &&
git checkout --recurse-submodules FETCH_HEAD
)
--
2.19.0.397.gdd90340f6a-goog
^ permalink raw reply related [relevance 23%]
* [PATCH 8/9] fetch: retry fetching submodules if sha1 were not fetched
2018-09-11 23:49 10% [PATCH 0/9] fetch: make sure submodule oids are fetched Stefan Beller
` (4 preceding siblings ...)
2018-09-11 23:49 26% ` [PATCH 7/9] submodule: fetch in submodules git directory instead of in worktree Stefan Beller
@ 2018-09-11 23:49 18% ` Stefan Beller
2018-09-12 19:03 6% ` Junio C Hamano
2018-09-11 23:49 23% ` [PATCH 9/9] builtin/fetch: check for submodule updates for non branch fetches Stefan Beller
2018-09-17 21:35 9% ` [PATCHv2 0/9] fetch: make sure submodule oids are fetched Stefan Beller
7 siblings, 1 reply; 200+ results
From: Stefan Beller @ 2018-09-11 23:49 UTC (permalink / raw)
To: git; +Cc: Stefan Beller
Currently when git-fetch is asked to recurse into submodules, it dispatches
a plain "git-fetch -C <submodule-dir>" (and some submodule related options
such as prefix and recusing strategy, but) without any information of the
remote or the tip that should be fetched.
This works surprisingly well in some workflows (such as using submodules
as a third party library), while not so well in other scenarios, such
as in a Gerrit topic-based workflow, that can tie together changes
(potentially across repositories) on the server side. One of the parts
of such a Gerrit workflow is to download a change when wanting to examine
it, and you'd want to have its submodule changes that are in the same
topic downloaded as well. However these submodule changes reside in their
own repository in their own ref (refs/changes/<int>).
Retry fetching a submodule if the object id that the superproject points
to, cannot be found.
This doesn't support fetching to FETCH_HEAD yet, but only into a local
branch. To make fetching into FETCH_HEAD work, we need some refactoring
in builtin/fetch.c to adjust the calls to 'check_for_new_submodule_commits'
that is coming in the next patch.
Signed-off-by: Stefan Beller <sbeller@google.com>
---
builtin/fetch.c | 9 ++--
submodule.c | 87 ++++++++++++++++++++++++++++++++++++-
t/t5526-fetch-submodules.sh | 16 +++++++
3 files changed, 104 insertions(+), 8 deletions(-)
diff --git a/builtin/fetch.c b/builtin/fetch.c
index 61bec5d213d..95c44bf6ffa 100644
--- a/builtin/fetch.c
+++ b/builtin/fetch.c
@@ -700,8 +700,7 @@ static int update_local_ref(struct ref *ref,
what = _("[new ref]");
}
- if ((recurse_submodules != RECURSE_SUBMODULES_OFF) &&
- (recurse_submodules != RECURSE_SUBMODULES_ON))
+ if (recurse_submodules != RECURSE_SUBMODULES_OFF)
check_for_new_submodule_commits(&ref->new_oid);
r = s_update_ref(msg, ref, 0);
format_display(display, r ? '!' : '*', what,
@@ -716,8 +715,7 @@ static int update_local_ref(struct ref *ref,
strbuf_add_unique_abbrev(&quickref, ¤t->object.oid, DEFAULT_ABBREV);
strbuf_addstr(&quickref, "..");
strbuf_add_unique_abbrev(&quickref, &ref->new_oid, DEFAULT_ABBREV);
- if ((recurse_submodules != RECURSE_SUBMODULES_OFF) &&
- (recurse_submodules != RECURSE_SUBMODULES_ON))
+ if (recurse_submodules != RECURSE_SUBMODULES_OFF)
check_for_new_submodule_commits(&ref->new_oid);
r = s_update_ref("fast-forward", ref, 1);
format_display(display, r ? '!' : ' ', quickref.buf,
@@ -731,8 +729,7 @@ static int update_local_ref(struct ref *ref,
strbuf_add_unique_abbrev(&quickref, ¤t->object.oid, DEFAULT_ABBREV);
strbuf_addstr(&quickref, "...");
strbuf_add_unique_abbrev(&quickref, &ref->new_oid, DEFAULT_ABBREV);
- if ((recurse_submodules != RECURSE_SUBMODULES_OFF) &&
- (recurse_submodules != RECURSE_SUBMODULES_ON))
+ if (recurse_submodules != RECURSE_SUBMODULES_OFF)
check_for_new_submodule_commits(&ref->new_oid);
r = s_update_ref("forced-update", ref, 1);
format_display(display, r ? '!' : '+', quickref.buf,
diff --git a/submodule.c b/submodule.c
index 1e6781504f0..a75146e89cf 100644
--- a/submodule.c
+++ b/submodule.c
@@ -1127,8 +1127,11 @@ struct submodule_parallel_fetch {
int result;
struct string_list changed_submodule_names;
+ struct string_list retry;
};
-#define SPF_INIT {0, ARGV_ARRAY_INIT, NULL, NULL, 0, 0, 0, 0, STRING_LIST_INIT_DUP }
+#define SPF_INIT {0, ARGV_ARRAY_INIT, NULL, NULL, 0, 0, 0, 0, \
+ STRING_LIST_INIT_DUP, \
+ STRING_LIST_INIT_NODUP}
static void calculate_changed_submodule_paths(
struct submodule_parallel_fetch *spf)
@@ -1259,8 +1262,10 @@ static int get_next_submodule(struct child_process *cp,
{
int ret = 0;
struct submodule_parallel_fetch *spf = data;
+ struct string_list_item *it;
for (; spf->count < spf->r->index->cache_nr; spf->count++) {
+ int recurse_config;
struct strbuf submodule_prefix = STRBUF_INIT;
const struct cache_entry *ce = spf->r->index->cache[spf->count];
const char *git_dir, *default_argv;
@@ -1280,7 +1285,9 @@ static int get_next_submodule(struct child_process *cp,
}
}
- switch (get_fetch_recurse_config(submodule, spf))
+ recurse_config = get_fetch_recurse_config(submodule, spf);
+
+ switch (recurse_config)
{
default:
case RECURSE_SUBMODULES_DEFAULT:
@@ -1319,9 +1326,50 @@ static int get_next_submodule(struct child_process *cp,
strbuf_release(&submodule_prefix);
if (ret) {
spf->count++;
+ if (submodule != &default_submodule)
+ /* discard const-ness: */
+ *task_cb = (void*)submodule;
return 1;
}
}
+
+retry_next:
+
+ if (spf->retry.nr) {
+ struct strbuf submodule_prefix = STRBUF_INIT;
+ const struct submodule *sub;
+
+ it = string_list_last(&spf->retry);
+ sub = submodule_from_name(spf->r, &null_oid,
+ it->string);
+
+ child_process_init(cp);
+ cp->dir = get_submodule_git_dir(spf->r, sub->path);
+ if (!cp->dir) {
+ string_list_pop(&spf->retry, 0);
+ goto retry_next;
+ }
+ prepare_submodule_repo_env_in_gitdir(&cp->env_array);
+ cp->git_cmd = 1;
+
+ strbuf_addf(&submodule_prefix, "%s%s/", spf->prefix, sub->path);
+ argv_array_init(&cp->args);
+ argv_array_pushv(&cp->args, spf->args.argv);
+ argv_array_push(&cp->args, "on-demand");
+ argv_array_push(&cp->args, "--submodule-prefix");
+ argv_array_push(&cp->args, submodule_prefix.buf);
+
+ /* NEEDSWORK: have get_default_remote from s--h */
+ argv_array_push(&cp->args, "origin");
+ oid_array_for_each_unique(it->util,
+ append_oid_to_argv, &cp->args);
+
+ *task_cb = NULL; /* make sure we do not recurse forever */
+ strbuf_release(&submodule_prefix);
+ string_list_pop(&spf->retry, 0);
+ return 1;
+ }
+
return 0;
}
@@ -1335,14 +1383,49 @@ static int fetch_start_failure(struct strbuf *err,
return 0;
}
+static int commit_exists_in_sub(const struct object_id *oid, void *data)
+{
+ struct repository *subrepo = data;
+
+ enum object_type type = oid_object_info(subrepo, oid, NULL);
+
+ return type != OBJ_COMMIT;
+}
+
static int fetch_finish(int retvalue, struct strbuf *err,
void *cb, void *task_cb)
{
struct submodule_parallel_fetch *spf = cb;
+ struct submodule *sub = task_cb;
+ struct repository subrepo;
if (retvalue)
spf->result = 1;
+ if (!sub)
+ return 0;
+
+ if (repo_submodule_init(&subrepo, spf->r, sub->path) < 0)
+ warning(_("Could not get submodule repository for submodule '%s' in repository '%s'"),
+ sub->path, spf->r->worktree);
+ else {
+ struct string_list_item *it;
+ struct oid_array *commits;
+
+ it = string_list_lookup(&spf->changed_submodule_names, sub->name);
+ if (!it)
+ return 0;
+
+ commits = it->util;
+ oid_array_filter(commits,
+ commit_exists_in_sub,
+ &subrepo);
+
+ if (commits->nr)
+ string_list_append(&spf->retry, sub->name)
+ ->util = commits;
+ }
+
return 0;
}
diff --git a/t/t5526-fetch-submodules.sh b/t/t5526-fetch-submodules.sh
index 42692219a1a..af12c50e7dd 100755
--- a/t/t5526-fetch-submodules.sh
+++ b/t/t5526-fetch-submodules.sh
@@ -605,4 +605,20 @@ test_expect_success "fetch new commits when submodule got renamed" '
test_cmp expect actual
'
+test_expect_success "fetch new commits on-demand when they are not reachable" '
+ git checkout --detach &&
+ C=$(git -C submodule commit-tree -m "new change outside refs/heads" HEAD^{tree}) &&
+ git -C submodule update-ref refs/changes/1 $C &&
+ git update-index --cacheinfo 160000 $C submodule &&
+ git commit -m "updated submodule outside of refs/heads" &&
+ D=$(git rev-parse HEAD) &&
+ git update-ref refs/changes/2 $D &&
+ (
+ cd downstream &&
+ git fetch --recurse-submodules --recurse-submodules-default on-demand origin refs/changes/2:refs/heads/my_branch &&
+ git -C submodule cat-file -t $C &&
+ git checkout --recurse-submodules FETCH_HEAD
+ )
+'
+
test_done
--
2.19.0.397.gdd90340f6a-goog
^ permalink raw reply related [relevance 18%]
* [PATCH 7/9] submodule: fetch in submodules git directory instead of in worktree
2018-09-11 23:49 10% [PATCH 0/9] fetch: make sure submodule oids are fetched Stefan Beller
` (3 preceding siblings ...)
2018-09-11 23:49 17% ` [PATCH 6/9] submodule.c: do not copy around submodule list Stefan Beller
@ 2018-09-11 23:49 26% ` Stefan Beller
2018-09-12 18:36 7% ` Junio C Hamano
2018-09-11 23:49 18% ` [PATCH 8/9] fetch: retry fetching submodules if sha1 were not fetched Stefan Beller
` (2 subsequent siblings)
7 siblings, 1 reply; 200+ results
From: Stefan Beller @ 2018-09-11 23:49 UTC (permalink / raw)
To: git; +Cc: Stefan Beller
This patch started as a refactoring to make 'get_next_submodule' more
readable, but upon doing so, I realized that git-fetch actually doesn't
need to be run in the worktree. So let's run it in the git dir instead.
That should pave the way towards fetching submodules that are currently
not checked out.
Signed-off-by: Stefan Beller <sbeller@google.com>
---
submodule.c | 43 ++++++++++++++++++++++++++-----------
t/t5526-fetch-submodules.sh | 7 +++++-
2 files changed, 37 insertions(+), 13 deletions(-)
diff --git a/submodule.c b/submodule.c
index 00a9a3c6b12..1e6781504f0 100644
--- a/submodule.c
+++ b/submodule.c
@@ -481,6 +481,12 @@ void prepare_submodule_repo_env(struct argv_array *out)
DEFAULT_GIT_DIR_ENVIRONMENT);
}
+static void prepare_submodule_repo_env_in_gitdir(struct argv_array *out)
+{
+ prepare_submodule_repo_env_no_git_dir(out);
+ argv_array_pushf(out, "%s=.", GIT_DIR_ENVIRONMENT);
+}
+
/* Helper function to display the submodule header line prior to the full
* summary output. If it can locate the submodule objects directory it will
* attempt to lookup both the left and right commits and put them into the
@@ -1227,6 +1233,27 @@ static int get_fetch_recurse_config(const struct submodule *submodule,
return spf->default_option;
}
+static const char *get_submodule_git_dir(struct repository *r, const char *path)
+{
+ struct repository subrepo;
+ const char *ret;
+
+ if (repo_submodule_init(&subrepo, r, path)) {
+ /* no entry in .gitmodules? */
+ struct strbuf gitdir = STRBUF_INIT;
+ strbuf_repo_worktree_path(&gitdir, r, "%s/.git", path);
+ if (repo_init(&subrepo, gitdir.buf, NULL)) {
+ strbuf_release(&gitdir);
+ return NULL;
+ }
+ }
+
+ ret = xstrdup(subrepo.gitdir);
+ repo_clear(&subrepo);
+
+ return ret;
+}
+
static int get_next_submodule(struct child_process *cp,
struct strbuf *err, void *data, void **task_cb)
{
@@ -1234,8 +1261,6 @@ static int get_next_submodule(struct child_process *cp,
struct submodule_parallel_fetch *spf = data;
for (; spf->count < spf->r->index->cache_nr; spf->count++) {
- struct strbuf submodule_path = STRBUF_INIT;
- struct strbuf submodule_git_dir = STRBUF_INIT;
struct strbuf submodule_prefix = STRBUF_INIT;
const struct cache_entry *ce = spf->r->index->cache[spf->count];
const char *git_dir, *default_argv;
@@ -1274,16 +1299,12 @@ static int get_next_submodule(struct child_process *cp,
continue;
}
- strbuf_repo_worktree_path(&submodule_path, spf->r, "%s", ce->name);
- strbuf_addf(&submodule_git_dir, "%s/.git", submodule_path.buf);
strbuf_addf(&submodule_prefix, "%s%s/", spf->prefix, ce->name);
- git_dir = read_gitfile(submodule_git_dir.buf);
- if (!git_dir)
- git_dir = submodule_git_dir.buf;
- if (is_directory(git_dir)) {
+ git_dir = get_submodule_git_dir(spf->r, ce->name);
+ if (git_dir) {
child_process_init(cp);
- cp->dir = strbuf_detach(&submodule_path, NULL);
- prepare_submodule_repo_env(&cp->env_array);
+ prepare_submodule_repo_env_in_gitdir(&cp->env_array);
+ cp->dir = git_dir;
cp->git_cmd = 1;
if (!spf->quiet)
strbuf_addf(err, "Fetching submodule %s%s\n",
@@ -1295,8 +1316,6 @@ static int get_next_submodule(struct child_process *cp,
argv_array_push(&cp->args, submodule_prefix.buf);
ret = 1;
}
- strbuf_release(&submodule_path);
- strbuf_release(&submodule_git_dir);
strbuf_release(&submodule_prefix);
if (ret) {
spf->count++;
diff --git a/t/t5526-fetch-submodules.sh b/t/t5526-fetch-submodules.sh
index 6c2f9b2ba26..42692219a1a 100755
--- a/t/t5526-fetch-submodules.sh
+++ b/t/t5526-fetch-submodules.sh
@@ -566,7 +566,12 @@ test_expect_success 'fetching submodule into a broken repository' '
test_must_fail git -C dst status &&
test_must_fail git -C dst diff &&
- test_must_fail git -C dst fetch --recurse-submodules
+
+ # git-fetch cannot find the git directory of the submodule,
+ # so it will do nothing, successfully, as it cannot distinguish between
+ # this broken submodule and a submodule that was just set active but
+ # not cloned yet
+ git -C dst fetch --recurse-submodules
'
test_expect_success "fetch new commits when submodule got renamed" '
--
2.19.0.397.gdd90340f6a-goog
^ permalink raw reply related [relevance 26%]
* [PATCH 6/9] submodule.c: do not copy around submodule list
2018-09-11 23:49 10% [PATCH 0/9] fetch: make sure submodule oids are fetched Stefan Beller
` (2 preceding siblings ...)
2018-09-11 23:49 16% ` [PATCH 5/9] submodule: move global changed_submodule_names into fetch submodule struct Stefan Beller
@ 2018-09-11 23:49 17% ` Stefan Beller
2018-09-12 2:25 5% ` Ramsay Jones
2018-09-11 23:49 26% ` [PATCH 7/9] submodule: fetch in submodules git directory instead of in worktree Stefan Beller
` (3 subsequent siblings)
7 siblings, 1 reply; 200+ results
From: Stefan Beller @ 2018-09-11 23:49 UTC (permalink / raw)
To: git; +Cc: Stefan Beller
'calculate_changed_submodule_paths' uses a local list to compute the
changed submodules, and then produces the result by copying appropriate
items into the result list.
Instead use the result list directly and prune items afterwards
using string_list_remove_empty_items.
By doin so we'll have access to the util pointer for longer that
contains the commits that we need to fetch, which will be
useful in a later patch.
Signed-off-by: Stefan Beller <sbeller@google.com>
---
submodule.c | 19 ++++++++++---------
1 file changed, 10 insertions(+), 9 deletions(-)
diff --git a/submodule.c b/submodule.c
index 3520dd76bdf..00a9a3c6b12 100644
--- a/submodule.c
+++ b/submodule.c
@@ -1128,8 +1128,7 @@ static void calculate_changed_submodule_paths(
struct submodule_parallel_fetch *spf)
{
struct argv_array argv = ARGV_ARRAY_INIT;
- struct string_list changed_submodules = STRING_LIST_INIT_DUP;
- const struct string_list_item *name;
+ struct string_list_item *name;
/* No need to check if there are no submodules configured */
if (!submodule_from_path(the_repository, NULL, NULL))
@@ -1146,9 +1145,9 @@ static void calculate_changed_submodule_paths(
* Collect all submodules (whether checked out or not) for which new
* commits have been recorded upstream in "changed_submodule_names".
*/
- collect_changed_submodules(&changed_submodules, &argv);
+ collect_changed_submodules(&spf->changed_submodule_names, &argv);
- for_each_string_list_item(name, &changed_submodules) {
+ for_each_string_list_item(name, &spf->changed_submodule_names) {
struct oid_array *commits = name->util;
const struct submodule *submodule;
const char *path = NULL;
@@ -1162,12 +1161,14 @@ static void calculate_changed_submodule_paths(
if (!path)
continue;
- if (!submodule_has_commits(path, commits))
- string_list_append(&spf->changed_submodule_names,
- name->string);
+ if (submodule_has_commits(path, commits)) {
+ oid_array_clear(commits);
+ *name->string = '\0';
+ }
}
- free_submodules_oids(&changed_submodules);
+ string_list_remove_empty_items(&spf->changed_submodule_names, 1);
+
argv_array_clear(&argv);
oid_array_clear(&ref_tips_before_fetch);
oid_array_clear(&ref_tips_after_fetch);
@@ -1363,7 +1364,7 @@ int fetch_populated_submodules(struct repository *r,
argv_array_clear(&spf.args);
out:
- string_list_clear(&spf.changed_submodule_names, 1);
+ free_submodules_oids(&spf.changed_submodule_names);
return spf.result;
}
--
2.19.0.397.gdd90340f6a-goog
^ permalink raw reply related [relevance 17%]
* [PATCH 5/9] submodule: move global changed_submodule_names into fetch submodule struct
2018-09-11 23:49 10% [PATCH 0/9] fetch: make sure submodule oids are fetched Stefan Beller
2018-09-11 23:49 24% ` [PATCH 3/9] submodule.c: fix indentation Stefan Beller
2018-09-11 23:49 17% ` [PATCH 4/9] submodule.c: sort changed_submodule_names before searching it Stefan Beller
@ 2018-09-11 23:49 16% ` Stefan Beller
2018-09-11 23:49 17% ` [PATCH 6/9] submodule.c: do not copy around submodule list Stefan Beller
` (4 subsequent siblings)
7 siblings, 0 replies; 200+ results
From: Stefan Beller @ 2018-09-11 23:49 UTC (permalink / raw)
To: git; +Cc: Stefan Beller
The `changed_submodule_names` are only used for fetching, so let's make it
part of the struct that is passed around for fetching submodules.
Signed-off-by: Stefan Beller <sbeller@google.com>
---
submodule.c | 42 +++++++++++++++++++++++-------------------
1 file changed, 23 insertions(+), 19 deletions(-)
diff --git a/submodule.c b/submodule.c
index c6eff7699f3..3520dd76bdf 100644
--- a/submodule.c
+++ b/submodule.c
@@ -24,7 +24,7 @@
#include "object-store.h"
static int config_update_recurse_submodules = RECURSE_SUBMODULES_OFF;
-static struct string_list changed_submodule_names = STRING_LIST_INIT_DUP;
+
static int initialized_fetch_ref_tips;
static struct oid_array ref_tips_before_fetch;
static struct oid_array ref_tips_after_fetch;
@@ -1110,7 +1110,22 @@ void check_for_new_submodule_commits(struct object_id *oid)
oid_array_append(&ref_tips_after_fetch, oid);
}
-static void calculate_changed_submodule_paths(void)
+struct submodule_parallel_fetch {
+ int count;
+ struct argv_array args;
+ struct repository *r;
+ const char *prefix;
+ int command_line_option;
+ int default_option;
+ int quiet;
+ int result;
+
+ struct string_list changed_submodule_names;
+};
+#define SPF_INIT {0, ARGV_ARRAY_INIT, NULL, NULL, 0, 0, 0, 0, STRING_LIST_INIT_DUP }
+
+static void calculate_changed_submodule_paths(
+ struct submodule_parallel_fetch *spf)
{
struct argv_array argv = ARGV_ARRAY_INIT;
struct string_list changed_submodules = STRING_LIST_INIT_DUP;
@@ -1148,7 +1163,8 @@ static void calculate_changed_submodule_paths(void)
continue;
if (!submodule_has_commits(path, commits))
- string_list_append(&changed_submodule_names, name->string);
+ string_list_append(&spf->changed_submodule_names,
+ name->string);
}
free_submodules_oids(&changed_submodules);
@@ -1185,18 +1201,6 @@ int submodule_touches_in_range(struct object_id *excl_oid,
return ret;
}
-struct submodule_parallel_fetch {
- int count;
- struct argv_array args;
- struct repository *r;
- const char *prefix;
- int command_line_option;
- int default_option;
- int quiet;
- int result;
-};
-#define SPF_INIT {0, ARGV_ARRAY_INIT, NULL, NULL, 0, 0, 0, 0}
-
static int get_fetch_recurse_config(const struct submodule *submodule,
struct submodule_parallel_fetch *spf)
{
@@ -1257,7 +1261,7 @@ static int get_next_submodule(struct child_process *cp,
case RECURSE_SUBMODULES_ON_DEMAND:
if (!submodule ||
!string_list_lookup(
- &changed_submodule_names,
+ &spf->changed_submodule_names,
submodule->name))
continue;
default_argv = "on-demand";
@@ -1349,8 +1353,8 @@ int fetch_populated_submodules(struct repository *r,
argv_array_push(&spf.args, "--recurse-submodules-default");
/* default value, "--submodule-prefix" and its value are added later */
- calculate_changed_submodule_paths();
- string_list_sort(&changed_submodule_names);
+ calculate_changed_submodule_paths(&spf);
+ string_list_sort(&spf.changed_submodule_names);
run_processes_parallel(max_parallel_jobs,
get_next_submodule,
fetch_start_failure,
@@ -1359,7 +1363,7 @@ int fetch_populated_submodules(struct repository *r,
argv_array_clear(&spf.args);
out:
- string_list_clear(&changed_submodule_names, 1);
+ string_list_clear(&spf.changed_submodule_names, 1);
return spf.result;
}
--
2.19.0.397.gdd90340f6a-goog
^ permalink raw reply related [relevance 16%]
* [PATCH 4/9] submodule.c: sort changed_submodule_names before searching it
2018-09-11 23:49 10% [PATCH 0/9] fetch: make sure submodule oids are fetched Stefan Beller
2018-09-11 23:49 24% ` [PATCH 3/9] submodule.c: fix indentation Stefan Beller
@ 2018-09-11 23:49 17% ` Stefan Beller
2018-09-12 18:18 6% ` Junio C Hamano
2018-09-11 23:49 16% ` [PATCH 5/9] submodule: move global changed_submodule_names into fetch submodule struct Stefan Beller
` (5 subsequent siblings)
7 siblings, 1 reply; 200+ results
From: Stefan Beller @ 2018-09-11 23:49 UTC (permalink / raw)
To: git; +Cc: Stefan Beller
We can string_list_insert() to maintain sorted-ness of the
list as we find new items, or we can string_list_append() to
build an unsorted list and sort it at the end just once.
To pick which one is more appropriate, we notice the fact
that we discover new items more or less in the already
sorted order. That makes "append then sort" more
appropriate.
Signed-off-by: Stefan Beller <sbeller@google.com>
---
submodule.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/submodule.c b/submodule.c
index d29dfa3d1f5..c6eff7699f3 100644
--- a/submodule.c
+++ b/submodule.c
@@ -1256,7 +1256,7 @@ static int get_next_submodule(struct child_process *cp,
case RECURSE_SUBMODULES_DEFAULT:
case RECURSE_SUBMODULES_ON_DEMAND:
if (!submodule ||
- !unsorted_string_list_lookup(
+ !string_list_lookup(
&changed_submodule_names,
submodule->name))
continue;
@@ -1350,6 +1350,7 @@ int fetch_populated_submodules(struct repository *r,
/* default value, "--submodule-prefix" and its value are added later */
calculate_changed_submodule_paths();
+ string_list_sort(&changed_submodule_names);
run_processes_parallel(max_parallel_jobs,
get_next_submodule,
fetch_start_failure,
--
2.19.0.397.gdd90340f6a-goog
^ permalink raw reply related [relevance 17%]
* [PATCH 3/9] submodule.c: fix indentation
2018-09-11 23:49 10% [PATCH 0/9] fetch: make sure submodule oids are fetched Stefan Beller
@ 2018-09-11 23:49 24% ` Stefan Beller
2018-09-12 18:02 4% ` Junio C Hamano
2018-09-11 23:49 17% ` [PATCH 4/9] submodule.c: sort changed_submodule_names before searching it Stefan Beller
` (6 subsequent siblings)
7 siblings, 1 reply; 200+ results
From: Stefan Beller @ 2018-09-11 23:49 UTC (permalink / raw)
To: git; +Cc: Stefan Beller
The submodule subsystem is really bad at staying within 80 characters.
Fix it while we are here.
Signed-off-by: Stefan Beller <sbeller@google.com>
---
submodule.c | 9 ++++++---
1 file changed, 6 insertions(+), 3 deletions(-)
diff --git a/submodule.c b/submodule.c
index a2b266fbfae..d29dfa3d1f5 100644
--- a/submodule.c
+++ b/submodule.c
@@ -1244,7 +1244,8 @@ static int get_next_submodule(struct child_process *cp,
if (!submodule) {
const char *name = default_name_or_path(ce->name);
if (name) {
- default_submodule.path = default_submodule.name = name;
+ default_submodule.path = name;
+ default_submodule.name = name;
submodule = &default_submodule;
}
}
@@ -1254,8 +1255,10 @@ static int get_next_submodule(struct child_process *cp,
default:
case RECURSE_SUBMODULES_DEFAULT:
case RECURSE_SUBMODULES_ON_DEMAND:
- if (!submodule || !unsorted_string_list_lookup(&changed_submodule_names,
- submodule->name))
+ if (!submodule ||
+ !unsorted_string_list_lookup(
+ &changed_submodule_names,
+ submodule->name))
continue;
default_argv = "on-demand";
break;
--
2.19.0.397.gdd90340f6a-goog
^ permalink raw reply related [relevance 24%]
* [PATCH 0/9] fetch: make sure submodule oids are fetched
@ 2018-09-11 23:49 10% Stefan Beller
2018-09-11 23:49 24% ` [PATCH 3/9] submodule.c: fix indentation Stefan Beller
` (7 more replies)
0 siblings, 8 replies; 200+ results
From: Stefan Beller @ 2018-09-11 23:49 UTC (permalink / raw)
To: git; +Cc: Stefan Beller
Currently when git-fetch is asked to recurse into submodules, it dispatches
a plain "git-fetch -C <submodule-dir>" (and some submodule related options
such as prefix and recusing strategy, but) without any information of the
remote or the tip that should be fetched.
This works surprisingly well in some workflows, not so well in others,
which this series aims to fix.
The first patches provide new basic functionality and do some refactoring;
the interesting part is in the two last patches.
This was discussed in
https://public-inbox.org/git/20180808221752.195419-1-sbeller@google.com/
and I think I addressed all feedback so far.
Thanks,
Stefan
Stefan Beller (9):
string-list: add string_list_{pop, last} functions
sha1-array: provide oid_array_filter
submodule.c: fix indentation
submodule.c: sort changed_submodule_names before searching it
submodule: move global changed_submodule_names into fetch submodule
struct
submodule.c: do not copy around submodule list
submodule: fetch in submodules git directory instead of in worktree
fetch: retry fetching submodules if sha1 were not fetched
builtin/fetch: check for submodule updates for non branch fetches
builtin/fetch.c | 14 +--
sha1-array.c | 18 ++++
sha1-array.h | 5 +
string-list.c | 14 +++
string-list.h | 11 +++
submodule.c | 189 ++++++++++++++++++++++++++++--------
t/t5526-fetch-submodules.sh | 23 ++++-
7 files changed, 227 insertions(+), 47 deletions(-)
--
2.19.0.397.gdd90340f6a-goog
^ permalink raw reply [relevance 10%]
* Re: 2.19.0.rc2.windows.1: stash fails with dirty submodule
@ 2018-09-11 22:24 4% ` Stefan Beller
0 siblings, 0 replies; 200+ results
From: Stefan Beller @ 2018-09-11 22:24 UTC (permalink / raw)
To: Jonathan Nieder
Cc: Thomas Braun, git, Philip Oakley, git-for-windows,
Johannes Schindelin
On Fri, Sep 7, 2018 at 5:35 PM Jonathan Nieder <jrnieder@gmail.com> wrote:
>
> Hi,
>
> Thomas Braun wrote:
>
> > I'm using git with stash and rebase builtins.
> >
> > $ git --version --build-options
> >
> > git version 2.19.0.rc2.windows.1
> [...]
> > mkdir test
> > cd test
> > git init
> > echo 1 > file
> > git add file
> > git commit file -m "message"
> > git submodule add ./ mysubmod
> > git commit -m "Add submodule"
> > echo 2 > mysubmod/file
> > git checkout -b mybranch
> > git rebase -i --autosquash master
> [...]
> > fatal: Unexpected stash response: ''
> >
> > and that used to work with older git versions.
>
> Thanks for reporting. I'm cc-ing Dscho, who has been looking for
> reports of issues with the new experimental stash and rebase code[2].
(It tests fine on my machine and I have no Windows machine at hand,
so ...)
This finally gave me an opportunity to play around with gitgitgadget
and its integrated CI for all major OS, see
https://github.com/gitgitgadget/git/pull/38
which is just the bug report put into our test suite.
^ permalink raw reply [relevance 4%]
* Re: [PATCH 06/11] submodule.c: sort changed_submodule_names before searching it
@ 2018-09-11 18:31 7% ` Stefan Beller
0 siblings, 0 replies; 200+ results
From: Stefan Beller @ 2018-09-11 18:31 UTC (permalink / raw)
To: Junio C Hamano; +Cc: git
On Thu, Sep 6, 2018 at 11:03 AM Junio C Hamano <gitster@pobox.com> wrote:
>
> Stefan Beller <sbeller@google.com> writes:
>
> > Instead of sorting it after we created an unsorted list, we could insert
> > correctly into the list.
>
> It is unclear what problem you are solving, especially with
> subjunctive "could" there. We are creating an unsorted list and
> then sorting it and you see it as a problem because it is just as
> easy and efficient to do the insertion sort while building up the
> list? (don't react and answer without reading all the way to the
> end; I think I know what is going on).
>
> > As the unsorted append is in order of cache entry
> > names, this is already sorted if names were equal to paths for submodules.
>
> That may be a statement of a fact, but it is unclear how that fact
> relates to what the code is doing before this patch, or what the
> code updated by this patch is doing.
>
> > As submodule names are often the same as their path, the input is sorted
> > pretty well already, so let's just do the sort afterwards.
>
> It is unclear what (performance?) trade-off this senttence is trying
> to make. It sounds as if it is claiming this:
>
> We can string_list_insert() to maintain sorted-ness of the
> list as we find new items, or we can string_list_append() to
> build an unsorted list and sort it at the end just once.
>
> To pick which one is more appropriate, we notice the fact
> that we discover new items more or less in the already
> sorted order. That makes "append then sort" more
> appropriate.
>
> But is that reasoning sensible?
>
> I'd imagine that append-then-sort would always be more efficient
> than insert-at-the-right-place-as-we-go, and the reason why we
> sometimes would need to do the latter is when we need to look up
> elements in the middle of building the list (e.g. we may want to
> de-dup, which requires us to look up a name from the ones we
> collected so far).
If we come across a (mostly) sorted list, then the insert-at-the-right-place
we'd come across the best case of insertion sort which is O(n), which
sounds better than append-then-sort as our sorting is a merge sort,
which has O(n log n) even in its best case (and needs to copy stuff
into a temp buffer and back).
By having the submodules named after its path, I strongly suspect
we have a mostly sorted list in nearly all cases except some really
interesting corner cases out there.
> And in this application, calculate_changed_submodule_paths()
> discover paths by calling collect_changed_submodules() which finds a
> mapping <submodule name, oid of commits> into a list sorted by
> submodule name, and then goes through that list and builds a list of
> submodules paths (which could be different from submodule names) by
> appending. Only after this list is fully built, get_next_submodule()
> gets called, so making the latter use string_list_lookup() that assumes
> a sorted list is safe if we built the list by append-then-sort (iow,
> sortedness while building the list does not matter).
>
> Having analysed all that, I find it somewhat iffy that _append() is
> used there in the calculate_changed_submodule_paths() function.
Note that this is fixed in the later patch
"submodule.c: do not copy around submodule list"
> It
> would cause the resulting changed_submodule_names list to hold the
> same name twice (or more),
This would be possible if there is a submodule at path A and another
submodule (at a different path) named "A", as we'll try hard to collect
names, but are also okay with path as we want to keep supporting the
historical use case of submodules.
> but I do not know if that would pose a
> problem to the consumer of the list. Using "accumulate then sort
> before calling look-up" would not change it as string_list_sort()
> would not dedup, so I do not think this patch would introduce a new
> problem, though.
Yes, that is true, so we'd want to extend the message above to
mention the potential duplicates.
Thanks,
Stefan
^ permalink raw reply [relevance 7%]
* [ANNOUNCE] Git v2.19.0
@ 2018-09-10 20:11 3% Junio C Hamano
0 siblings, 0 replies; 200+ results
From: Junio C Hamano @ 2018-09-10 20:11 UTC (permalink / raw)
To: git; +Cc: Linux Kernel, git-packagers
The latest feature release Git v2.19.0 is now available at the
usual places. It is comprised of 769 non-merge commits since
v2.18.0, contributed by 72 people, 16 of which are new faces.
The tarballs are found at:
https://www.kernel.org/pub/software/scm/git/
The following public repositories all have a copy of the 'v2.19.0'
tag and the 'master' branch that the tag points at:
url = https://kernel.googlesource.com/pub/scm/git/git
url = git://repo.or.cz/alt-git.git
url = https://github.com/gitster/git
New contributors whose contributions weren't in v2.18.0 are as follows.
Welcome to the Git development community!
Aleksandr Makarov, Andrei Rybak, Chen Bin, Henning Schild,
Isabella Stephens, Josh Steadmon, Jules Maselbas, Kana Natsuno,
Marc Strapetz, Masaya Suzuki, Nicholas Guriev, Raphaël Hertzog,
Samuel Maftoul, Sebastian Kisela, Vladimir Parfinenko, and
William Chargin.
Returning contributors who helped this release are as follows.
Thanks for your continued support.
Aaron Schrab, Ævar Arnfjörð Bjarmason, Alban Gruin, Alejandro
R. Sedeño, Alexander Shopov, Anthony Sottile, Antonio Ospite,
Beat Bolli, Ben Peart, Brandon Williams, brian m. carlson,
Christian Couder, Christopher Díaz Riveros, Derrick Stolee,
Dimitriy Ryazantcev, Elia Pinto, Elijah Newren, Eric Sunshine,
Han-Wen Nienhuys, Jameson Miller, Jean-Noël Avila, Jeff
Hostetler, Jeff King, Jiang Xin, Johannes Schindelin, Johannes
Sixt, Jonathan Nieder, Jonathan Tan, Junio C Hamano, Kim Gybels,
Kirill Smelkov, Kyle Meyer, Luis Marsano, Łukasz Stelmach,
Luke Diamand, Martin Ågren, Max Kirillov, Michael Barabanov,
Mike Hommey, Nguyễn Thái Ngọc Duy, Olga Telezhnaya, Peter
Krefting, Phillip Wood, Prathamesh Chavan, Ralf Thielow, Ramsay
Jones, René Scharfe, Stefan Beller, SZEDER Gábor, Taylor Blau,
Thomas Rast, Tobias Klauser, Todd Zullinger, Trần Ngọc Quân,
Ville Skyttä, and Xiaolong Ye.
----------------------------------------------------------------
Git 2.19 Release Notes
======================
Updates since v2.18
-------------------
UI, Workflows & Features
* "git diff" compares the index and the working tree. For paths
added with intent-to-add bit, the command shows the full contents
of them as added, but the paths themselves were not marked as new
files. They are now shown as new by default.
"git apply" learned the "--intent-to-add" option so that an
otherwise working-tree-only application of a patch will add new
paths to the index marked with the "intent-to-add" bit.
* "git grep" learned the "--column" option that gives not just the
line number but the column number of the hit.
* The "-l" option in "git branch -l" is an unfortunate short-hand for
"--create-reflog", but many users, both old and new, somehow expect
it to be something else, perhaps "--list". This step warns when "-l"
is used as a short-hand for "--create-reflog" and warns about the
future repurposing of the it when it is used.
* The userdiff pattern for .php has been updated.
* The content-transfer-encoding of the message "git send-email" sends
out by default was 8bit, which can cause trouble when there is an
overlong line to bust RFC 5322/2822 limit. A new option 'auto' to
automatically switch to quoted-printable when there is such a line
in the payload has been introduced and is made the default.
* "git checkout" and "git worktree add" learned to honor
checkout.defaultRemote when auto-vivifying a local branch out of a
remote tracking branch in a repository with multiple remotes that
have tracking branches that share the same names.
(merge 8d7b558bae ab/checkout-default-remote later to maint).
* "git grep" learned the "--only-matching" option.
* "git rebase --rebase-merges" mode now handles octopus merges as
well.
* Add a server-side knob to skip commits in exponential/fibbonacci
stride in an attempt to cover wider swath of history with a smaller
number of iterations, potentially accepting a larger packfile
transfer, instead of going back one commit a time during common
ancestor discovery during the "git fetch" transaction.
(merge 42cc7485a2 jt/fetch-negotiator-skipping later to maint).
* A new configuration variable core.usereplacerefs has been added,
primarily to help server installations that want to ignore the
replace mechanism altogether.
* Teach "git tag -s" etc. a few configuration variables (gpg.format
that can be set to "openpgp" or "x509", and gpg.<format>.program
that is used to specify what program to use to deal with the format)
to allow x.509 certs with CMS via "gpgsm" to be used instead of
openpgp via "gnupg".
* Many more strings are prepared for l10n.
* "git p4 submit" learns to ask its own pre-submit hook if it should
continue with submitting.
* The test performed at the receiving end of "git push" to prevent
bad objects from entering repository can be customized via
receive.fsck.* configuration variables; we now have gained a
counterpart to do the same on the "git fetch" side, with
fetch.fsck.* configuration variables.
* "git pull --rebase=interactive" learned "i" as a short-hand for
"interactive".
* "git instaweb" has been adjusted to run better with newer Apache on
RedHat based distros.
* "git range-diff" is a reimplementation of "git tbdiff" that lets us
compare individual patches in two iterations of a topic.
* The sideband code learned to optionally paint selected keywords at
the beginning of incoming lines on the receiving end.
* "git branch --list" learned to take the default sort order from the
'branch.sort' configuration variable, just like "git tag --list"
pays attention to 'tag.sort'.
* "git worktree" command learned "--quiet" option to make it less
verbose.
Performance, Internal Implementation, Development Support etc.
* The bulk of "git submodule foreach" has been rewritten in C.
* The in-core "commit" object had an all-purpose "void *util" field,
which was tricky to use especially in library-ish part of the
code. All of the existing uses of the field has been migrated to a
more dedicated "commit-slab" mechanism and the field is eliminated.
* A less often used command "git show-index" has been modernized.
(merge fb3010c31f jk/show-index later to maint).
* The conversion to pass "the_repository" and then "a_repository"
throughout the object access API continues.
* Continuing with the idea to programatically enumerate various
pieces of data required for command line completion, teach the
codebase to report the list of configuration variables
subcommands care about to help complete them.
* Separate "rebase -p" codepath out of "rebase -i" implementation to
slim down the latter and make it easier to manage.
* Make refspec parsing codepath more robust.
* Some flaky tests have been fixed.
* Continuing with the idea to programmatically enumerate various
pieces of data required for command line completion, the codebase
has been taught to enumerate options prefixed with "--no-" to
negate them.
* Build and test procedure for netrc credential helper (in contrib/)
has been updated.
* Remove unused function definitions and declarations from ewah
bitmap subsystem.
* Code preparation to make "git p4" closer to be usable with Python 3.
* Tighten the API to make it harder to misuse in-tree .gitmodules
file, even though it shares the same syntax with configuration
files, to read random configuration items from it.
* "git fast-import" has been updated to avoid attempting to create
delta against a zero-byte-long string, which is pointless.
* The codebase has been updated to compile cleanly with -pedantic
option.
(merge 2b647a05d7 bb/pedantic later to maint).
* The character display width table has been updated to match the
latest Unicode standard.
(merge 570951eea2 bb/unicode-11-width later to maint).
* test-lint now looks for broken use of "VAR=VAL shell_func" in test
scripts.
* Conversion from uchar[40] to struct object_id continues.
* Recent "security fix" to pay attention to contents of ".gitmodules"
while accepting "git push" was a bit overly strict than necessary,
which has been adjusted.
* "git fsck" learns to make sure the optional commit-graph file is in
a sane state.
* "git diff --color-moved" feature has further been tweaked.
* Code restructuring and a small fix to transport protocol v2 during
fetching.
* Parsing of -L[<N>][,[<M>]] parameters "git blame" and "git log"
take has been tweaked.
* lookup_commit_reference() and friends have been updated to find
in-core object for a specific in-core repository instance.
* Various glitches in the heuristics of merge-recursive strategy have
been documented in new tests.
* "git fetch" learned a new option "--negotiation-tip" to limit the
set of commits it tells the other end as "have", to reduce wasted
bandwidth and cycles, which would be helpful when the receiving
repository has a lot of refs that have little to do with the
history at the remote it is fetching from.
* For a large tree, the index needs to hold many cache entries
allocated on heap. These cache entries are now allocated out of a
dedicated memory pool to amortize malloc(3) overhead.
* Tests to cover various conflicting cases have been added for
merge-recursive.
* Tests to cover conflict cases that involve submodules have been
added for merge-recursive.
* Look for broken "&&" chains that are hidden in subshell, many of
which have been found and corrected.
* The singleton commit-graph in-core instance is made per in-core
repository instance.
* "make DEVELOPER=1 DEVOPTS=pedantic" allows developers to compile
with -pedantic option, which may catch more problematic program
constructs and potential bugs.
* Preparatory code to later add json output for telemetry data has
been added.
* Update the way we use Coccinelle to find out-of-style code that
need to be modernised.
* It is too easy to misuse system API functions such as strcat();
these selected functions are now forbidden in this codebase and
will cause a compilation failure.
* Add a script (in contrib/) to help users of VSCode work better with
our codebase.
* The Travis CI scripts were taught to ship back the test data from
failed tests.
(merge aea8879a6a sg/travis-retrieve-trash-upon-failure later to maint).
* The parse-options machinery learned to refrain from enclosing
placeholder string inside a "<bra" and "ket>" pair automatically
without PARSE_OPT_LITERAL_ARGHELP. Existing help text for option
arguments that are not formatted correctly have been identified and
fixed.
(merge 5f0df44cd7 rs/parse-opt-lithelp later to maint).
* Noiseword "extern" has been removed from function decls in the
header files.
* A few atoms like %(objecttype) and %(objectsize) in the format
specifier of "for-each-ref --format=<format>" can be filled without
getting the full contents of the object, but just with the object
header. These cases have been optimized by calling
oid_object_info() API (instead of reading and inspecting the data).
* The end result of documentation update has been made to be
inspected more easily to help developers.
* The API to iterate over all objects learned to optionally list
objects in the order they appear in packfiles, which helps locality
of access if the caller accesses these objects while as objects are
enumerated.
* Improve built-in facility to catch broken &&-chain in the tests.
* The more library-ish parts of the codebase learned to work on the
in-core index-state instance that is passed in by their callers,
instead of always working on the singleton "the_index" instance.
* A test prerequisite defined by various test scripts with slightly
different semantics has been consolidated into a single copy and
made into a lazily defined one.
(merge 6ec633059a wc/make-funnynames-shared-lazy-prereq later to maint).
* After a partial clone, repeated fetches from promisor remote would
have accumulated many packfiles marked with .promisor bit without
getting them coalesced into fewer packfiles, hurting performance.
"git repack" now learned to repack them.
* Partially revert the support for multiple hash functions to regain
hash comparison performance; we'd think of a way to do this better
in the next cycle.
* "git help --config" (which is used in command line completion)
missed the configuration variables not described in the main
config.txt file but are described in another file that is included
by it, which has been corrected.
* The test linter code has learned that the end of here-doc mark
"EOF" can be quoted in a double-quote pair, not just in a
single-quote pair.
Fixes since v2.18
-----------------
* "git remote update" can take both a single remote nickname and a
nickname for remote groups, and the completion script (in contrib/)
has been taught about it.
(merge 9cd4382ad5 ls/complete-remote-update-names later to maint).
* "git fetch --shallow-since=<cutoff>" that specifies the cut-off
point that is newer than the existing history used to end up
grabbing the entire history. Such a request now errors out.
(merge e34de73c56 nd/reject-empty-shallow-request later to maint).
* Fix for 2.17-era regression around `core.safecrlf`.
(merge 6cb09125be as/safecrlf-quiet-fix later to maint).
* The recent addition of "partial clone" experimental feature kicked
in when it shouldn't, namely, when there is no partial-clone filter
defined even if extensions.partialclone is set.
(merge cac1137dc4 jh/partial-clone later to maint).
* "git send-pack --signed" (hence "git push --signed" over the http
transport) did not read user ident from the config mechanism to
determine whom to sign the push certificate as, which has been
corrected.
(merge d067d98887 ms/send-pack-honor-config later to maint).
* "git fetch-pack --all" used to unnecessarily fail upon seeing an
annotated tag that points at an object other than a commit.
(merge c12c9df527 jk/fetch-all-peeled-fix later to maint).
* When user edits the patch in "git add -p" and the user's editor is
set to strip trailing whitespaces indiscriminately, an empty line
that is unchanged in the patch would become completely empty
(instead of a line with a sole SP on it). The code introduced in
Git 2.17 timeframe failed to parse such a patch, but now it learned
to notice the situation and cope with it.
(merge f4d35a6b49 pw/add-p-recount later to maint).
* The code to try seeing if a fetch is necessary in a submodule
during a fetch with --recurse-submodules got confused when the path
to the submodule was changed in the range of commits in the
superproject, sometimes showing "(null)". This has been corrected.
* Bugfix for "rebase -i" corner case regression.
(merge a9279c6785 pw/rebase-i-keep-reword-after-conflict later to maint).
* Recently added "--base" option to "git format-patch" command did
not correctly generate prereq patch ids.
(merge 15b76c1fb3 xy/format-patch-prereq-patch-id-fix later to maint).
* POSIX portability fix in Makefile to fix a glitch introduced a few
releases ago.
(merge 6600054e9b dj/runtime-prefix later to maint).
* "git filter-branch" when used with the "--state-branch" option
still attempted to rewrite the commits whose filtered result is
known from the previous attempt (which is recorded on the state
branch); the command has been corrected not to waste cycles doing
so.
(merge 709cfe848a mb/filter-branch-optim later to maint).
* Clarify that setting core.ignoreCase to deviate from reality would
not turn a case-incapable filesystem into a case-capable one.
(merge 48294b512a ms/core-icase-doc later to maint).
* "fsck.skipList" did not prevent a blob object listed there from
being inspected for is contents (e.g. we recently started to
inspect the contents of ".gitmodules" for certain malicious
patterns), which has been corrected.
(merge fb16287719 rj/submodule-fsck-skip later to maint).
* "git checkout --recurse-submodules another-branch" did not report
in which submodule it failed to update the working tree, which
resulted in an unhelpful error message.
(merge ba95d4e4bd sb/submodule-move-head-error-msg later to maint).
* "git rebase" behaved slightly differently depending on which one of
the three backends gets used; this has been documented and an
effort to make them more uniform has begun.
(merge b00bf1c9a8 en/rebase-consistency later to maint).
* The "--ignore-case" option of "git for-each-ref" (and its friends)
did not work correctly, which has been fixed.
(merge e674eb2528 jk/for-each-ref-icase later to maint).
* "git fetch" failed to correctly validate the set of objects it
received when making a shallow history deeper, which has been
corrected.
(merge cf1e7c0770 jt/connectivity-check-after-unshallow later to maint).
* Partial clone support of "git clone" has been updated to correctly
validate the objects it receives from the other side. The server
side has been corrected to send objects that are directly
requested, even if they may match the filtering criteria (e.g. when
doing a "lazy blob" partial clone).
(merge a7e67c11b8 jt/partial-clone-fsck-connectivity later to maint).
* Handling of an empty range by "git cherry-pick" was inconsistent
depending on how the range ended up to be empty, which has been
corrected.
(merge c5e358d073 jk/empty-pick-fix later to maint).
* "git reset --merge" (hence "git merge ---abort") and "git reset --hard"
had trouble working correctly in a sparsely checked out working
tree after a conflict, which has been corrected.
(merge b33fdfc34c mk/merge-in-sparse-checkout later to maint).
* Correct a broken use of "VAR=VAL shell_func" in a test.
(merge 650161a277 jc/t3404-one-shot-export-fix later to maint).
* "git rev-parse ':/substring'" did not consider the history leading
only to HEAD when looking for a commit with the given substring,
when the HEAD is detached. This has been fixed.
(merge 6b3351e799 wc/find-commit-with-pattern-on-detached-head later to maint).
* Build doc update for Windows.
(merge ede8d89bb1 nd/command-list later to maint).
* core.commentchar is now honored when preparing the list of commits
to replay in "rebase -i".
* "git pull --rebase" on a corrupt HEAD caused a segfault. In
general we substitute an empty tree object when running the in-core
equivalent of the diff-index command, and the codepath has been
corrected to do so as well to fix this issue.
(merge 3506dc9445 jk/has-uncommitted-changes-fix later to maint).
* httpd tests saw occasional breakage due to the way its access log
gets inspected by the tests, which has been updated to make them
less flaky.
(merge e8b3b2e275 sg/httpd-test-unflake later to maint).
* Tests to cover more D/F conflict cases have been added for
merge-recursive.
* "git gc --auto" opens file descriptors for the packfiles before
spawning "git repack/prune", which would upset Windows that does
not want a process to work on a file that is open by another
process. The issue has been worked around.
(merge 12e73a3ce4 kg/gc-auto-windows-workaround later to maint).
* The recursive merge strategy did not properly ensure there was no
change between HEAD and the index before performing its operation,
which has been corrected.
(merge 55f39cf755 en/dirty-merge-fixes later to maint).
* "git rebase" started exporting GIT_DIR environment variable and
exposing it to hook scripts when part of it got rewritten in C.
Instead of matching the old scripted Porcelains' behaviour,
compensate by also exporting GIT_WORK_TREE environment as well to
lessen the damage. This can harm existing hooks that want to
operate on different repository, but the current behaviour is
already broken for them anyway.
(merge ab5e67d751 bc/sequencer-export-work-tree-as-well later to maint).
* "git send-email" when using in a batched mode that limits the
number of messages sent in a single SMTP session lost the contents
of the variable used to choose between tls/ssl, unable to send the
second and later batches, which has been fixed.
(merge 636f3d7ac5 jm/send-email-tls-auth-on-batch later to maint).
* The lazy clone support had a few places where missing but promised
objects were not correctly tolerated, which have been fixed.
* One of the "diff --color-moved" mode "dimmed_zebra" that was named
in an unusual way has been deprecated and replaced by
"dimmed-zebra".
(merge e3f2f5f9cd es/diff-color-moved-fix later to maint).
* The wire-protocol v2 relies on the client to send "ref prefixes" to
limit the bandwidth spent on the initial ref advertisement. "git
clone" when learned to speak v2 forgot to do so, which has been
corrected.
(merge 402c47d939 bw/clone-ref-prefixes later to maint).
* "git diff --histogram" had a bad memory usage pattern, which has
been rearranged to reduce the peak usage.
(merge 79cb2ebb92 sb/histogram-less-memory later to maint).
* Code clean-up to use size_t/ssize_t when they are the right type.
(merge 7726d360b5 jk/size-t later to maint).
* The wire-protocol v2 relies on the client to send "ref prefixes" to
limit the bandwidth spent on the initial ref advertisement. "git
fetch $remote branch:branch" that asks tags that point into the
history leading to the "branch" automatically followed sent to
narrow prefix and broke the tag following, which has been fixed.
(merge 2b554353a5 jt/tag-following-with-proto-v2-fix later to maint).
* When the sparse checkout feature is in use, "git cherry-pick" and
other mergy operations lost the skip_worktree bit when a path that
is excluded from checkout requires content level merge, which is
resolved as the same as the HEAD version, without materializing the
merge result in the working tree, which made the path appear as
deleted. This has been corrected by preserving the skip_worktree
bit (and not materializing the file in the working tree).
(merge 2b75fb601c en/merge-recursive-skip-fix later to maint).
* The "author-script" file "git rebase -i" creates got broken when
we started to move the command away from shell script, which is
getting fixed now.
(merge 5522bbac20 es/rebase-i-author-script-fix later to maint).
* The automatic tree-matching in "git merge -s subtree" was broken 5
years ago and nobody has noticed since then, which is now fixed.
(merge 2ec4150713 jk/merge-subtree-heuristics later to maint).
* "git fetch $there refs/heads/s" ought to fetch the tip of the
branch 's', but when "refs/heads/refs/heads/s", i.e. a branch whose
name is "refs/heads/s" exists at the same time, fetched that one
instead by mistake. This has been corrected to honor the usual
disambiguation rules for abbreviated refnames.
(merge 60650a48c0 jt/refspec-dwim-precedence-fix later to maint).
* Futureproofing a helper function that can easily be misused.
(merge 65bb21e77e es/want-color-fd-defensive later to maint).
* The http-backend (used for smart-http transport) used to slurp the
whole input until EOF, without paying attention to CONTENT_LENGTH
that is supplied in the environment and instead expecting the Web
server to close the input stream. This has been fixed.
(merge eebfe40962 mk/http-backend-content-length later to maint).
* "git merge --abort" etc. did not clean things up properly when
there were conflicted entries in the index in certain order that
are involved in D/F conflicts. This has been corrected.
(merge ad3762042a en/abort-df-conflict-fixes later to maint).
* "git diff --indent-heuristic" had a bad corner case performance.
(merge 301ef85401 sb/indent-heuristic-optim later to maint).
* The "--exec" option to "git rebase --rebase-merges" placed the exec
commands at wrong places, which has been corrected.
* "git verify-tag" and "git verify-commit" have been taught to use
the exit status of underlying "gpg --verify" to signal bad or
untrusted signature they found.
(merge 4e5dc9ca17 jc/gpg-status later to maint).
* "git mergetool" stopped and gave an extra prompt to continue after
the last path has been handled, which did not make much sense.
(merge d651a54b8a ng/mergetool-lose-final-prompt later to maint).
* Among the three codepaths we use O_APPEND to open a file for
appending, one used for writing GIT_TRACE output requires O_APPEND
implementation that behaves sensibly when multiple processes are
writing to the same file. POSIX emulation used in the Windows port
has been updated to improve in this area.
(merge d641097589 js/mingw-o-append later to maint).
* "git pull --rebase -v" in a repository with a submodule barfed as
an intermediate process did not understand what "-v(erbose)" flag
meant, which has been fixed.
(merge e84c3cf3dc sb/pull-rebase-submodule later to maint).
* Recent update to "git config" broke updating variable in a
subsection, which has been corrected.
(merge bff7df7a87 sb/config-write-fix later to maint).
* When "git rebase -i" is told to squash two or more commits into
one, it labeled the log message for each commit with its number.
It correctly called the first one "1st commit", but the next one
was "commit #1", which was off-by-one. This has been corrected.
(merge dd2e36ebac pw/rebase-i-squash-number-fix later to maint).
* "git rebase -i", when a 'merge <branch>' insn in its todo list
fails, segfaulted, which has been (minimally) corrected.
(merge bc9238bb09 pw/rebase-i-merge-segv-fix later to maint).
* "git cherry-pick --quit" failed to remove CHERRY_PICK_HEAD even
though we won't be in a cherry-pick session after it returns, which
has been corrected.
(merge 3e7dd99208 nd/cherry-pick-quit-fix later to maint).
* In a recent update in 2.18 era, "git pack-objects" started
producing a larger than necessary packfiles by missing
opportunities to use large deltas. This has been corrected.
* The meaning of the possible values the "core.checkStat"
configuration variable can take were not adequately documented,
which has been fixed.
(merge 9bf5d4c4e2 nd/config-core-checkstat-doc later to maint).
* Recent "git rebase -i" update started to write bogusly formatted
author-script, with a matching broken reading code. These are
fixed.
* Recent addition of "directory rename" heuristics to the
merge-recursive backend makes the command susceptible to false
positives and false negatives. In the context of "git am -3",
which does not know about surrounding unmodified paths and thus
cannot inform the merge machinery about the full trees involved,
this risk is particularly severe. As such, the heuristic is
disabled for "git am -3" to keep the machinery "more stupid but
predictable".
* "git merge-base" in 2.19-rc1 has performance regression when the
(experimental) commit-graph feature is in use, which has been
mitigated.
* Code cleanup, docfix, build fix, etc.
(merge aee9be2ebe sg/update-ref-stdin-cleanup later to maint).
(merge 037714252f jc/clean-after-sanity-tests later to maint).
(merge 5b26c3c941 en/merge-recursive-cleanup later to maint).
(merge 0dcbc0392e bw/config-refer-to-gitsubmodules-doc later to maint).
(merge bb4d000e87 bw/protocol-v2 later to maint).
(merge 928f0ab4ba vs/typofixes later to maint).
(merge d7f590be84 en/rebase-i-microfixes later to maint).
(merge 81d395cc85 js/rebase-recreate-merge later to maint).
(merge 51d1863168 tz/exclude-doc-smallfixes later to maint).
(merge a9aa3c0927 ds/commit-graph later to maint).
(merge 5cf8e06474 js/enhanced-version-info later to maint).
(merge 6aaded5509 tb/config-default later to maint).
(merge 022d2ac1f3 sb/blame-color later to maint).
(merge 5a06a20e0c bp/test-drop-caches-for-windows later to maint).
(merge dd61cc1c2e jk/ui-color-always-to-auto later to maint).
(merge 1e83b9bfdd sb/trailers-docfix later to maint).
(merge ab29f1b329 sg/fast-import-dump-refs-on-checkpoint-fix later to maint).
(merge 6a8ad880f0 jn/subtree-test-fixes later to maint).
(merge ffbd51cc60 nd/pack-objects-threading-doc later to maint).
(merge e9dac7be60 es/mw-to-git-chain-fix later to maint).
(merge fe583c6c7a rs/remote-mv-leakfix later to maint).
(merge 69885ab015 en/t3031-title-fix later to maint).
(merge 8578037bed nd/config-blame-sort later to maint).
(merge 8ad169c4ba hn/config-in-code-comment later to maint).
(merge b7446fcfdf ar/t4150-am-scissors-test-fix later to maint).
(merge a8132410ee js/typofixes later to maint).
(merge 388d0ff6e5 en/update-index-doc later to maint).
(merge e05aa688dd jc/update-index-doc later to maint).
(merge 10c600172c sg/t5310-empty-input-fix later to maint).
(merge 5641eb9465 jh/partial-clone-doc later to maint).
(merge 2711b1ad5e ab/submodule-relative-url-tests later to maint).
(merge ce528de023 ab/unconditional-free-and-null later to maint).
(merge bbc072f5d8 rs/opt-updates later to maint).
(merge 69d846f053 jk/use-compat-util-in-test-tool later to maint).
(merge 1820703045 js/larger-timestamps later to maint).
(merge c8b35b95e1 sg/t4051-fix later to maint).
(merge 30612cb670 sg/t0020-conversion-fix later to maint).
(merge 15da753709 sg/t7501-thinkofix later to maint).
(merge 79b04f9b60 sg/t3903-missing-fix later to maint).
(merge 2745817028 sg/t3420-autostash-fix later to maint).
(merge 7afb0d6777 sg/test-rebase-editor-fix later to maint).
(merge 6c6ce21baa es/freebsd-iconv-portability later to maint).
----------------------------------------------------------------
Changes since v2.18.0 are as follows:
Aaron Schrab (1):
sequencer: use configured comment character
Alban Gruin (4):
rebase: introduce a dedicated backend for --preserve-merges
rebase: strip unused code in git-rebase--preserve-merges.sh
rebase: use the new git-rebase--preserve-merges.sh
rebase: remove -p code from git-rebase--interactive.sh
Alejandro R. Sedeño (1):
Makefile: tweak sed invocation
Aleksandr Makarov (1):
for-each-ref: consistently pass WM_IGNORECASE flag
Alexander Shopov (1):
l10n: bg.po: Updated Bulgarian translation (3958t)
Andrei Rybak (2):
Documentation: fix --color option formatting
t4150: fix broken test for am --scissors
Anthony Sottile (1):
config.c: fix regression for core.safecrlf false
Antonio Ospite (6):
config: move config_from_gitmodules to submodule-config.c
submodule-config: add helper function to get 'fetch' config from .gitmodules
submodule-config: add helper to get 'update-clone' config from .gitmodules
submodule-config: make 'config_from_gitmodules' private
submodule-config: pass repository as argument to config_from_gitmodules
submodule-config: reuse config_from_gitmodules in repo_read_gitmodules
Beat Bolli (10):
builtin/config: work around an unsized array forward declaration
unicode: update the width tables to Unicode 11
connect.h: avoid forward declaration of an enum
refs/refs-internal.h: avoid forward declaration of an enum
convert.c: replace "\e" escapes with "\033".
sequencer.c: avoid empty statements at top level
string-list.c: avoid conversion from void * to function pointer
utf8.c: avoid char overflow
Makefile: add a DEVOPTS flag to get pedantic compilation
packfile: ensure that enum object_type is defined
Ben Peart (3):
convert log_ref_write_fd() to use strbuf
handle lower case drive letters on Windows
t3507: add a testcase showing failure with sparse checkout
Brandon Williams (15):
commit: convert commit_graft_pos() to handle arbitrary repositories
commit: convert register_commit_graft to handle arbitrary repositories
commit: convert read_graft_file to handle arbitrary repositories
test-pkt-line: add unpack-sideband subcommand
docs: link to gitsubmodules
upload-pack: implement ref-in-want
upload-pack: test negotiation with changing repository
fetch: refactor the population of peer ref OIDs
fetch: refactor fetch_refs into two functions
fetch: refactor to make function args narrower
fetch-pack: put shallow info in output parameter
fetch-pack: implement ref-in-want
clone: send ref-prefixes when using protocol v2
fetch-pack: mark die strings for translation
pack-protocol: mention and point to docs for protocol v2
Chen Bin (1):
git-p4: add the `p4-pre-submit` hook
Christian Couder (1):
t9104: kosherly remove remote refs
Christopher Díaz Riveros (1):
l10n: es.po v2.19.0 round 2
Derrick Stolee (46):
ref-filter: fix outdated comment on in_commit_list
commit: add generation number to struct commit
commit-graph: compute generation numbers
commit: use generations in paint_down_to_common()
commit-graph: always load commit-graph information
ref-filter: use generation number for --contains
commit: use generation numbers for in_merge_bases()
commit: add short-circuit to paint_down_to_common()
commit: use generation number in remove_redundant()
merge: check config before loading commits
commit-graph.txt: update design document
commit-graph: fix UX issue when .lock file exists
ewah/bitmap.c: delete unused 'bitmap_clear()'
ewah/bitmap.c: delete unused 'bitmap_each_bit()'
ewah_bitmap: delete unused 'ewah_and()'
ewah_bitmap: delete unused 'ewah_and_not()'
ewah_bitmap: delete unused 'ewah_not()'
ewah_bitmap: delete unused 'ewah_or()'
ewah_io: delete unused 'ewah_serialize()'
t5318-commit-graph.sh: use core.commitGraph
commit-graph: UNLEAK before die()
commit-graph: fix GRAPH_MIN_SIZE
commit-graph: parse commit from chosen graph
commit: force commit to parse from object database
commit-graph: load a root tree from specific graph
commit-graph: add 'verify' subcommand
commit-graph: verify catches corrupt signature
commit-graph: verify required chunks are present
commit-graph: verify corrupt OID fanout and lookup
commit-graph: verify objects exist
commit-graph: verify root tree OIDs
commit-graph: verify parent list
commit-graph: verify generation number
commit-graph: verify commit date
commit-graph: test for corrupted octopus edge
commit-graph: verify contents match checksum
fsck: verify commit-graph
commit-graph: use string-list API for input
commit-graph: add '--reachable' option
gc: automatically write commit-graph files
commit-graph: update design document
commit-graph: fix documentation inconsistencies
coccinelle: update commit.cocci
commit: use timestamp_t for author_date_slab
config: fix commit-graph related config docs
commit: don't use generation numbers if not needed
Dimitriy Ryazantcev (1):
l10n: ru.po: update Russian translation
Elia Pinto (1):
worktree: add --quiet option
Elijah Newren (66):
t6036, t6042: use test_create_repo to keep tests independent
t6036, t6042: use test_line_count instead of wc -l
t6036, t6042: prefer test_path_is_file, test_path_is_missing
t6036, t6042: prefer test_cmp to sequences of test
t6036: prefer test_when_finished to manual cleanup in following test
merge-recursive: fix miscellaneous grammar error in comment
merge-recursive: fix numerous argument alignment issues
merge-recursive: align labels with their respective code blocks
merge-recursive: clarify the rename_dir/RENAME_DIR meaning
merge-recursive: rename conflict_rename_*() family of functions
merge-recursive: add pointer about unduly complex looking code
git-rebase.txt: document incompatible options
git-rebase.sh: update help messages a bit
t3422: new testcases for checking when incompatible options passed
git-rebase: error out when incompatible options passed
git-rebase.txt: address confusion between --no-ff vs --force-rebase
directory-rename-detection.txt: technical docs on abilities and limitations
git-rebase.txt: document behavioral differences between modes
t3401: add directory rename testcases for rebase and am
git-rebase: make --allow-empty-message the default
t3418: add testcase showing problems with rebase -i and strategy options
Fix use of strategy options with interactive rebases
git-rebase--merge: modernize "git-$cmd" to "git $cmd"
apply: fix grammar error in comment
t5407: fix test to cover intended arguments
read-cache.c: move index_has_changes() from merge.c
index_has_changes(): avoid assuming operating on the_index
t6044: verify that merges expected to abort actually abort
t6036: add a failed conflict detection case with symlink modify/modify
t6036: add a failed conflict detection case with symlink add/add
t6036: add a failed conflict detection case with submodule modify/modify
t6036: add a failed conflict detection case with submodule add/add
t6036: add a failed conflict detection case with conflicting types
t6042: add testcase covering rename/add/delete conflict type
t6042: add testcase covering rename/rename(2to1)/delete/delete conflict
t6042: add testcase covering long chains of rename conflicts
t6036: add lots of detail for directory/file conflicts in recursive case
t6036: add a failed conflict detection case: regular files, different modes
t6044: add a testcase for index matching head, when head doesn't match HEAD
merge-recursive: make sure when we say we abort that we actually abort
merge-recursive: fix assumption that head tree being merged is HEAD
t6044: add more testcases with staged changes before a merge is invoked
merge-recursive: enforce rule that index matches head before merging
merge: fix misleading pre-merge check documentation
t7405: add a file/submodule conflict
t7405: add a directory/submodule conflict
t7405: verify 'merge --abort' works after submodule/path conflicts
merge-recursive: preserve skip_worktree bit when necessary
t1015: demonstrate directory/file conflict recovery failures
read-cache: fix directory/file conflict handling in read_index_unmerged()
t3031: update test description to mention desired behavior
t7406: fix call that was failing for the wrong reason
t7406: simplify by using diff --name-only instead of diff --raw
t7406: avoid having git commands upstream of a pipe
t7406: prefer test_* helper functions to test -[feds]
t7406: avoid using test_must_fail for commands other than git
git-update-index.txt: reword possibly confusing example
Add missing includes and forward declarations
alloc: make allocate_alloc_state and clear_alloc_state more consistent
Move definition of enum branch_track from cache.h to branch.h
urlmatch.h: fix include guard
compat/precompose_utf8.h: use more common include guard style
Remove forward declaration of an enum
t3401: add another directory rename testcase for rebase and am
merge-recursive: add ability to turn off directory rename detection
am: avoid directory rename detection when calling recursive merge machinery
Eric Sunshine (55):
t: use test_might_fail() instead of manipulating exit code manually
t: use test_write_lines() instead of series of 'echo' commands
t: use sane_unset() rather than 'unset' with broken &&-chain
t: drop unnecessary terminating semicolon in subshell
t/lib-submodule-update: fix "absorbing" test
t5405: use test_must_fail() instead of checking exit code manually
t5406: use write_script() instead of birthing shell script manually
t5505: modernize and simplify hard-to-digest test
t6036: fix broken "merge fails but has appropriate contents" tests
t7201: drop pointless "exit 0" at end of subshell
t7400: fix broken "submodule add/reconfigure --force" test
t7810: use test_expect_code() instead of hand-rolled comparison
t9001: fix broken "invoke hook" test
t9814: simplify convoluted check that command correctly errors out
t0000-t0999: fix broken &&-chains
t1000-t1999: fix broken &&-chains
t2000-t2999: fix broken &&-chains
t3000-t3999: fix broken &&-chains
t3030: fix broken &&-chains
t4000-t4999: fix broken &&-chains
t5000-t5999: fix broken &&-chains
t6000-t6999: fix broken &&-chains
t7000-t7999: fix broken &&-chains
t9000-t9999: fix broken &&-chains
t9119: fix broken &&-chains
t6046/t9833: fix use of "VAR=VAL cmd" with a shell function
t/check-non-portable-shell: stop being so polite
t/check-non-portable-shell: make error messages more compact
t/check-non-portable-shell: detect "FOO=bar shell_func"
t/test-lib: teach --chain-lint to detect broken &&-chains in subshells
t/Makefile: add machinery to check correctness of chainlint.sed
t/chainlint: add chainlint "basic" test cases
t/chainlint: add chainlint "whitespace" test cases
t/chainlint: add chainlint "one-liner" test cases
t/chainlint: add chainlint "nested subshell" test cases
t/chainlint: add chainlint "loop" and "conditional" test cases
t/chainlint: add chainlint "cuddled" test cases
t/chainlint: add chainlint "complex" test cases
t/chainlint: add chainlint "specialized" test cases
diff: --color-moved: rename "dimmed_zebra" to "dimmed-zebra"
mw-to-git/t9360: fix broken &&-chain
t/chainlint.sed: drop extra spaces from regex character class
sequencer: fix "rebase -i --root" corrupting author header
sequencer: fix "rebase -i --root" corrupting author header timezone
sequencer: fix "rebase -i --root" corrupting author header timestamp
sequencer: don't die() on bogus user-edited timestamp
color: protect against out-of-bounds reads and writes
chainlint: match arbitrary here-docs tags rather than hard-coded names
chainlint: match 'quoted' here-doc tags
chainlint: recognize multi-line $(...) when command cuddled with "$("
chainlint: let here-doc and multi-line string commence on same line
chainlint: recognize multi-line quoted strings more robustly
chainlint: add test of pathological case which triggered false positive
chainlint: match "quoted" here-doc tags
config.mak.uname: resolve FreeBSD iconv-related compilation warning
Han-Wen Nienhuys (2):
config: document git config getter return value
sideband: highlight keywords in remote sideband output
Henning Schild (9):
builtin/receive-pack: use check_signature from gpg-interface
gpg-interface: make parse_gpg_output static and remove from interface header
gpg-interface: add new config to select how to sign a commit
t/t7510: check the validation of the new config gpg.format
gpg-interface: introduce an abstraction for multiple gpg formats
gpg-interface: do not hardcode the key string len anymore
gpg-interface: introduce new config to select per gpg format program
gpg-interface: introduce new signature format "x509" using gpgsm
gpg-interface t: extend the existing GPG tests with GPGSM
Isabella Stephens (2):
blame: prevent error if range ends past end of file
log: prevent error if line range ends past end of file
Jameson Miller (8):
read-cache: teach refresh_cache_entry to take istate
read-cache: teach make_cache_entry to take object_id
block alloc: add lifecycle APIs for cache_entry structs
mem-pool: only search head block for available space
mem-pool: add life cycle management functions
mem-pool: fill out functionality
block alloc: allocate cache entries from mem_pool
block alloc: add validations around cache_entry lifecyle
Jean-Noël Avila (3):
i18n: fix mistakes in translated strings
l10n: fr.po v2.19.0 rnd 1
l10n: fr.po v2.19.0 rnd 2
Jeff Hostetler (1):
json_writer: new routines to create JSON data
Jeff King (50):
make show-index a builtin
show-index: update documentation for index v2
fetch-pack: don't try to fetch peel values with --all
ewah: drop ewah_deserialize function
ewah: drop ewah_serialize_native function
t3200: unset core.logallrefupdates when testing reflog creation
t: switch "branch -l" to "branch --create-reflog"
branch: deprecate "-l" option
config: turn die_on_error into caller-facing enum
config: add CONFIG_ERROR_SILENT handler
config: add options parameter to git_config_from_mem
fsck: silence stderr when parsing .gitmodules
t6300: add a test for --ignore-case
ref-filter: avoid backend filtering with --ignore-case
t5500: prettify non-commit tag tests
sequencer: handle empty-set cases consistently
sequencer: don't say BUG on bogus input
has_uncommitted_changes(): fall back to empty tree
fsck: split ".gitmodules too large" error from parse failure
fsck: downgrade gitmodulesParse default to "info"
blame: prefer xsnprintf to strcpy for colors
check_replace_refs: fix outdated comment
check_replace_refs: rename to read_replace_refs
add core.usereplacerefs config option
reencode_string: use st_add/st_mult helpers
reencode_string: use size_t for string lengths
strbuf: use size_t for length in intermediate variables
strbuf_readlink: use ssize_t
pass st.st_size as hint for strbuf_readlink()
strbuf_humanise: use unsigned variables
automatically ban strcpy()
banned.h: mark strcat() as banned
banned.h: mark sprintf() as banned
banned.h: mark strncpy() as banned
score_trees(): fix iteration over trees with missing entries
add a script to diff rendered documentation
t5552: suppress upload-pack trace output
for_each_*_object: store flag definitions in a single location
for_each_*_object: take flag arguments as enum
for_each_*_object: give more comprehensive docstrings
for_each_packed_object: support iterating in pack-order
t1006: test cat-file --batch-all-objects with duplicates
cat-file: rename batch_{loose,packed}_object callbacks
cat-file: support "unordered" output for --batch-all-objects
cat-file: use oidset check-and-insert
cat-file: split batch "buf" into two variables
cat-file: use a single strbuf for all output
for_each_*_object: move declarations to object-store.h
test-tool.h: include git-compat-util.h
hashcmp: assert constant hash size
Jiang Xin (4):
l10n: zh_CN: review for git 2.18.0
l10n: git.pot: v2.19.0 round 1 (382 new, 30 removed)
l10n: git.pot: v2.19.0 round 2 (3 new, 5 removed)
l10n: zh_CN: for git v2.19.0 l10n round 1 to 2
Johannes Schindelin (41):
Makefile: fix the "built from commit" code
merge: allow reading the merge commit message from a file
rebase --rebase-merges: add support for octopus merges
rebase --rebase-merges: adjust man page for octopus support
vcbuild/README: update to accommodate for missing common-cmds.h
t7406: avoid failures solely due to timing issues
contrib: add a script to initialize VS Code configuration
vscode: hard-code a couple defines
cache.h: extract enum declaration from inside a struct declaration
mingw: define WIN32 explicitly
vscode: only overwrite C/C++ settings
vscode: wrap commit messages at column 72 by default
vscode: use 8-space tabs, no trailing ws, etc for Git's source code
vscode: add a dictionary for cSpell
vscode: let cSpell work on commit messages, too
pull --rebase=<type>: allow single-letter abbreviations for the type
t3430: demonstrate what -r, --autosquash & --exec should do
git-compat-util.h: fix typo
remote-curl: remove spurious period
rebase --exec: make it work with --rebase-merges
linear-assignment: a function to solve least-cost assignment problems
Introduce `range-diff` to compare iterations of a topic branch
range-diff: first rudimentary implementation
range-diff: improve the order of the shown commits
range-diff: also show the diff between patches
range-diff: right-trim commit messages
range-diff: indent the diffs just like tbdiff
range-diff: suppress the diff headers
range-diff: adjust the output of the commit pairs
range-diff: do not show "function names" in hunk headers
range-diff: use color for the commit pairs
color: add the meta color GIT_COLOR_REVERSE
diff: add an internal option to dual-color diffs of diffs
range-diff: offer to dual-color the diffs
range-diff --dual-color: skip white-space warnings
range-diff: populate the man page
completion: support `git range-diff`
range-diff: left-pad patch numbers
range-diff: make --dual-color the default mode
range-diff: use dim/bold cues to improve dual color mode
chainlint: fix for core.autocrlf=true
Johannes Sixt (1):
mingw: enable atomic O_APPEND
Jonathan Nieder (12):
object: add repository argument to grow_object_hash
object: move grafts to object parser
commit: add repository argument to commit_graft_pos
commit: add repository argument to register_commit_graft
commit: add repository argument to read_graft_file
commit: add repository argument to prepare_commit_graft
commit: add repository argument to lookup_commit_graft
subtree test: add missing && to &&-chain
subtree test: simplify preparation of expected results
doc hash-function-transition: pick SHA-256 as NewHash
partial-clone: render design doc using asciidoc
Revert "Merge branch 'sb/submodule-core-worktree'"
Jonathan Tan (28):
list-objects: check if filter is NULL before using
fetch-pack: split up everything_local()
fetch-pack: clear marks before re-marking
fetch-pack: directly end negotiation if ACK ready
fetch-pack: use ref adv. to prune "have" sent
fetch-pack: make negotiation-related vars local
fetch-pack: move common check and marking together
fetch-pack: introduce negotiator API
pack-bitmap: remove bitmap_git global variable
pack-bitmap: add free function
fetch-pack: write shallow, then check connectivity
fetch-pack: support negotiation tip whitelist
upload-pack: send refs' objects despite "filter"
clone: check connectivity even if clone is partial
revision: tolerate promised targets of tags
tag: don't warn if target is missing but promised
negotiator/skipping: skip commits during fetch
commit-graph: refactor preparing commit graph
object-store: add missing include
commit-graph: add missing forward declaration
commit-graph: add free_commit_graph
commit-graph: store graph in struct object_store
commit-graph: add repo arg to graph readers
t5702: test fetch with multiple refspecs at a time
fetch: send "refs/tags/" prefix upon CLI refspecs
fetch-pack: unify ref in and out param
repack: refactor setup of pack-objects cmd
repack: repack promisor objects if -a or -A is set
Josh Steadmon (1):
protocol-v2 doc: put HTTP headers after request
Jules Maselbas (1):
send-email: fix tls AUTH when sending batch
Junio C Hamano (23):
tests: clean after SANITY tests
ewah: delete unused 'rlwit_discharge_empty()'
Prepare to start 2.19 cycle
First batch for 2.19 cycle
Second batch for 2.19 cycle
fixup! connect.h: avoid forward declaration of an enum
fixup! refs/refs-internal.h: avoid forward declaration of an enum
t3404: fix use of "VAR=VAL cmd" with a shell function
Third batch for 2.19 cycle
Fourth batch for 2.19 cycle
remote: make refspec follow the same disambiguation rule as local refs
Fifth batch for 2.19 cycle
update-index: there no longer is `apply --index-info`
gpg-interface: propagate exit status from gpg back to the callers
Sixth batch for 2.19 cycle
config.txt: clarify core.checkStat
Seventh batch for 2.19 cycle
sideband: do not read beyond the end of input
Git 2.19-rc0
Getting ready for -rc1
Git 2.19-rc1
Git 2.19-rc2
Git 2.19
Kana Natsuno (2):
t4018: add missing test cases for PHP
userdiff: support new keywords in PHP hunk header
Kim Gybels (1):
gc --auto: release pack files before auto packing
Kirill Smelkov (1):
fetch-pack: test explicitly that --all can fetch tag references pointing to non-commits
Kyle Meyer (1):
range-diff: update stale summary of --no-dual-color
Luis Marsano (2):
git-credential-netrc: use in-tree Git.pm for tests
git-credential-netrc: fix exit status when tests fail
Luke Diamand (6):
git-p4: python3: replace <> with !=
git-p4: python3: replace dict.has_key(k) with "k in dict"
git-p4: python3: remove backticks
git-p4: python3: basestring workaround
git-p4: python3: use print() function
git-p4: python3: fix octal constants
Marc Strapetz (1):
Documentation: declare "core.ignoreCase" as internal variable
Martin Ågren (1):
refspec: initalize `refspec_item` in `valid_fetch_refspec()`
Masaya Suzuki (2):
builtin/send-pack: populate the default configs
doc: fix want-capability separator
Max Kirillov (5):
http-backend: cleanup writing to child process
http-backend: respect CONTENT_LENGTH as specified by rfc3875
unpack-trees: do not fail reset because of unmerged skipped entry
http-backend: respect CONTENT_LENGTH for receive-pack
http-backend: allow empty CONTENT_LENGTH
Michael Barabanov (1):
filter-branch: skip commits present on --state-branch
Mike Hommey (1):
fast-import: do not call diff_delta() with empty buffer
Nguyễn Thái Ngọc Duy (100):
commit-slab.h: code split
commit-slab: support shared commit-slab
blame: use commit-slab for blame suspects instead of commit->util
describe: use commit-slab for commit names instead of commit->util
shallow.c: use commit-slab for commit depth instead of commit->util
sequencer.c: use commit-slab to mark seen commits
sequencer.c: use commit-slab to associate todo items to commits
revision.c: use commit-slab for show_source
bisect.c: use commit-slab for commit weight instead of commit->util
name-rev: use commit-slab for rev-name instead of commit->util
show-branch: use commit-slab for commit-name instead of commit->util
show-branch: note about its object flags usage
log: use commit-slab in prepare_bases() instead of commit->util
merge: use commit-slab in merge remote desc instead of commit->util
commit.h: delete 'util' field in struct commit
diff: ignore --ita-[in]visible-in-index when diffing worktree-to-tree
diff: turn --ita-invisible-in-index on by default
t2203: add a test about "diff HEAD" case
apply: add --intent-to-add
parse-options: option to let --git-completion-helper show negative form
completion: suppress some -no- options
Add and use generic name->id mapping code for color slot parsing
grep: keep all colors in an array
fsck: factor out msg_id_info[] lazy initialization code
help: add --config to list all available config
fsck: produce camelCase config key names
advice: keep config name in camelCase in advice_config[]
am: move advice.amWorkDir parsing back to advice.c
completion: drop the hard coded list of config vars
completion: keep other config var completion in camelCase
completion: support case-insensitive config vars
log-tree: allow to customize 'grafted' color
completion: complete general config vars in two steps
upload-pack: reject shallow requests that would return nothing
completion: collapse extra --no-.. options
pack-objects: fix performance issues on packing large deltas
Update messages in preparation for i18n
archive-tar.c: mark more strings for translation
archive-zip.c: mark more strings for translation
builtin/config.c: mark more strings for translation
builtin/grep.c: mark strings for translation
builtin/pack-objects.c: mark more strings for translation
builtin/replace.c: mark more strings for translation
commit-graph.c: mark more strings for translation
config.c: mark more strings for translation
connect.c: mark more strings for translation
convert.c: mark more strings for translation
dir.c: mark more strings for translation
environment.c: mark more strings for translation
exec-cmd.c: mark more strings for translation
object.c: mark more strings for translation
pkt-line.c: mark more strings for translation
refs.c: mark more strings for translation
refspec.c: mark more strings for translation
replace-object.c: mark more strings for translation
sequencer.c: mark more strings for translation
sha1-file.c: mark more strings for translation
transport.c: mark more strings for translation
transport-helper.c: mark more strings for translation
pack-objects: document about thread synchronization
apply.h: drop extern on func declaration
attr.h: drop extern from function declaration
blame.h: drop extern on func declaration
cache-tree.h: drop extern from function declaration
convert.h: drop 'extern' from function declaration
diffcore.h: drop extern from function declaration
diff.h: remove extern from function declaration
line-range.h: drop extern from function declaration
rerere.h: drop extern from function declaration
repository.h: drop extern from function declaration
revision.h: drop extern from function declaration
submodule.h: drop extern from function declaration
config.txt: reorder blame stuff to keep config keys sorted
Makefile: add missing dependency for command-list.h
diff.c: move read_index() code back to the caller
cache-tree: wrap the_index based wrappers with #ifdef
attr: remove an implicit dependency on the_index
convert.c: remove an implicit dependency on the_index
dir.c: remove an implicit dependency on the_index in pathspec code
preload-index.c: use the right index instead of the_index
ls-files: correct index argument to get_convert_attr_ascii()
unpack-trees: remove 'extern' on function declaration
unpack-trees: add a note about path invalidation
unpack-trees: don't shadow global var the_index
unpack-trees: convert clear_ce_flags* to avoid the_index
unpack-trees: avoid the_index in verify_absent()
pathspec.c: use the right index instead of the_index
submodule.c: use the right index instead of the_index
entry.c: use the right index instead of the_index
attr: remove index from git_attr_set_direction()
grep: use the right index instead of the_index
archive.c: avoid access to the_index
archive-*.c: use the right repository
resolve-undo.c: use the right index instead of the_index
apply.c: pass struct apply_state to more functions
apply.c: make init_apply_state() take a struct repository
apply.c: remove implicit dependency on the_index
blame.c: remove implicit dependency on the_index
cherry-pick: fix --quit not deleting CHERRY_PICK_HEAD
generate-cmdlist.sh: collect config from all config.txt files
Nicholas Guriev (1):
mergetool: don't suggest to continue after last file
Olga Telezhnaya (5):
ref-filter: add info_source to valid_atom
ref-filter: fill empty fields with empty values
ref-filter: initialize eaten variable
ref-filter: merge get_obj and get_object
ref-filter: use oid_object_info() to get object
Peter Krefting (2):
l10n: sv.po: Update Swedish translation(3608t0f0u)
l10n: sv.po: Update Swedish translation (3958t0f0u)
Phillip Wood (7):
add -p: fix counting empty context lines in edited patches
sequencer: do not squash 'reword' commits when we hit conflicts
sequencer: handle errors from read_author_ident()
sequencer: fix quoting in write_author_script
rebase -i: fix numbering in squash message
t3430: add conflicting commit
rebase -i: fix SIGSEGV when 'merge <branch>' fails
Prathamesh Chavan (4):
submodule foreach: correct '$path' in nested submodules from a subdirectory
submodule foreach: document '$sm_path' instead of '$path'
submodule foreach: document variable '$displaypath'
submodule: port submodule subcommand 'foreach' from shell to C
Ralf Thielow (1):
l10n: de.po: translate 108 new messages
Ramsay Jones (3):
fsck: check skiplist for object in fsck_blob()
t6036: fix broken && chain in sub-shell
t5562: avoid non-portable "export FOO=bar" construct
Raphaël Hertzog (1):
l10n: fr: fix a message seen in git bisect
René Scharfe (10):
remote: clear string_list after use in mv()
add, update-index: fix --chmod argument help
difftool: remove angular brackets from argument help
pack-objects: specify --index-version argument help explicitly
send-pack: specify --force-with-lease argument help explicitly
shortlog: correct option help for -w
parse-options: automatically infer PARSE_OPT_LITERAL_ARGHELP
checkout-index: improve argument help for --stage
remote: improve argument help for add --mirror
parseopt: group literal string alternatives in argument help
SZEDER Gábor (30):
update-ref --stdin: use skip_prefix()
t7510-signed-commit: use 'test_must_fail'
tests: make forging GPG signed commits and tags more robust
t5541: clean up truncating access log
t/lib-httpd: add the strip_access_log() helper function
t/lib-httpd: avoid occasional failures when checking access.log
t5608: fix broken &&-chain
t9300: wait for background fast-import process to die after killing it
travis-ci: run Coccinelle static analysis with two parallel jobs
travis-ci: fail if Coccinelle static analysis found something to transform
coccinelle: mark the 'coccicheck' make target as .PHONY
coccinelle: use $(addsuffix) in 'coccicheck' make target
coccinelle: exclude sha1dc source files from static analysis
coccinelle: put sane filenames into output patches
coccinelle: extract dedicated make target to clean Coccinelle's results
travis-ci: include the trash directories of failed tests in the trace log
t5318: use 'test_cmp_bin' to compare commit-graph files
t5318: avoid unnecessary command substitutions
t5310-pack-bitmaps: fix bogus 'pack-objects to file can use bitmap' test
tests: use 'test_must_be_empty' instead of '! test -s'
tests: use 'test_must_be_empty' instead of 'test ! -s'
tests: use 'test_must_be_empty' instead of 'test_cmp /dev/null <out>'
tests: use 'test_must_be_empty' instead of 'test_cmp <empty> <out>'
t7501-commit: drop silly command substitution
t0020-crlf: check the right file
t4051-diff-function-context: read the right file
t6018-rev-list-glob: fix 'empty stdin' test
t3903-stash: don't try to grep non-existing file
t3420-rebase-autostash: don't try to grep non-existing files
t/lib-rebase.sh: support explicit 'pick' commands in 'fake_editor.sh'
Samuel Maftoul (1):
branch: support configuring --sort via .gitconfig
Sebastian Kisela (2):
git-instaweb: support Fedora/Red Hat apache module path
git-instaweb: fix apache2 config with apache >= 2.4
Stefan Beller (87):
repository: introduce parsed objects field
object: add repository argument to create_object
alloc: add repository argument to alloc_blob_node
alloc: add repository argument to alloc_tree_node
alloc: add repository argument to alloc_commit_node
alloc: add repository argument to alloc_tag_node
alloc: add repository argument to alloc_object_node
alloc: add repository argument to alloc_report
alloc: add repository argument to alloc_commit_index
object: allow grow_object_hash to handle arbitrary repositories
object: allow create_object to handle arbitrary repositories
alloc: allow arbitrary repositories for alloc functions
object-store: move object access functions to object-store.h
shallow: add repository argument to set_alternate_shallow_file
shallow: add repository argument to register_shallow
shallow: add repository argument to check_shallow_file_for_update
shallow: add repository argument to is_repository_shallow
cache: convert get_graft_file to handle arbitrary repositories
path.c: migrate global git_path_* to take a repository argument
shallow: migrate shallow information into the object parser
commit: allow prepare_commit_graft to handle arbitrary repositories
commit: allow lookup_commit_graft to handle arbitrary repositories
refs/packed-backend.c: close fd of empty file
submodule--helper: plug mem leak in print_default_remote
sequencer.c: plug leaks in do_pick_commit
submodule: fix NULL correctness in renamed broken submodules
t5526: test recursive submodules when fetching moved submodules
submodule: unset core.worktree if no working tree is present
submodule: ensure core.worktree is set after update
submodule deinit: unset core.worktree
submodule.c: report the submodule that an error occurs in
sequencer.c: plug mem leak in git_sequencer_config
.mailmap: merge different spellings of names
object: add repository argument to parse_object
object: add repository argument to lookup_object
object: add repository argument to parse_object_buffer
object: add repository argument to object_as_type
blob: add repository argument to lookup_blob
tree: add repository argument to lookup_tree
commit: add repository argument to lookup_commit_reference_gently
commit: add repository argument to lookup_commit_reference
commit: add repository argument to lookup_commit
commit: add repository argument to parse_commit_buffer
commit: add repository argument to set_commit_buffer
commit: add repository argument to get_cached_commit_buffer
tag: add repository argument to lookup_tag
tag: add repository argument to parse_tag_buffer
tag: add repository argument to deref_tag
object: allow object_as_type to handle arbitrary repositories
object: allow lookup_object to handle arbitrary repositories
blob: allow lookup_blob to handle arbitrary repositories
tree: allow lookup_tree to handle arbitrary repositories
commit: allow lookup_commit to handle arbitrary repositories
tag: allow lookup_tag to handle arbitrary repositories
tag: allow parse_tag_buffer to handle arbitrary repositories
commit.c: allow parse_commit_buffer to handle arbitrary repositories
commit-slabs: remove realloc counter outside of slab struct
commit.c: migrate the commit buffer to the parsed object store
commit.c: allow set_commit_buffer to handle arbitrary repositories
commit.c: allow get_cached_commit_buffer to handle arbitrary repositories
object.c: allow parse_object_buffer to handle arbitrary repositories
object.c: allow parse_object to handle arbitrary repositories
tag.c: allow deref_tag to handle arbitrary repositories
commit.c: allow lookup_commit_reference_gently to handle arbitrary repositories
commit.c: allow lookup_commit_reference to handle arbitrary repositories
xdiff/xdiff.h: remove unused flags
xdiff/xdiffi.c: remove unneeded function declarations
t4015: avoid git as a pipe input
diff.c: do not pass diff options as keydata to hashmap
diff.c: adjust hash function signature to match hashmap expectation
diff.c: add a blocks mode for moved code detection
diff.c: decouple white space treatment from move detection algorithm
diff.c: factor advance_or_nullify out of mark_color_as_moved
diff.c: add white space mode to move detection that allows indent changes
diff.c: offer config option to control ws handling in move detection
xdiff/xhistogram: pass arguments directly to fall_back_to_classic_diff
xdiff/xhistogram: factor out memory cleanup into free_index()
xdiff/xhistogram: move index allocation into find_lcs
Documentation/git-interpret-trailers: explain possible values
xdiff/histogram: remove tail recursion
t1300: document current behavior of setting options
xdiff: reduce indent heuristic overhead
config: fix case sensitive subsection names on writing
git-config: document accidental multi-line setting in deprecated syntax
git-submodule.sh: accept verbose flag in cmd_update to be non-quiet
t7410: update to new style
builtin/submodule--helper: remove stray new line
Taylor Blau (9):
Documentation/config.txt: camel-case lineNumber for consistency
grep.c: expose {,inverted} match column in match_line()
grep.[ch]: extend grep_opt to allow showing matched column
grep.c: display column number of first match
builtin/grep.c: add '--column' option to 'git-grep(1)'
grep.c: add configuration variables to show matched option
contrib/git-jump/git-jump: jump to exact location
grep.c: extract show_line_header()
grep.c: teach 'git grep --only-matching'
Thomas Rast (1):
range-diff: add tests
Tobias Klauser (1):
git-rebase--preserve-merges: fix formatting of todo help message
Todd Zullinger (4):
git-credential-netrc: minor whitespace cleanup in test script
git-credential-netrc: make "all" default target of Makefile
gitignore.txt: clarify default core.excludesfile path
dir.c: fix typos in core.excludesfile comment
Trần Ngọc Quân (1):
l10n: vi.po(3958t): updated Vietnamese translation v2.19.0 round 2
Ville Skyttä (1):
Documentation: spelling and grammar fixes
Vladimir Parfinenko (1):
rebase: fix documentation formatting
William Chargin (2):
sha1-name.c: for ":/", find detached HEAD commits
t: factor out FUNNYNAMES as shared lazy prereq
Xiaolong Ye (1):
format-patch: clear UNINTERESTING flag before prepare_bases
brian m. carlson (21):
send-email: add an auto option for transfer encoding
send-email: accept long lines with suitable transfer encoding
send-email: automatically determine transfer-encoding
docs: correct RFC specifying email line length
sequencer: pass absolute GIT_WORK_TREE to exec commands
cache: update object ID functions for the_hash_algo
tree-walk: replace hard-coded constants with the_hash_algo
hex: switch to using the_hash_algo
commit: express tree entry constants in terms of the_hash_algo
strbuf: allocate space with GIT_MAX_HEXSZ
sha1-name: use the_hash_algo when parsing object names
refs/files-backend: use the_hash_algo for writing refs
builtin/update-index: convert to using the_hash_algo
builtin/update-index: simplify parsing of cacheinfo
builtin/fmt-merge-msg: make hash independent
builtin/merge: switch to use the_hash_algo
builtin/merge-recursive: make hash independent
diff: switch GIT_SHA1_HEXSZ to use the_hash_algo
log-tree: switch GIT_SHA1_HEXSZ to the_hash_algo->hexsz
sha1-file: convert constants to uses of the_hash_algo
pretty: switch hard-coded constants to the_hash_algo
Ævar Arnfjörð Bjarmason (45):
checkout tests: index should be clean after dwim checkout
checkout.h: wrap the arguments to unique_tracking_name()
checkout.c: introduce an *_INIT macro
checkout.c: change "unique" member to "num_matches"
checkout: pass the "num_matches" up to callers
builtin/checkout.c: use "ret" variable for return
checkout: add advice for ambiguous "checkout <branch>"
checkout & worktree: introduce checkout.defaultRemote
refspec: s/refspec_item_init/&_or_die/g
refspec: add back a refspec_item_init() function
doc hash-function-transition: note the lack of a changelog
receive.fsck.<msg-id> tests: remove dead code
config doc: don't describe *.fetchObjects twice
config doc: unify the description of fsck.* and receive.fsck.*
config doc: elaborate on what transfer.fsckObjects does
config doc: elaborate on fetch.fsckObjects security
transfer.fsckObjects tests: untangle confusing setup
fetch: implement fetch.fsck.*
fsck: test & document {fetch,receive}.fsck.* config fallback
fsck: add stress tests for fsck.skipList
fsck: test and document unknown fsck.<msg-id> values
tests: make use of the test_must_be_empty function
tests: make use of the test_must_be_empty function
fetch tests: change "Tag" test tag to "testTag"
push tests: remove redundant 'git push' invocation
push tests: fix logic error in "push" test assertion
push tests: add more testing for forced tag pushing
push tests: assert re-pushing annotated tags
negotiator: unknown fetch.negotiationAlgorithm should error out
fetch doc: cross-link two new negotiation options
sha1dc: update from upstream
push: use PARSE_OPT_LITERAL_ARGHELP instead of unbalanced brackets
fetch tests: correct a comment "remove it" -> "remove them"
pull doc: fix a long-standing grammar error
submodule: add more exhaustive up-path testing
refactor various if (x) FREE_AND_NULL(x) to just FREE_AND_NULL(x)
t2024: mark test using "checkout -p" with PERL prerequisite
tests: fix and add lint for non-portable head -c N
tests: fix and add lint for non-portable seq
tests: fix comment syntax in chainlint.sed for AIX sed
tests: use shorter labels in chainlint.sed for AIX sed
tests: fix version-specific portability issue in Perl JSON
tests: fix and add lint for non-portable grep --file
tests: fix non-portable "${var:-"str"}" construct
tests: fix non-portable iconv invocation
Łukasz Stelmach (1):
completion: complete remote names too
^ permalink raw reply [relevance 3%]
* Re: [PATCH v1] git-mv: allow submodules and fsmonitor to work together
2018-09-10 17:07 6% ` Stefan Beller
@ 2018-09-10 19:38 4% ` Ben Peart
0 siblings, 0 replies; 200+ results
From: Ben Peart @ 2018-09-10 19:38 UTC (permalink / raw)
To: Stefan Beller, Ben Peart
Cc: Ben Peart, Ævar Arnfjörð Bjarmason, git,
Junio C Hamano
On 9/10/2018 1:07 PM, Stefan Beller wrote:
> On Mon, Sep 10, 2018 at 9:29 AM Ben Peart <benpeart@microsoft.com> wrote:
>>
>> It was reported that
>>
>> GIT_FSMONITOR_TEST=$PWD/t7519/fsmonitor-all ./t7411-submodule-config.sh
>>
>> breaks as the fsmonitor data is out of sync with the state of the .gitmodules
>> file. Update is_staging_gitmodules_ok() so that it no longer tells
>> ie_match_stat() to ignore refreshing the fsmonitor data.
>
> Wondering how this came to be,
> 7da9aba4178 (submodule: used correct index in is_staging_gitmodules_ok,
> 2017-12-12) last touched this line, but is unrelated as the fsmonitor
> behavior was
> there before.
>
> Before that, we have 883e248b8a0 (fsmonitor: teach git to optionally utilize a
> file system monitor to speed up detecting new or changed files., 2017-09-22)
> that was written by you, who knows the fsmonitor better than I do (or Brandon
> who wrote the commit referenced above).
>
> Looking through the archive, it seems that we might have more such hidden
> gems?
Fortunately, the only one left is the one in preload_index() which is
what the flag was created to handle so I think we're ok.
>
> https://public-inbox.org/git/f50825a4-fa15-9f28-a079-853e78ee8e2e@gmail.com/
>
> Anyway, I think this is a better fix than what I proposed for sure.
>
> Thanks for looking into this!
>
> Stefan
>
>>
>> Reported-by: Ævar Arnfjörð Bjarmason <avarab@gmail.com>
>> Helped-by: Stefan Beller <sbeller@google.com>
>> Signed-off-by: Ben Peart <benpeart@microsoft.com>
>> ---
>>
>> Notes:
>> Base Ref: v2.19.0-rc2
>> Web-Diff: https://github.com/benpeart/git/commit/ed30e1a885
>> Checkout: git fetch https://github.com/benpeart/git fsmonitor-t7411-v1 && git checkout ed30e1a885
>>
>> submodule.c | 3 +--
>> 1 file changed, 1 insertion(+), 2 deletions(-)
>>
>> diff --git a/submodule.c b/submodule.c
>> index 50cbf5f13e..1e7194af28 100644
>> --- a/submodule.c
>> +++ b/submodule.c
>> @@ -65,8 +65,7 @@ int is_staging_gitmodules_ok(struct index_state *istate)
>> if ((pos >= 0) && (pos < istate->cache_nr)) {
>> struct stat st;
>> if (lstat(GITMODULES_FILE, &st) == 0 &&
>> - ie_match_stat(istate, istate->cache[pos], &st,
>> - CE_MATCH_IGNORE_FSMONITOR) & DATA_CHANGED)
>> + ie_match_stat(istate, istate->cache[pos], &st, 0) & DATA_CHANGED)
>> return 0;
>> }
>>
>>
>> base-commit: c05048d43925ab8edcb36663752c2b4541911231
>> --
>> 2.18.0.windows.1
>>
^ permalink raw reply [relevance 4%]
* Re: [PATCH] Revert "Merge branch 'sb/submodule-core-worktree'" (was Re: Old submodules broken in 2.19rc1 and 2.19rc2)
2018-09-08 18:39 10% ` Johannes Sixt
@ 2018-09-10 17:11 5% ` Stefan Beller
0 siblings, 0 replies; 200+ results
From: Stefan Beller @ 2018-09-10 17:11 UTC (permalink / raw)
To: Johannes Sixt; +Cc: Junio C Hamano, Jonathan Nieder, allan.jensen, git
> >> Like this (generated using "git revert -m1)?
> >
> > OK. Thanks for taking care of it.
Yes that looks good to me, thanks!
>
> Please don't forget to remove the corresponding release notes entry.
Makes sense, too.
Thanks,
Stefan
^ permalink raw reply [relevance 5%]
* Re: [PATCH v1] git-mv: allow submodules and fsmonitor to work together
2018-09-10 16:29 11% ` [PATCH v1] " Ben Peart
@ 2018-09-10 17:07 6% ` Stefan Beller
2018-09-10 19:38 4% ` Ben Peart
0 siblings, 1 reply; 200+ results
From: Stefan Beller @ 2018-09-10 17:07 UTC (permalink / raw)
To: Ben Peart
Cc: Ben Peart, Ben Peart, Ævar Arnfjörð Bjarmason, git,
Junio C Hamano
On Mon, Sep 10, 2018 at 9:29 AM Ben Peart <benpeart@microsoft.com> wrote:
>
> It was reported that
>
> GIT_FSMONITOR_TEST=$PWD/t7519/fsmonitor-all ./t7411-submodule-config.sh
>
> breaks as the fsmonitor data is out of sync with the state of the .gitmodules
> file. Update is_staging_gitmodules_ok() so that it no longer tells
> ie_match_stat() to ignore refreshing the fsmonitor data.
Wondering how this came to be,
7da9aba4178 (submodule: used correct index in is_staging_gitmodules_ok,
2017-12-12) last touched this line, but is unrelated as the fsmonitor
behavior was
there before.
Before that, we have 883e248b8a0 (fsmonitor: teach git to optionally utilize a
file system monitor to speed up detecting new or changed files., 2017-09-22)
that was written by you, who knows the fsmonitor better than I do (or Brandon
who wrote the commit referenced above).
Looking through the archive, it seems that we might have more such hidden
gems?
https://public-inbox.org/git/f50825a4-fa15-9f28-a079-853e78ee8e2e@gmail.com/
Anyway, I think this is a better fix than what I proposed for sure.
Thanks for looking into this!
Stefan
>
> Reported-by: Ævar Arnfjörð Bjarmason <avarab@gmail.com>
> Helped-by: Stefan Beller <sbeller@google.com>
> Signed-off-by: Ben Peart <benpeart@microsoft.com>
> ---
>
> Notes:
> Base Ref: v2.19.0-rc2
> Web-Diff: https://github.com/benpeart/git/commit/ed30e1a885
> Checkout: git fetch https://github.com/benpeart/git fsmonitor-t7411-v1 && git checkout ed30e1a885
>
> submodule.c | 3 +--
> 1 file changed, 1 insertion(+), 2 deletions(-)
>
> diff --git a/submodule.c b/submodule.c
> index 50cbf5f13e..1e7194af28 100644
> --- a/submodule.c
> +++ b/submodule.c
> @@ -65,8 +65,7 @@ int is_staging_gitmodules_ok(struct index_state *istate)
> if ((pos >= 0) && (pos < istate->cache_nr)) {
> struct stat st;
> if (lstat(GITMODULES_FILE, &st) == 0 &&
> - ie_match_stat(istate, istate->cache[pos], &st,
> - CE_MATCH_IGNORE_FSMONITOR) & DATA_CHANGED)
> + ie_match_stat(istate, istate->cache[pos], &st, 0) & DATA_CHANGED)
> return 0;
> }
>
>
> base-commit: c05048d43925ab8edcb36663752c2b4541911231
> --
> 2.18.0.windows.1
>
^ permalink raw reply [relevance 6%]
* [PATCH v1] git-mv: allow submodules and fsmonitor to work together
2018-09-10 15:58 10% ` Ben Peart
@ 2018-09-10 16:29 11% ` Ben Peart
2018-09-10 17:07 6% ` Stefan Beller
0 siblings, 1 reply; 200+ results
From: Ben Peart @ 2018-09-10 16:29 UTC (permalink / raw)
To: peartben@gmail.com
Cc: Ben Peart, avarab@gmail.com, git@vger.kernel.org,
gitster@pobox.com, sbeller@google.com, Ben Peart
It was reported that
GIT_FSMONITOR_TEST=$PWD/t7519/fsmonitor-all ./t7411-submodule-config.sh
breaks as the fsmonitor data is out of sync with the state of the .gitmodules
file. Update is_staging_gitmodules_ok() so that it no longer tells
ie_match_stat() to ignore refreshing the fsmonitor data.
Reported-by: Ævar Arnfjörð Bjarmason <avarab@gmail.com>
Helped-by: Stefan Beller <sbeller@google.com>
Signed-off-by: Ben Peart <benpeart@microsoft.com>
---
Notes:
Base Ref: v2.19.0-rc2
Web-Diff: https://github.com/benpeart/git/commit/ed30e1a885
Checkout: git fetch https://github.com/benpeart/git fsmonitor-t7411-v1 && git checkout ed30e1a885
submodule.c | 3 +--
1 file changed, 1 insertion(+), 2 deletions(-)
diff --git a/submodule.c b/submodule.c
index 50cbf5f13e..1e7194af28 100644
--- a/submodule.c
+++ b/submodule.c
@@ -65,8 +65,7 @@ int is_staging_gitmodules_ok(struct index_state *istate)
if ((pos >= 0) && (pos < istate->cache_nr)) {
struct stat st;
if (lstat(GITMODULES_FILE, &st) == 0 &&
- ie_match_stat(istate, istate->cache[pos], &st,
- CE_MATCH_IGNORE_FSMONITOR) & DATA_CHANGED)
+ ie_match_stat(istate, istate->cache[pos], &st, 0) & DATA_CHANGED)
return 0;
}
base-commit: c05048d43925ab8edcb36663752c2b4541911231
--
2.18.0.windows.1
^ permalink raw reply related [relevance 11%]
* Re: [PATCH] git-mv: allow submodules and fsmonitor to work together
@ 2018-09-10 15:58 10% ` Ben Peart
2018-09-10 16:29 11% ` [PATCH v1] " Ben Peart
0 siblings, 1 reply; 200+ results
From: Ben Peart @ 2018-09-10 15:58 UTC (permalink / raw)
To: Stefan Beller, git; +Cc: Ben.Peart, avarab, gitster
On 9/6/2018 4:34 PM, Stefan Beller wrote:
> It was reported that
>
> GIT_FSMONITOR_TEST=$PWD/t7519/fsmonitor-all ./t7411-submodule-config.sh
>
> breaks as the .gitmodules file is modified and staged after the fsmonitor
> considers it clean. Mark the .gitmodules file to be not clean before
> staging.
>
> Reported-by: Ævar Arnfjörð Bjarmason <avarab@gmail.com>
> Inspired-by: Ben Peart <benpeart@microsoft.com>
> Signed-off-by: Stefan Beller <sbeller@google.com>
> ---
>
> I am not quite sure if this is the correct approach and handling of the
> fsmonitor API, but it unbreaks the test.
>
>> Just naively adding mark_fsmonitor_invalid doesn't work, as then ...
>
> Adding it before the staging, works.
>
> Please double check!
I took a look at this bug/patch and wondered why add_file_to_index()
wasn't properly handling the .gitmodules file. On investigation, I
chased it down to what looks like a faulty test in
is_staging_gitmodules_ok().
I believe the following is a better patch for this bug:
diff --git a/submodule.c b/submodule.c
index 50cbf5f13e..1e7194af28 100644
--- a/submodule.c
+++ b/submodule.c
@@ -65,8 +65,7 @@ int is_staging_gitmodules_ok(struct index_state *istate)
if ((pos >= 0) && (pos < istate->cache_nr)) {
struct stat st;
if (lstat(GITMODULES_FILE, &st) == 0 &&
- ie_match_stat(istate, istate->cache[pos], &st,
- CE_MATCH_IGNORE_FSMONITOR) & DATA_CHANGED)
+ ie_match_stat(istate, istate->cache[pos], &st, 0) &
DATA_CHANGED)
return 0;
}
Please double check but I just don't understand why the .gitmodules file
should force the fsmonitor data to be ignored. This flag was added to
enable proper behavior in the preload_thread() logic and I don't believe
it is appropriate here.
Ben
>
> Thanks,
> Stefan
>
> submodule.c | 10 ++++++++++
> 1 file changed, 10 insertions(+)
>
> diff --git a/submodule.c b/submodule.c
> index 50cbf5f13ed..56b0d5fe24e 100644
> --- a/submodule.c
> +++ b/submodule.c
> @@ -22,6 +22,7 @@
> #include "worktree.h"
> #include "parse-options.h"
> #include "object-store.h"
> +#include "fsmonitor.h"
>
> static int config_update_recurse_submodules = RECURSE_SUBMODULES_OFF;
> static struct string_list changed_submodule_names = STRING_LIST_INIT_DUP;
> @@ -149,6 +150,15 @@ int remove_path_from_gitmodules(const char *path)
>
> void stage_updated_gitmodules(struct index_state *istate)
> {
> + struct cache_entry *ce;
> + int pos;
> +
> + pos = index_name_pos(istate, GITMODULES_FILE, strlen(GITMODULES_FILE));
> + ce = (0 <= pos) ? istate->cache[pos] : NULL;
> +
> + if (ce)
> + mark_fsmonitor_invalid(istate, ce);
> +
> if (add_file_to_index(istate, GITMODULES_FILE, 0))
> die(_("staging updated .gitmodules failed"));
> }
>
^ permalink raw reply related [relevance 10%]
* Re: [PATCH] Revert "Merge branch 'sb/submodule-core-worktree'" (was Re: Old submodules broken in 2.19rc1 and 2.19rc2)
2018-09-08 2:04 2% ` Junio C Hamano
@ 2018-09-08 18:39 10% ` Johannes Sixt
2018-09-10 17:11 5% ` Stefan Beller
0 siblings, 1 reply; 200+ results
From: Johannes Sixt @ 2018-09-08 18:39 UTC (permalink / raw)
To: Junio C Hamano, Jonathan Nieder; +Cc: Stefan Beller, allan.jensen, git
Am 08.09.2018 um 04:04 schrieb Junio C Hamano:
> Jonathan Nieder <jrnieder@gmail.com> writes:
>
>> It is late in the release cycle, so revert the whole 3-patch series.
>> We can try again later for 2.20.
>>
>> Reported-by: Allan Sandfeld Jensen <allan.jensen@qt.io>
>> Helped-by: Stefan Beller <sbeller@google.com>
>> Signed-off-by: Jonathan Nieder <jrnieder@gmail.com>
>> ---
>> Stefan Beller wrote:
>>> Jonathan Nieder wrote:
>>
>>>> I think we
>>>> should revert e98317508c0 in "master" (for 2.19) and keep making use
>>>> of that 'second try' in "next" (for 2.20).
>>>
>>> Actually I'd rather revert the whole topic leading up to
>>> 7e25437d35a (Merge branch 'sb/submodule-core-worktree', 2018-07-18)
>>> as the last patch in there doesn't work well without e98317508c0 IIRC.
>>>
>>> And having only the first patch would bring an inconsistent state as
>>> then different commands behave differently w.r.t. setting core.worktree.
>>
>> Like this (generated using "git revert -m1)?
>
> OK. Thanks for taking care of it.
Please don't forget to remove the corresponding release notes entry.
diff --git a/Documentation/RelNotes/2.19.0.txt b/Documentation/RelNotes/2.19.0.txt
index bcbfbc2041..834454ffb9 100644
--- a/Documentation/RelNotes/2.19.0.txt
+++ b/Documentation/RelNotes/2.19.0.txt
@@ -296,12 +296,6 @@ Fixes since v2.18
to the submodule was changed in the range of commits in the
superproject, sometimes showing "(null)". This has been corrected.
- * "git submodule" did not correctly adjust core.worktree setting that
- indicates whether/where a submodule repository has its associated
- working tree across various state transitions, which has been
- corrected.
- (merge 984cd77ddb sb/submodule-core-worktree later to maint).
-
* Bugfix for "rebase -i" corner case regression.
(merge a9279c6785 pw/rebase-i-keep-reword-after-conflict later to maint).
^ permalink raw reply related [relevance 10%]
* Re: [PATCH] Revert "Merge branch 'sb/submodule-core-worktree'" (was Re: Old submodules broken in 2.19rc1 and 2.19rc2)
2018-09-08 0:09 22% ` [PATCH] Revert "Merge branch 'sb/submodule-core-worktree'" (was Re: Old submodules broken in 2.19rc1 and 2.19rc2) Jonathan Nieder
@ 2018-09-08 2:04 2% ` Junio C Hamano
2018-09-08 18:39 10% ` Johannes Sixt
0 siblings, 1 reply; 200+ results
From: Junio C Hamano @ 2018-09-08 2:04 UTC (permalink / raw)
To: Jonathan Nieder; +Cc: Stefan Beller, allan.jensen, git
Jonathan Nieder <jrnieder@gmail.com> writes:
> It is late in the release cycle, so revert the whole 3-patch series.
> We can try again later for 2.20.
>
> Reported-by: Allan Sandfeld Jensen <allan.jensen@qt.io>
> Helped-by: Stefan Beller <sbeller@google.com>
> Signed-off-by: Jonathan Nieder <jrnieder@gmail.com>
> ---
> Stefan Beller wrote:
>> Jonathan Nieder wrote:
>
>>> I think we
>>> should revert e98317508c0 in "master" (for 2.19) and keep making use
>>> of that 'second try' in "next" (for 2.20).
>>
>> Actually I'd rather revert the whole topic leading up to
>> 7e25437d35a (Merge branch 'sb/submodule-core-worktree', 2018-07-18)
>> as the last patch in there doesn't work well without e98317508c0 IIRC.
>>
>> And having only the first patch would bring an inconsistent state as
>> then different commands behave differently w.r.t. setting core.worktree.
>
> Like this (generated using "git revert -m1)?
OK. Thanks for taking care of it.
>
> builtin/submodule--helper.c | 26 --------------------------
> git-submodule.sh | 5 -----
> submodule.c | 14 --------------
> submodule.h | 2 --
> t/lib-submodule-update.sh | 5 ++---
> t/t7400-submodule-basic.sh | 5 -----
> 6 files changed, 2 insertions(+), 55 deletions(-)
>
> diff --git a/builtin/submodule--helper.c b/builtin/submodule--helper.c
> index b56028ba9d..f6fb8991f3 100644
> --- a/builtin/submodule--helper.c
> +++ b/builtin/submodule--helper.c
> @@ -1123,8 +1123,6 @@ static void deinit_submodule(const char *path, const char *prefix,
> if (!(flags & OPT_QUIET))
> printf(format, displaypath);
>
> - submodule_unset_core_worktree(sub);
> -
> strbuf_release(&sb_rm);
> }
>
> @@ -2005,29 +2003,6 @@ static int check_name(int argc, const char **argv, const char *prefix)
> return 0;
> }
>
> -static int connect_gitdir_workingtree(int argc, const char **argv, const char *prefix)
> -{
> - struct strbuf sb = STRBUF_INIT;
> - const char *name, *path;
> - char *sm_gitdir;
> -
> - if (argc != 3)
> - BUG("submodule--helper connect-gitdir-workingtree <name> <path>");
> -
> - name = argv[1];
> - path = argv[2];
> -
> - strbuf_addf(&sb, "%s/modules/%s", get_git_dir(), name);
> - sm_gitdir = absolute_pathdup(sb.buf);
> -
> - connect_work_tree_and_git_dir(path, sm_gitdir, 0);
> -
> - strbuf_release(&sb);
> - free(sm_gitdir);
> -
> - return 0;
> -}
> -
> #define SUPPORT_SUPER_PREFIX (1<<0)
>
> struct cmd_struct {
> @@ -2041,7 +2016,6 @@ static struct cmd_struct commands[] = {
> {"name", module_name, 0},
> {"clone", module_clone, 0},
> {"update-clone", update_clone, 0},
> - {"connect-gitdir-workingtree", connect_gitdir_workingtree, 0},
> {"relative-path", resolve_relative_path, 0},
> {"resolve-relative-url", resolve_relative_url, 0},
> {"resolve-relative-url-test", resolve_relative_url_test, 0},
> diff --git a/git-submodule.sh b/git-submodule.sh
> index f7fd80345c..1cb2c0a31b 100755
> --- a/git-submodule.sh
> +++ b/git-submodule.sh
> @@ -580,11 +580,6 @@ cmd_update()
> die "$(eval_gettext "Unable to find current \${remote_name}/\${branch} revision in submodule path '\$sm_path'")"
> fi
>
> - if ! $(git config -f "$(git rev-parse --git-common-dir)/modules/$name/config" core.worktree) 2>/dev/null
> - then
> - git submodule--helper connect-gitdir-workingtree "$name" "$sm_path"
> - fi
> -
> if test "$subsha1" != "$sha1" || test -n "$force"
> then
> subforce=$force
> diff --git a/submodule.c b/submodule.c
> index 50cbf5f13e..a2b266fbfa 100644
> --- a/submodule.c
> +++ b/submodule.c
> @@ -1534,18 +1534,6 @@ int bad_to_remove_submodule(const char *path, unsigned flags)
> return ret;
> }
>
> -void submodule_unset_core_worktree(const struct submodule *sub)
> -{
> - char *config_path = xstrfmt("%s/modules/%s/config",
> - get_git_common_dir(), sub->name);
> -
> - if (git_config_set_in_file_gently(config_path, "core.worktree", NULL))
> - warning(_("Could not unset core.worktree setting in submodule '%s'"),
> - sub->path);
> -
> - free(config_path);
> -}
> -
> static const char *get_super_prefix_or_empty(void)
> {
> const char *s = get_super_prefix();
> @@ -1711,8 +1699,6 @@ int submodule_move_head(const char *path,
>
> if (is_empty_dir(path))
> rmdir_or_warn(path);
> -
> - submodule_unset_core_worktree(sub);
> }
> }
> out:
> diff --git a/submodule.h b/submodule.h
> index 7d476cefa7..e452919aa4 100644
> --- a/submodule.h
> +++ b/submodule.h
> @@ -127,8 +127,6 @@ int submodule_move_head(const char *path,
> const char *new_head,
> unsigned flags);
>
> -void submodule_unset_core_worktree(const struct submodule *sub);
> -
> /*
> * Prepare the "env_array" parameter of a "struct child_process" for executing
> * a submodule by clearing any repo-specific environment variables, but
> diff --git a/t/lib-submodule-update.sh b/t/lib-submodule-update.sh
> index 5b56b23166..016391723c 100755
> --- a/t/lib-submodule-update.sh
> +++ b/t/lib-submodule-update.sh
> @@ -235,7 +235,7 @@ reset_work_tree_to_interested () {
> then
> mkdir -p submodule_update/.git/modules/sub1/modules &&
> cp -r submodule_update_repo/.git/modules/sub1/modules/sub2 submodule_update/.git/modules/sub1/modules/sub2
> - # core.worktree is unset for sub2 as it is not checked out
> + GIT_WORK_TREE=. git -C submodule_update/.git/modules/sub1/modules/sub2 config --unset core.worktree
> fi &&
> # indicate we are interested in the submodule:
> git -C submodule_update config submodule.sub1.url "bogus" &&
> @@ -709,8 +709,7 @@ test_submodule_recursing_with_args_common() {
> git branch -t remove_sub1 origin/remove_sub1 &&
> $command remove_sub1 &&
> test_superproject_content origin/remove_sub1 &&
> - ! test -e sub1 &&
> - test_must_fail git config -f .git/modules/sub1/config core.worktree
> + ! test -e sub1
> )
> '
> # ... absorbing a .git directory along the way.
> diff --git a/t/t7400-submodule-basic.sh b/t/t7400-submodule-basic.sh
> index 7d3d984210..c0ffc1022a 100755
> --- a/t/t7400-submodule-basic.sh
> +++ b/t/t7400-submodule-basic.sh
> @@ -984,11 +984,6 @@ test_expect_success 'submodule deinit should remove the whole submodule section
> rmdir init
> '
>
> -test_expect_success 'submodule deinit should unset core.worktree' '
> - test_path_is_file .git/modules/example/config &&
> - test_must_fail git config -f .git/modules/example/config core.worktree
> -'
> -
> test_expect_success 'submodule deinit from subdirectory' '
> git submodule update --init &&
> git config submodule.example.foo bar &&
^ permalink raw reply [relevance 2%]
* [PATCH] Revert "Merge branch 'sb/submodule-core-worktree'" (was Re: Old submodules broken in 2.19rc1 and 2.19rc2)
2018-09-07 22:45 7% ` Stefan Beller
@ 2018-09-08 0:09 22% ` Jonathan Nieder
2018-09-08 2:04 2% ` Junio C Hamano
0 siblings, 1 reply; 200+ results
From: Jonathan Nieder @ 2018-09-08 0:09 UTC (permalink / raw)
To: Stefan Beller; +Cc: allan.jensen, git, Junio C Hamano
Subject: Revert "Merge branch 'sb/submodule-core-worktree'"
This reverts commit 7e25437d35a70791b345872af202eabfb3e1a8bc, reversing
changes made to 00624d608cc69bd62801c93e74d1ea7a7ddd6598.
v2.19.0-rc0~165^2~1 (submodule: ensure core.worktree is set after
update, 2018-06-18) assumes an "absorbed" submodule layout, where the
submodule's Git directory is in the superproject's .git/modules/
directory and .git in the submodule worktree is a .git file pointing
there. In particular, it uses $GIT_DIR/modules/$name to find the
submodule to find out whether it already has core.worktree set, and it
uses connect_work_tree_and_git_dir if not, resulting in
fatal: could not open sub/.git for writing
The context behind that patch: v2.19.0-rc0~165^2~2 (submodule: unset
core.worktree if no working tree is present, 2018-06-12) unsets
core.worktree when running commands like "git checkout
--recurse-submodules" to switch to a branch without the submodule. If
a user then uses "git checkout --no-recurse-submodules" to switch back
to a branch with the submodule and runs "git submodule update", this
patch is needed to ensure that commands using the submodule directly
are aware of the path to the worktree.
It is late in the release cycle, so revert the whole 3-patch series.
We can try again later for 2.20.
Reported-by: Allan Sandfeld Jensen <allan.jensen@qt.io>
Helped-by: Stefan Beller <sbeller@google.com>
Signed-off-by: Jonathan Nieder <jrnieder@gmail.com>
---
Stefan Beller wrote:
> Jonathan Nieder wrote:
>> I think we
>> should revert e98317508c0 in "master" (for 2.19) and keep making use
>> of that 'second try' in "next" (for 2.20).
>
> Actually I'd rather revert the whole topic leading up to
> 7e25437d35a (Merge branch 'sb/submodule-core-worktree', 2018-07-18)
> as the last patch in there doesn't work well without e98317508c0 IIRC.
>
> And having only the first patch would bring an inconsistent state as
> then different commands behave differently w.r.t. setting core.worktree.
Like this (generated using "git revert -m1)?
builtin/submodule--helper.c | 26 --------------------------
git-submodule.sh | 5 -----
submodule.c | 14 --------------
submodule.h | 2 --
t/lib-submodule-update.sh | 5 ++---
t/t7400-submodule-basic.sh | 5 -----
6 files changed, 2 insertions(+), 55 deletions(-)
diff --git a/builtin/submodule--helper.c b/builtin/submodule--helper.c
index b56028ba9d..f6fb8991f3 100644
--- a/builtin/submodule--helper.c
+++ b/builtin/submodule--helper.c
@@ -1123,8 +1123,6 @@ static void deinit_submodule(const char *path, const char *prefix,
if (!(flags & OPT_QUIET))
printf(format, displaypath);
- submodule_unset_core_worktree(sub);
-
strbuf_release(&sb_rm);
}
@@ -2005,29 +2003,6 @@ static int check_name(int argc, const char **argv, const char *prefix)
return 0;
}
-static int connect_gitdir_workingtree(int argc, const char **argv, const char *prefix)
-{
- struct strbuf sb = STRBUF_INIT;
- const char *name, *path;
- char *sm_gitdir;
-
- if (argc != 3)
- BUG("submodule--helper connect-gitdir-workingtree <name> <path>");
-
- name = argv[1];
- path = argv[2];
-
- strbuf_addf(&sb, "%s/modules/%s", get_git_dir(), name);
- sm_gitdir = absolute_pathdup(sb.buf);
-
- connect_work_tree_and_git_dir(path, sm_gitdir, 0);
-
- strbuf_release(&sb);
- free(sm_gitdir);
-
- return 0;
-}
-
#define SUPPORT_SUPER_PREFIX (1<<0)
struct cmd_struct {
@@ -2041,7 +2016,6 @@ static struct cmd_struct commands[] = {
{"name", module_name, 0},
{"clone", module_clone, 0},
{"update-clone", update_clone, 0},
- {"connect-gitdir-workingtree", connect_gitdir_workingtree, 0},
{"relative-path", resolve_relative_path, 0},
{"resolve-relative-url", resolve_relative_url, 0},
{"resolve-relative-url-test", resolve_relative_url_test, 0},
diff --git a/git-submodule.sh b/git-submodule.sh
index f7fd80345c..1cb2c0a31b 100755
--- a/git-submodule.sh
+++ b/git-submodule.sh
@@ -580,11 +580,6 @@ cmd_update()
die "$(eval_gettext "Unable to find current \${remote_name}/\${branch} revision in submodule path '\$sm_path'")"
fi
- if ! $(git config -f "$(git rev-parse --git-common-dir)/modules/$name/config" core.worktree) 2>/dev/null
- then
- git submodule--helper connect-gitdir-workingtree "$name" "$sm_path"
- fi
-
if test "$subsha1" != "$sha1" || test -n "$force"
then
subforce=$force
diff --git a/submodule.c b/submodule.c
index 50cbf5f13e..a2b266fbfa 100644
--- a/submodule.c
+++ b/submodule.c
@@ -1534,18 +1534,6 @@ int bad_to_remove_submodule(const char *path, unsigned flags)
return ret;
}
-void submodule_unset_core_worktree(const struct submodule *sub)
-{
- char *config_path = xstrfmt("%s/modules/%s/config",
- get_git_common_dir(), sub->name);
-
- if (git_config_set_in_file_gently(config_path, "core.worktree", NULL))
- warning(_("Could not unset core.worktree setting in submodule '%s'"),
- sub->path);
-
- free(config_path);
-}
-
static const char *get_super_prefix_or_empty(void)
{
const char *s = get_super_prefix();
@@ -1711,8 +1699,6 @@ int submodule_move_head(const char *path,
if (is_empty_dir(path))
rmdir_or_warn(path);
-
- submodule_unset_core_worktree(sub);
}
}
out:
diff --git a/submodule.h b/submodule.h
index 7d476cefa7..e452919aa4 100644
--- a/submodule.h
+++ b/submodule.h
@@ -127,8 +127,6 @@ int submodule_move_head(const char *path,
const char *new_head,
unsigned flags);
-void submodule_unset_core_worktree(const struct submodule *sub);
-
/*
* Prepare the "env_array" parameter of a "struct child_process" for executing
* a submodule by clearing any repo-specific environment variables, but
diff --git a/t/lib-submodule-update.sh b/t/lib-submodule-update.sh
index 5b56b23166..016391723c 100755
--- a/t/lib-submodule-update.sh
+++ b/t/lib-submodule-update.sh
@@ -235,7 +235,7 @@ reset_work_tree_to_interested () {
then
mkdir -p submodule_update/.git/modules/sub1/modules &&
cp -r submodule_update_repo/.git/modules/sub1/modules/sub2 submodule_update/.git/modules/sub1/modules/sub2
- # core.worktree is unset for sub2 as it is not checked out
+ GIT_WORK_TREE=. git -C submodule_update/.git/modules/sub1/modules/sub2 config --unset core.worktree
fi &&
# indicate we are interested in the submodule:
git -C submodule_update config submodule.sub1.url "bogus" &&
@@ -709,8 +709,7 @@ test_submodule_recursing_with_args_common() {
git branch -t remove_sub1 origin/remove_sub1 &&
$command remove_sub1 &&
test_superproject_content origin/remove_sub1 &&
- ! test -e sub1 &&
- test_must_fail git config -f .git/modules/sub1/config core.worktree
+ ! test -e sub1
)
'
# ... absorbing a .git directory along the way.
diff --git a/t/t7400-submodule-basic.sh b/t/t7400-submodule-basic.sh
index 7d3d984210..c0ffc1022a 100755
--- a/t/t7400-submodule-basic.sh
+++ b/t/t7400-submodule-basic.sh
@@ -984,11 +984,6 @@ test_expect_success 'submodule deinit should remove the whole submodule section
rmdir init
'
-test_expect_success 'submodule deinit should unset core.worktree' '
- test_path_is_file .git/modules/example/config &&
- test_must_fail git config -f .git/modules/example/config core.worktree
-'
-
test_expect_success 'submodule deinit from subdirectory' '
git submodule update --init &&
git config submodule.example.foo bar &&
--
2.19.0.rc2.392.g5ba43deb5a
^ permalink raw reply related [relevance 22%]
* Re: Old submodules broken in 2.19rc1 and 2.19rc2
2018-09-07 22:35 7% ` Jonathan Nieder
@ 2018-09-07 22:45 7% ` Stefan Beller
2018-09-08 0:09 22% ` [PATCH] Revert "Merge branch 'sb/submodule-core-worktree'" (was Re: Old submodules broken in 2.19rc1 and 2.19rc2) Jonathan Nieder
0 siblings, 1 reply; 200+ results
From: Stefan Beller @ 2018-09-07 22:45 UTC (permalink / raw)
To: Jonathan Nieder; +Cc: allan.jensen, git
I think we
> should revert e98317508c0 in "master" (for 2.19) and keep making use
> of that 'second try' in "next" (for 2.20).
Actually I'd rather revert the whole topic leading up to
7e25437d35a (Merge branch 'sb/submodule-core-worktree', 2018-07-18)
as the last patch in there doesn't work well without e98317508c0 IIRC.
And having only the first patch would bring an inconsistent state as
then different commands behave differently w.r.t. setting core.worktree.
So for this release we'd
git revert 984cd77ddbf0 e98317508c 4fa4f90ccd85
and then can
git cherry-pick 4fa4f90ccd8 984cd77ddbf0
on top of sb/submodule-update-in-c, as that re-instates the behavior
going forward.
Thoughts?
Thanks,
Stefan
^ permalink raw reply [relevance 7%]
* Re: Old submodules broken in 2.19rc1 and 2.19rc2
2018-09-07 17:08 7% ` Stefan Beller
2018-09-07 20:20 6% ` Jonathan Nieder
@ 2018-09-07 22:33 3% ` Allan Sandfeld Jensen
2018-09-07 22:35 7% ` Jonathan Nieder
2 siblings, 0 replies; 200+ results
From: Allan Sandfeld Jensen @ 2018-09-07 22:33 UTC (permalink / raw)
To: Stefan Beller; +Cc: git
[-- Attachment #1: Type: text/plain, Size: 1272 bytes --]
On Freitag, 7. September 2018 19:08:43 CEST Stefan Beller wrote:
> On Fri, Sep 7, 2018 at 2:53 AM Allan Sandfeld Jensen <allan.jensen@qt.io>
wrote:
> > Submodules checked out with older versions of git not longer works in the
> > latest 2.19 releases. A "git submodule update --recursive" command wil
> > fail
> > for each submodule with a line saying "fatal: could not open
> > '<submodule>/.git' for writing> Is a directory.
>
> Can you run the update again with
>
> GIT_TRACE=1 git submodule update ....
>
> and post the output?
>
> I have the suspicion that e98317508c0 (submodule:
> ensure core.worktree is set after update, 2018-06-18)
> might be the offender.
>
> Could you try reverting that commit and check as well?
>
> git clone https://github.com/git/git && cd git
> git revert e98317508c0
> make install # installs to you home dir at ~/bin
>
> and then try again, as well?
> (though bisection may be more fruitful if this doesn't pan out)
Okay. I had the issue on my workstation at work which I won't be back to until
friday next week, but I managed to reproduce the exact same issue on separate
machine running Ubuntu, and a freshly built git from git master, on another
roughly one year old checkout of qt5.git with submodules
[-- Attachment #2: git-trace.txt --]
[-- Type: text/plain, Size: 68895 bytes --]
[127] carewolf@twilight% GIT_TRACE=1 ~src/git/git submodule update --recursive [5.11.2] ~qt5
00:28:32.234453 git.c:659 trace: exec: git-submodule update --recursive
00:28:32.234491 run-command.c:637 trace: run_command: git-submodule update --recursive
00:28:32.240792 git.c:415 trace: built-in: git rev-parse --git-dir
00:28:32.241981 git.c:415 trace: built-in: git rev-parse --git-path objects
00:28:32.242905 git.c:415 trace: built-in: git rev-parse -q --git-dir
00:28:32.244794 git.c:415 trace: built-in: git rev-parse --show-prefix
00:28:32.245699 git.c:415 trace: built-in: git rev-parse --show-toplevel
00:28:32.247181 git.c:415 trace: built-in: git submodule--helper update-clone
00:28:32.247463 run-command.c:1553 run_processes_parallel: preparing to run up to 1 tasks
00:28:32.247764 run-command.c:1585 run_processes_parallel: done
00:28:32.248626 git.c:415 trace: built-in: git submodule--helper name qt3d
00:28:32.249751 git.c:415 trace: built-in: git config submodule.qt3d.update
00:28:32.250608 git.c:415 trace: built-in: git submodule--helper relative-path qt3d
00:28:32.251399 git.c:415 trace: built-in: git rev-parse --local-env-vars
00:28:32.252120 git.c:415 trace: built-in: git rev-parse --verify HEAD
00:28:32.253096 git.c:415 trace: built-in: git rev-parse --git-common-dir
00:28:32.253982 git.c:415 trace: built-in: git config -f .git/modules/qt3d/config core.worktree
00:28:32.254771 git.c:415 trace: built-in: git submodule--helper connect-gitdir-workingtree qt3d qt3d
fatal: could not open 'qt3d/.git' for writing: Is a directory
00:28:32.255930 git.c:415 trace: built-in: git submodule--helper relative-path qt3d/
00:28:32.256712 git.c:415 trace: built-in: git rev-parse --local-env-vars
00:28:32.257878 git.c:415 trace: built-in: git submodule--helper update-clone --recursive-prefix qt3d/
00:28:32.258510 run-command.c:1553 run_processes_parallel: preparing to run up to 1 tasks
00:28:32.258544 run-command.c:1585 run_processes_parallel: done
00:28:32.260356 git.c:415 trace: built-in: git submodule--helper name qtactiveqt
00:28:32.261475 git.c:415 trace: built-in: git config submodule.qtactiveqt.update
00:28:32.262321 git.c:415 trace: built-in: git submodule--helper relative-path qtactiveqt
00:28:32.263259 git.c:415 trace: built-in: git rev-parse --local-env-vars
00:28:32.263987 git.c:415 trace: built-in: git rev-parse --verify HEAD
00:28:32.264929 git.c:415 trace: built-in: git rev-parse --git-common-dir
00:28:32.265775 git.c:415 trace: built-in: git config -f .git/modules/qtactiveqt/config core.worktree
00:28:32.266563 git.c:415 trace: built-in: git submodule--helper connect-gitdir-workingtree qtactiveqt qtactiveqt
fatal: could not open 'qtactiveqt/.git' for writing: Is a directory
00:28:32.267595 git.c:415 trace: built-in: git submodule--helper relative-path qtactiveqt/
00:28:32.268310 git.c:415 trace: built-in: git rev-parse --local-env-vars
00:28:32.269481 git.c:415 trace: built-in: git submodule--helper update-clone --recursive-prefix qtactiveqt/
00:28:32.269626 run-command.c:1553 run_processes_parallel: preparing to run up to 1 tasks
00:28:32.269652 run-command.c:1585 run_processes_parallel: done
00:28:32.270893 git.c:415 trace: built-in: git submodule--helper name qtandroidextras
00:28:32.272151 git.c:415 trace: built-in: git config submodule.qtandroidextras.update
00:28:32.273213 git.c:415 trace: built-in: git submodule--helper relative-path qtandroidextras
00:28:32.274162 git.c:415 trace: built-in: git rev-parse --local-env-vars
00:28:32.274856 git.c:415 trace: built-in: git rev-parse --verify HEAD
00:28:32.275879 git.c:415 trace: built-in: git rev-parse --git-common-dir
00:28:32.276995 git.c:415 trace: built-in: git config -f .git/modules/qtandroidextras/config core.worktree
00:28:32.277883 git.c:415 trace: built-in: git submodule--helper connect-gitdir-workingtree qtandroidextras qtandroidextras
fatal: could not open 'qtandroidextras/.git' for writing: Is a directory
00:28:32.278974 git.c:415 trace: built-in: git submodule--helper relative-path qtandroidextras/
00:28:32.279721 git.c:415 trace: built-in: git rev-parse --local-env-vars
00:28:32.280735 git.c:415 trace: built-in: git submodule--helper update-clone --recursive-prefix qtandroidextras/
00:28:32.280854 run-command.c:1553 run_processes_parallel: preparing to run up to 1 tasks
00:28:32.280881 run-command.c:1585 run_processes_parallel: done
00:28:32.281938 git.c:415 trace: built-in: git submodule--helper name qtbase
00:28:32.283009 git.c:415 trace: built-in: git config submodule.qtbase.update
00:28:32.283947 git.c:415 trace: built-in: git submodule--helper relative-path qtbase
00:28:32.284835 git.c:415 trace: built-in: git rev-parse --local-env-vars
00:28:32.285594 git.c:415 trace: built-in: git rev-parse --verify HEAD
00:28:32.286639 git.c:415 trace: built-in: git rev-parse --git-common-dir
00:28:32.287548 git.c:415 trace: built-in: git config -f .git/modules/qtbase/config core.worktree
00:28:32.288399 git.c:415 trace: built-in: git submodule--helper connect-gitdir-workingtree qtbase qtbase
fatal: could not open 'qtbase/.git' for writing: Is a directory
00:28:32.289379 git.c:415 trace: built-in: git submodule--helper relative-path qtbase/
00:28:32.290073 git.c:415 trace: built-in: git rev-parse --local-env-vars
00:28:32.291003 git.c:415 trace: built-in: git submodule--helper update-clone --recursive-prefix qtbase/
00:28:32.293437 run-command.c:1553 run_processes_parallel: preparing to run up to 1 tasks
00:28:32.293450 run-command.c:1585 run_processes_parallel: done
00:28:32.294761 git.c:415 trace: built-in: git submodule--helper name qtcanvas3d
00:28:32.295955 git.c:415 trace: built-in: git config submodule.qtcanvas3d.update
00:28:32.296830 git.c:415 trace: built-in: git submodule--helper relative-path qtcanvas3d
00:28:32.297737 git.c:415 trace: built-in: git rev-parse --local-env-vars
00:28:32.298323 git.c:415 trace: built-in: git rev-parse --verify HEAD
00:28:32.299308 git.c:415 trace: built-in: git rev-parse --git-common-dir
00:28:32.300424 git.c:415 trace: built-in: git config -f .git/modules/qtcanvas3d/config core.worktree
00:28:32.301219 git.c:415 trace: built-in: git submodule--helper connect-gitdir-workingtree qtcanvas3d qtcanvas3d
fatal: could not open 'qtcanvas3d/.git' for writing: Is a directory
00:28:32.302298 git.c:415 trace: built-in: git submodule--helper relative-path qtcanvas3d/
00:28:32.303135 git.c:415 trace: built-in: git rev-parse --local-env-vars
00:28:32.304144 git.c:415 trace: built-in: git submodule--helper update-clone --recursive-prefix qtcanvas3d/
00:28:32.304305 run-command.c:1553 run_processes_parallel: preparing to run up to 1 tasks
00:28:32.304312 run-command.c:1585 run_processes_parallel: done
00:28:32.305436 git.c:415 trace: built-in: git submodule--helper name qtcharts
00:28:32.306623 git.c:415 trace: built-in: git config submodule.qtcharts.update
00:28:32.307733 git.c:415 trace: built-in: git submodule--helper relative-path qtcharts
00:28:32.309132 git.c:415 trace: built-in: git rev-parse --local-env-vars
00:28:32.309953 git.c:415 trace: built-in: git rev-parse --verify HEAD
00:28:32.310872 git.c:415 trace: built-in: git rev-parse --git-common-dir
00:28:32.311715 git.c:415 trace: built-in: git config -f .git/modules/qtcharts/config core.worktree
00:28:32.312694 git.c:415 trace: built-in: git submodule--helper connect-gitdir-workingtree qtcharts qtcharts
fatal: could not open 'qtcharts/.git' for writing: Is a directory
00:28:32.313739 git.c:415 trace: built-in: git submodule--helper relative-path qtcharts/
00:28:32.314476 git.c:415 trace: built-in: git rev-parse --local-env-vars
00:28:32.315474 git.c:415 trace: built-in: git submodule--helper update-clone --recursive-prefix qtcharts/
00:28:32.315706 run-command.c:1553 run_processes_parallel: preparing to run up to 1 tasks
00:28:32.315714 run-command.c:1585 run_processes_parallel: done
00:28:32.317048 git.c:415 trace: built-in: git submodule--helper name qtconnectivity
00:28:32.318220 git.c:415 trace: built-in: git config submodule.qtconnectivity.update
00:28:32.319278 git.c:415 trace: built-in: git submodule--helper relative-path qtconnectivity
00:28:32.320454 git.c:415 trace: built-in: git rev-parse --local-env-vars
00:28:32.321200 git.c:415 trace: built-in: git rev-parse --verify HEAD
00:28:32.322107 git.c:415 trace: built-in: git rev-parse --git-common-dir
00:28:32.323086 git.c:415 trace: built-in: git config -f .git/modules/qtconnectivity/config core.worktree
00:28:32.323983 git.c:415 trace: built-in: git submodule--helper connect-gitdir-workingtree qtconnectivity qtconnectivity
fatal: could not open 'qtconnectivity/.git' for writing: Is a directory
00:28:32.325043 git.c:415 trace: built-in: git submodule--helper relative-path qtconnectivity/
00:28:32.325810 git.c:415 trace: built-in: git rev-parse --local-env-vars
00:28:32.326789 git.c:415 trace: built-in: git submodule--helper update-clone --recursive-prefix qtconnectivity/
00:28:32.326978 run-command.c:1553 run_processes_parallel: preparing to run up to 1 tasks
00:28:32.326987 run-command.c:1585 run_processes_parallel: done
00:28:32.328155 git.c:415 trace: built-in: git submodule--helper name qtdatavis3d
00:28:32.329275 git.c:415 trace: built-in: git config submodule.qtdatavis3d.update
00:28:32.330125 git.c:415 trace: built-in: git submodule--helper relative-path qtdatavis3d
00:28:32.331069 git.c:415 trace: built-in: git rev-parse --local-env-vars
00:28:32.331678 git.c:415 trace: built-in: git rev-parse --verify HEAD
00:28:32.332616 git.c:415 trace: built-in: git rev-parse --git-common-dir
00:28:32.333516 git.c:415 trace: built-in: git config -f .git/modules/qtdatavis3d/config core.worktree
00:28:32.334445 git.c:415 trace: built-in: git submodule--helper connect-gitdir-workingtree qtdatavis3d qtdatavis3d
fatal: could not open 'qtdatavis3d/.git' for writing: Is a directory
00:28:32.335636 git.c:415 trace: built-in: git submodule--helper relative-path qtdatavis3d/
00:28:32.336442 git.c:415 trace: built-in: git rev-parse --local-env-vars
00:28:32.337468 git.c:415 trace: built-in: git submodule--helper update-clone --recursive-prefix qtdatavis3d/
00:28:32.337696 run-command.c:1553 run_processes_parallel: preparing to run up to 1 tasks
00:28:32.337724 run-command.c:1585 run_processes_parallel: done
00:28:32.338780 git.c:415 trace: built-in: git submodule--helper name qtdeclarative
00:28:32.339940 git.c:415 trace: built-in: git config submodule.qtdeclarative.update
00:28:32.340751 git.c:415 trace: built-in: git submodule--helper relative-path qtdeclarative
00:28:32.341537 git.c:415 trace: built-in: git rev-parse --local-env-vars
00:28:32.342183 git.c:415 trace: built-in: git rev-parse --verify HEAD
00:28:32.343111 git.c:415 trace: built-in: git rev-parse --git-common-dir
00:28:32.343997 git.c:415 trace: built-in: git config -f .git/modules/qtdeclarative/config core.worktree
00:28:32.344743 git.c:415 trace: built-in: git submodule--helper connect-gitdir-workingtree qtdeclarative qtdeclarative
fatal: could not open 'qtdeclarative/.git' for writing: Is a directory
00:28:32.345732 git.c:415 trace: built-in: git submodule--helper relative-path qtdeclarative/
00:28:32.346490 git.c:415 trace: built-in: git rev-parse --local-env-vars
00:28:32.347495 git.c:415 trace: built-in: git submodule--helper update-clone --recursive-prefix qtdeclarative/
00:28:32.348619 run-command.c:1553 run_processes_parallel: preparing to run up to 1 tasks
00:28:32.348690 run-command.c:1585 run_processes_parallel: done
00:28:32.349524 git.c:415 trace: built-in: git submodule--helper name tests/auto/qml/ecmascripttests/test262
00:28:32.351213 git.c:415 trace: built-in: git config submodule.tests/auto/qml/ecmascripttests/test262.update
00:28:32.352106 git.c:415 trace: built-in: git submodule--helper relative-path qtdeclarative/tests/auto/qml/ecmascripttests/test262
00:28:32.352996 git.c:415 trace: built-in: git rev-parse --local-env-vars
00:28:32.353646 git.c:415 trace: built-in: git rev-parse --verify HEAD
00:28:32.354640 git.c:415 trace: built-in: git rev-parse --git-common-dir
00:28:32.355437 git.c:415 trace: built-in: git config -f .git/modules/tests/auto/qml/ecmascripttests/test262/config core.worktree
00:28:32.356344 git.c:415 trace: built-in: git submodule--helper connect-gitdir-workingtree tests/auto/qml/ecmascripttests/test262 tests/auto/qml/ecmascripttests/test262
00:28:32.357659 git.c:415 trace: built-in: git submodule--helper relative-path qtdeclarative/tests/auto/qml/ecmascripttests/test262/
00:28:32.358693 git.c:415 trace: built-in: git rev-parse --local-env-vars
00:28:32.360070 git.c:415 trace: built-in: git submodule--helper update-clone --recursive-prefix qtdeclarative/tests/auto/qml/ecmascripttests/test262/
00:28:32.361641 run-command.c:1553 run_processes_parallel: preparing to run up to 1 tasks
00:28:32.361673 run-command.c:1585 run_processes_parallel: done
00:28:32.363347 git.c:415 trace: built-in: git submodule--helper name qtdoc
00:28:32.364557 git.c:415 trace: built-in: git config submodule.qtdoc.update
00:28:32.365526 git.c:415 trace: built-in: git submodule--helper relative-path qtdoc
00:28:32.366413 git.c:415 trace: built-in: git rev-parse --local-env-vars
00:28:32.367022 git.c:415 trace: built-in: git rev-parse --verify HEAD
00:28:32.368012 git.c:415 trace: built-in: git rev-parse --git-common-dir
00:28:32.368810 git.c:415 trace: built-in: git config -f .git/modules/qtdoc/config core.worktree
00:28:32.369720 git.c:415 trace: built-in: git submodule--helper connect-gitdir-workingtree qtdoc qtdoc
fatal: could not open 'qtdoc/.git' for writing: Is a directory
00:28:32.370824 git.c:415 trace: built-in: git submodule--helper relative-path qtdoc/
00:28:32.371573 git.c:415 trace: built-in: git rev-parse --local-env-vars
00:28:32.372563 git.c:415 trace: built-in: git submodule--helper update-clone --recursive-prefix qtdoc/
00:28:32.372766 run-command.c:1553 run_processes_parallel: preparing to run up to 1 tasks
00:28:32.372775 run-command.c:1585 run_processes_parallel: done
00:28:32.373799 git.c:415 trace: built-in: git submodule--helper name qtdocgallery
00:28:32.374757 git.c:415 trace: built-in: git config submodule.qtdocgallery.update
00:28:32.375489 git.c:415 trace: built-in: git submodule--helper relative-path qtdocgallery
00:28:32.376272 git.c:415 trace: built-in: git rev-parse --local-env-vars
00:28:32.376930 git.c:415 trace: built-in: git rev-parse --verify HEAD
00:28:32.377819 git.c:415 trace: built-in: git rev-parse --git-common-dir
00:28:32.378643 git.c:415 trace: built-in: git config -f .git/modules/qtdocgallery/config core.worktree
00:28:32.379545 git.c:415 trace: built-in: git submodule--helper connect-gitdir-workingtree qtdocgallery qtdocgallery
00:28:32.380811 git.c:415 trace: built-in: git submodule--helper relative-path qtdocgallery/
00:28:32.381563 git.c:415 trace: built-in: git rev-parse --local-env-vars
00:28:32.382596 git.c:415 trace: built-in: git submodule--helper update-clone --recursive-prefix qtdocgallery/
00:28:32.382712 run-command.c:1553 run_processes_parallel: preparing to run up to 1 tasks
00:28:32.382738 run-command.c:1585 run_processes_parallel: done
00:28:32.383858 git.c:415 trace: built-in: git submodule--helper name qtfeedback
00:28:32.384939 git.c:415 trace: built-in: git config submodule.qtfeedback.update
00:28:32.385747 git.c:415 trace: built-in: git submodule--helper relative-path qtfeedback
00:28:32.386554 git.c:415 trace: built-in: git rev-parse --local-env-vars
00:28:32.387206 git.c:415 trace: built-in: git rev-parse --verify HEAD
00:28:32.388173 git.c:415 trace: built-in: git rev-parse --git-common-dir
00:28:32.388989 git.c:415 trace: built-in: git config -f .git/modules/qtfeedback/config core.worktree
00:28:32.389997 git.c:415 trace: built-in: git submodule--helper connect-gitdir-workingtree qtfeedback qtfeedback
00:28:32.391155 git.c:415 trace: built-in: git submodule--helper relative-path qtfeedback/
00:28:32.391946 git.c:415 trace: built-in: git rev-parse --local-env-vars
00:28:32.392937 git.c:415 trace: built-in: git submodule--helper update-clone --recursive-prefix qtfeedback/
00:28:32.393042 run-command.c:1553 run_processes_parallel: preparing to run up to 1 tasks
00:28:32.393053 run-command.c:1585 run_processes_parallel: done
00:28:32.394070 git.c:415 trace: built-in: git submodule--helper name qtgamepad
00:28:32.395037 git.c:415 trace: built-in: git config submodule.qtgamepad.update
00:28:32.395778 git.c:415 trace: built-in: git submodule--helper relative-path qtgamepad
00:28:32.396736 git.c:415 trace: built-in: git rev-parse --local-env-vars
00:28:32.397423 git.c:415 trace: built-in: git rev-parse --verify HEAD
00:28:32.398334 git.c:415 trace: built-in: git rev-parse --git-common-dir
00:28:32.399191 git.c:415 trace: built-in: git config -f .git/modules/qtgamepad/config core.worktree
00:28:32.400069 git.c:415 trace: built-in: git submodule--helper connect-gitdir-workingtree qtgamepad qtgamepad
fatal: could not open 'qtgamepad/.git' for writing: Is a directory
00:28:32.401080 git.c:415 trace: built-in: git submodule--helper relative-path qtgamepad/
00:28:32.401843 git.c:415 trace: built-in: git rev-parse --local-env-vars
00:28:32.402868 git.c:415 trace: built-in: git submodule--helper update-clone --recursive-prefix qtgamepad/
00:28:32.402976 run-command.c:1553 run_processes_parallel: preparing to run up to 1 tasks
00:28:32.402988 run-command.c:1585 run_processes_parallel: done
00:28:32.404108 git.c:415 trace: built-in: git submodule--helper name qtgraphicaleffects
00:28:32.405122 git.c:415 trace: built-in: git config submodule.qtgraphicaleffects.update
00:28:32.406006 git.c:415 trace: built-in: git submodule--helper relative-path qtgraphicaleffects
00:28:32.407290 git.c:415 trace: built-in: git rev-parse --local-env-vars
00:28:32.408276 git.c:415 trace: built-in: git rev-parse --verify HEAD
00:28:32.409773 git.c:415 trace: built-in: git rev-parse --git-common-dir
00:28:32.410889 git.c:415 trace: built-in: git config -f .git/modules/qtgraphicaleffects/config core.worktree
00:28:32.411702 git.c:415 trace: built-in: git submodule--helper connect-gitdir-workingtree qtgraphicaleffects qtgraphicaleffects
fatal: could not open 'qtgraphicaleffects/.git' for writing: Is a directory
00:28:32.412798 git.c:415 trace: built-in: git submodule--helper relative-path qtgraphicaleffects/
00:28:32.413551 git.c:415 trace: built-in: git rev-parse --local-env-vars
00:28:32.414577 git.c:415 trace: built-in: git submodule--helper update-clone --recursive-prefix qtgraphicaleffects/
00:28:32.414724 run-command.c:1553 run_processes_parallel: preparing to run up to 1 tasks
00:28:32.414732 run-command.c:1585 run_processes_parallel: done
00:28:32.416013 git.c:415 trace: built-in: git submodule--helper name qtimageformats
00:28:32.417100 git.c:415 trace: built-in: git config submodule.qtimageformats.update
00:28:32.418063 git.c:415 trace: built-in: git submodule--helper relative-path qtimageformats
00:28:32.419038 git.c:415 trace: built-in: git rev-parse --local-env-vars
00:28:32.419763 git.c:415 trace: built-in: git rev-parse --verify HEAD
00:28:32.420765 git.c:415 trace: built-in: git rev-parse --git-common-dir
00:28:32.421611 git.c:415 trace: built-in: git config -f .git/modules/qtimageformats/config core.worktree
00:28:32.422395 git.c:415 trace: built-in: git submodule--helper connect-gitdir-workingtree qtimageformats qtimageformats
fatal: could not open 'qtimageformats/.git' for writing: Is a directory
00:28:32.423463 git.c:415 trace: built-in: git submodule--helper relative-path qtimageformats/
00:28:32.424255 git.c:415 trace: built-in: git rev-parse --local-env-vars
00:28:32.425276 git.c:415 trace: built-in: git submodule--helper update-clone --recursive-prefix qtimageformats/
00:28:32.425454 run-command.c:1553 run_processes_parallel: preparing to run up to 1 tasks
00:28:32.425462 run-command.c:1585 run_processes_parallel: done
00:28:32.426548 git.c:415 trace: built-in: git submodule--helper name qtlocation
00:28:32.427653 git.c:415 trace: built-in: git config submodule.qtlocation.update
00:28:32.428520 git.c:415 trace: built-in: git submodule--helper relative-path qtlocation
00:28:32.429392 git.c:415 trace: built-in: git rev-parse --local-env-vars
00:28:32.430081 git.c:415 trace: built-in: git rev-parse --verify HEAD
00:28:32.430948 git.c:415 trace: built-in: git rev-parse --git-common-dir
00:28:32.431883 git.c:415 trace: built-in: git config -f .git/modules/qtlocation/config core.worktree
00:28:32.432644 git.c:415 trace: built-in: git submodule--helper connect-gitdir-workingtree qtlocation qtlocation
fatal: could not open 'qtlocation/.git' for writing: Is a directory
00:28:32.433659 git.c:415 trace: built-in: git submodule--helper relative-path qtlocation/
00:28:32.434425 git.c:415 trace: built-in: git rev-parse --local-env-vars
00:28:32.435419 git.c:415 trace: built-in: git submodule--helper update-clone --recursive-prefix qtlocation/
00:28:32.435658 run-command.c:1553 run_processes_parallel: preparing to run up to 1 tasks
00:28:32.435689 run-command.c:1585 run_processes_parallel: done
00:28:32.436536 git.c:415 trace: built-in: git submodule--helper name src/3rdparty/mapbox-gl-native
00:28:32.437460 git.c:415 trace: built-in: git config submodule.src/3rdparty/mapbox-gl-native.update
00:28:32.438238 git.c:415 trace: built-in: git submodule--helper relative-path qtlocation/src/3rdparty/mapbox-gl-native
00:28:32.439093 git.c:415 trace: built-in: git rev-parse --local-env-vars
00:28:32.439705 git.c:415 trace: built-in: git rev-parse --verify HEAD
00:28:32.440802 git.c:415 trace: built-in: git rev-parse --git-common-dir
00:28:32.441577 git.c:415 trace: built-in: git config -f .git/modules/src/3rdparty/mapbox-gl-native/config core.worktree
00:28:32.442548 git.c:415 trace: built-in: git submodule--helper connect-gitdir-workingtree src/3rdparty/mapbox-gl-native src/3rdparty/mapbox-gl-native
00:28:32.443595 git.c:415 trace: built-in: git submodule--helper relative-path qtlocation/src/3rdparty/mapbox-gl-native/
00:28:32.444333 git.c:415 trace: built-in: git rev-parse --local-env-vars
00:28:32.445350 git.c:415 trace: built-in: git submodule--helper update-clone --recursive-prefix qtlocation/src/3rdparty/mapbox-gl-native/
00:28:32.445928 run-command.c:1553 run_processes_parallel: preparing to run up to 1 tasks
00:28:32.445938 run-command.c:1585 run_processes_parallel: done
00:28:32.447234 git.c:415 trace: built-in: git submodule--helper name qtmacextras
00:28:32.448286 git.c:415 trace: built-in: git config submodule.qtmacextras.update
00:28:32.449142 git.c:415 trace: built-in: git submodule--helper relative-path qtmacextras
00:28:32.450001 git.c:415 trace: built-in: git rev-parse --local-env-vars
00:28:32.450815 git.c:415 trace: built-in: git rev-parse --verify HEAD
00:28:32.451994 git.c:415 trace: built-in: git rev-parse --git-common-dir
00:28:32.452936 git.c:415 trace: built-in: git config -f .git/modules/qtmacextras/config core.worktree
00:28:32.453738 git.c:415 trace: built-in: git submodule--helper connect-gitdir-workingtree qtmacextras qtmacextras
fatal: could not open 'qtmacextras/.git' for writing: Is a directory
00:28:32.454821 git.c:415 trace: built-in: git submodule--helper relative-path qtmacextras/
00:28:32.455618 git.c:415 trace: built-in: git rev-parse --local-env-vars
00:28:32.456636 git.c:415 trace: built-in: git submodule--helper update-clone --recursive-prefix qtmacextras/
00:28:32.456803 run-command.c:1553 run_processes_parallel: preparing to run up to 1 tasks
00:28:32.456812 run-command.c:1585 run_processes_parallel: done
00:28:32.458217 git.c:415 trace: built-in: git submodule--helper name qtmultimedia
00:28:32.459897 git.c:415 trace: built-in: git config submodule.qtmultimedia.update
00:28:32.460908 git.c:415 trace: built-in: git submodule--helper relative-path qtmultimedia
00:28:32.461787 git.c:415 trace: built-in: git rev-parse --local-env-vars
00:28:32.462588 git.c:415 trace: built-in: git rev-parse --verify HEAD
00:28:32.463611 git.c:415 trace: built-in: git rev-parse --git-common-dir
00:28:32.464496 git.c:415 trace: built-in: git config -f .git/modules/qtmultimedia/config core.worktree
00:28:32.465265 git.c:415 trace: built-in: git submodule--helper connect-gitdir-workingtree qtmultimedia qtmultimedia
fatal: could not open 'qtmultimedia/.git' for writing: Is a directory
00:28:32.466329 git.c:415 trace: built-in: git submodule--helper relative-path qtmultimedia/
00:28:32.467121 git.c:415 trace: built-in: git rev-parse --local-env-vars
00:28:32.468129 git.c:415 trace: built-in: git submodule--helper update-clone --recursive-prefix qtmultimedia/
00:28:32.468431 run-command.c:1553 run_processes_parallel: preparing to run up to 1 tasks
00:28:32.468441 run-command.c:1585 run_processes_parallel: done
00:28:32.469470 git.c:415 trace: built-in: git submodule--helper name qtnetworkauth
00:28:32.470535 git.c:415 trace: built-in: git config submodule.qtnetworkauth.update
00:28:32.471371 git.c:415 trace: built-in: git submodule--helper relative-path qtnetworkauth
00:28:32.472263 git.c:415 trace: built-in: git rev-parse --local-env-vars
00:28:32.472913 git.c:415 trace: built-in: git rev-parse --verify HEAD
00:28:32.473816 git.c:415 trace: built-in: git rev-parse --git-common-dir
00:28:32.474659 git.c:415 trace: built-in: git config -f .git/modules/qtnetworkauth/config core.worktree
00:28:32.475434 git.c:415 trace: built-in: git submodule--helper connect-gitdir-workingtree qtnetworkauth qtnetworkauth
fatal: could not open 'qtnetworkauth/.git' for writing: Is a directory
00:28:32.476461 git.c:415 trace: built-in: git submodule--helper relative-path qtnetworkauth/
00:28:32.477322 git.c:415 trace: built-in: git rev-parse --local-env-vars
00:28:32.478451 git.c:415 trace: built-in: git submodule--helper update-clone --recursive-prefix qtnetworkauth/
00:28:32.478570 run-command.c:1553 run_processes_parallel: preparing to run up to 1 tasks
00:28:32.478597 run-command.c:1585 run_processes_parallel: done
00:28:32.479732 git.c:415 trace: built-in: git submodule--helper name qtpim
00:28:32.480914 git.c:415 trace: built-in: git config submodule.qtpim.update
00:28:32.481850 git.c:415 trace: built-in: git submodule--helper relative-path qtpim
00:28:32.482689 git.c:415 trace: built-in: git rev-parse --local-env-vars
00:28:32.483373 git.c:415 trace: built-in: git rev-parse --verify HEAD
00:28:32.484330 git.c:415 trace: built-in: git rev-parse --git-common-dir
00:28:32.485198 git.c:415 trace: built-in: git config -f .git/modules/qtpim/config core.worktree
00:28:32.486164 git.c:415 trace: built-in: git submodule--helper connect-gitdir-workingtree qtpim qtpim
00:28:32.487302 git.c:415 trace: built-in: git submodule--helper relative-path qtpim/
00:28:32.488150 git.c:415 trace: built-in: git rev-parse --local-env-vars
00:28:32.489140 git.c:415 trace: built-in: git submodule--helper update-clone --recursive-prefix qtpim/
00:28:32.489344 run-command.c:1553 run_processes_parallel: preparing to run up to 1 tasks
00:28:32.489352 run-command.c:1585 run_processes_parallel: done
00:28:32.490484 git.c:415 trace: built-in: git submodule--helper name qtpurchasing
00:28:32.491559 git.c:415 trace: built-in: git config submodule.qtpurchasing.update
00:28:32.492429 git.c:415 trace: built-in: git submodule--helper relative-path qtpurchasing
00:28:32.493215 git.c:415 trace: built-in: git rev-parse --local-env-vars
00:28:32.493847 git.c:415 trace: built-in: git rev-parse --verify HEAD
00:28:32.495112 git.c:415 trace: built-in: git rev-parse --git-common-dir
00:28:32.496041 git.c:415 trace: built-in: git config -f .git/modules/qtpurchasing/config core.worktree
00:28:32.496843 git.c:415 trace: built-in: git submodule--helper connect-gitdir-workingtree qtpurchasing qtpurchasing
fatal: could not open 'qtpurchasing/.git' for writing: Is a directory
00:28:32.497956 git.c:415 trace: built-in: git submodule--helper relative-path qtpurchasing/
00:28:32.498768 git.c:415 trace: built-in: git rev-parse --local-env-vars
00:28:32.499883 git.c:415 trace: built-in: git submodule--helper update-clone --recursive-prefix qtpurchasing/
00:28:32.500003 run-command.c:1553 run_processes_parallel: preparing to run up to 1 tasks
00:28:32.500011 run-command.c:1585 run_processes_parallel: done
00:28:32.501056 git.c:415 trace: built-in: git submodule--helper name qtqa
00:28:32.502126 git.c:415 trace: built-in: git config submodule.qtqa.update
00:28:32.502974 git.c:415 trace: built-in: git submodule--helper relative-path qtqa
00:28:32.503773 git.c:415 trace: built-in: git rev-parse --local-env-vars
00:28:32.504591 git.c:415 trace: built-in: git rev-parse --verify HEAD
00:28:32.505645 git.c:415 trace: built-in: git rev-parse --git-common-dir
00:28:32.506732 git.c:415 trace: built-in: git config -f .git/modules/qtqa/config core.worktree
00:28:32.507875 git.c:415 trace: built-in: git submodule--helper connect-gitdir-workingtree qtqa qtqa
fatal: could not open 'qtqa/.git' for writing: Is a directory
00:28:32.509175 git.c:415 trace: built-in: git submodule--helper relative-path qtqa/
00:28:32.509985 git.c:415 trace: built-in: git rev-parse --local-env-vars
00:28:32.510882 git.c:415 trace: built-in: git submodule--helper update-clone --recursive-prefix qtqa/
00:28:32.511002 run-command.c:1553 run_processes_parallel: preparing to run up to 1 tasks
00:28:32.511010 run-command.c:1585 run_processes_parallel: done
00:28:32.512184 git.c:415 trace: built-in: git submodule--helper name qtquickcontrols
00:28:32.513251 git.c:415 trace: built-in: git config submodule.qtquickcontrols.update
00:28:32.514082 git.c:415 trace: built-in: git submodule--helper relative-path qtquickcontrols
00:28:32.514961 git.c:415 trace: built-in: git rev-parse --local-env-vars
00:28:32.515611 git.c:415 trace: built-in: git rev-parse --verify HEAD
00:28:32.516612 git.c:415 trace: built-in: git rev-parse --git-common-dir
00:28:32.517380 git.c:415 trace: built-in: git config -f .git/modules/qtquickcontrols/config core.worktree
00:28:32.518168 git.c:415 trace: built-in: git submodule--helper connect-gitdir-workingtree qtquickcontrols qtquickcontrols
fatal: could not open 'qtquickcontrols/.git' for writing: Is a directory
00:28:32.519164 git.c:415 trace: built-in: git submodule--helper relative-path qtquickcontrols/
00:28:32.519987 git.c:415 trace: built-in: git rev-parse --local-env-vars
00:28:32.521262 git.c:415 trace: built-in: git submodule--helper update-clone --recursive-prefix qtquickcontrols/
00:28:32.521463 run-command.c:1553 run_processes_parallel: preparing to run up to 1 tasks
00:28:32.521472 run-command.c:1585 run_processes_parallel: done
00:28:32.522591 git.c:415 trace: built-in: git submodule--helper name qtquickcontrols2
00:28:32.523636 git.c:415 trace: built-in: git config submodule.qtquickcontrols2.update
00:28:32.524524 git.c:415 trace: built-in: git submodule--helper relative-path qtquickcontrols2
00:28:32.525398 git.c:415 trace: built-in: git rev-parse --local-env-vars
00:28:32.526081 git.c:415 trace: built-in: git rev-parse --verify HEAD
00:28:32.527044 git.c:415 trace: built-in: git rev-parse --git-common-dir
00:28:32.527925 git.c:415 trace: built-in: git config -f .git/modules/qtquickcontrols2/config core.worktree
00:28:32.528793 git.c:415 trace: built-in: git submodule--helper connect-gitdir-workingtree qtquickcontrols2 qtquickcontrols2
fatal: could not open 'qtquickcontrols2/.git' for writing: Is a directory
00:28:32.529877 git.c:415 trace: built-in: git submodule--helper relative-path qtquickcontrols2/
00:28:32.530626 git.c:415 trace: built-in: git rev-parse --local-env-vars
00:28:32.531571 git.c:415 trace: built-in: git submodule--helper update-clone --recursive-prefix qtquickcontrols2/
00:28:32.532079 run-command.c:1553 run_processes_parallel: preparing to run up to 1 tasks
00:28:32.532090 run-command.c:1585 run_processes_parallel: done
00:28:32.533173 git.c:415 trace: built-in: git submodule--helper name qtremoteobjects
00:28:32.534263 git.c:415 trace: built-in: git config submodule.qtremoteobjects.update
00:28:32.535105 git.c:415 trace: built-in: git submodule--helper relative-path qtremoteobjects
00:28:32.536033 git.c:415 trace: built-in: git rev-parse --local-env-vars
00:28:32.536725 git.c:415 trace: built-in: git rev-parse --verify HEAD
00:28:32.537674 git.c:415 trace: built-in: git rev-parse --git-common-dir
00:28:32.538553 git.c:415 trace: built-in: git config -f .git/modules/qtremoteobjects/config core.worktree
00:28:32.539863 git.c:415 trace: built-in: git submodule--helper connect-gitdir-workingtree qtremoteobjects qtremoteobjects
00:28:32.541153 git.c:415 trace: built-in: git submodule--helper relative-path qtremoteobjects/
00:28:32.541930 git.c:415 trace: built-in: git rev-parse --local-env-vars
00:28:32.543141 git.c:415 trace: built-in: git submodule--helper update-clone --recursive-prefix qtremoteobjects/
00:28:32.543315 run-command.c:1553 run_processes_parallel: preparing to run up to 1 tasks
00:28:32.543323 run-command.c:1585 run_processes_parallel: done
00:28:32.544545 git.c:415 trace: built-in: git submodule--helper name qtrepotools
00:28:32.545673 git.c:415 trace: built-in: git config submodule.qtrepotools.update
00:28:32.546525 git.c:415 trace: built-in: git submodule--helper relative-path qtrepotools
00:28:32.547372 git.c:415 trace: built-in: git rev-parse --local-env-vars
00:28:32.548026 git.c:415 trace: built-in: git rev-parse --verify HEAD
00:28:32.548941 git.c:415 trace: built-in: git rev-parse --git-common-dir
00:28:32.550163 git.c:415 trace: built-in: git config -f .git/modules/qtrepotools/config core.worktree
00:28:32.551057 git.c:415 trace: built-in: git submodule--helper connect-gitdir-workingtree qtrepotools qtrepotools
fatal: could not open 'qtrepotools/.git' for writing: Is a directory
00:28:32.552245 git.c:415 trace: built-in: git submodule--helper relative-path qtrepotools/
00:28:32.552994 git.c:415 trace: built-in: git rev-parse --local-env-vars
00:28:32.553990 git.c:415 trace: built-in: git submodule--helper update-clone --recursive-prefix qtrepotools/
00:28:32.554068 run-command.c:1553 run_processes_parallel: preparing to run up to 1 tasks
00:28:32.554078 run-command.c:1585 run_processes_parallel: done
00:28:32.555115 git.c:415 trace: built-in: git submodule--helper name qtscript
00:28:32.556249 git.c:415 trace: built-in: git config submodule.qtscript.update
00:28:32.557242 git.c:415 trace: built-in: git submodule--helper relative-path qtscript
00:28:32.558261 git.c:415 trace: built-in: git rev-parse --local-env-vars
00:28:32.559396 git.c:415 trace: built-in: git rev-parse --verify HEAD
00:28:32.560545 git.c:415 trace: built-in: git rev-parse --git-common-dir
00:28:32.561380 git.c:415 trace: built-in: git config -f .git/modules/qtscript/config core.worktree
00:28:32.562155 git.c:415 trace: built-in: git submodule--helper connect-gitdir-workingtree qtscript qtscript
fatal: could not open 'qtscript/.git' for writing: Is a directory
00:28:32.563124 git.c:415 trace: built-in: git submodule--helper relative-path qtscript/
00:28:32.563927 git.c:415 trace: built-in: git rev-parse --local-env-vars
00:28:32.564905 git.c:415 trace: built-in: git submodule--helper update-clone --recursive-prefix qtscript/
00:28:32.565221 run-command.c:1553 run_processes_parallel: preparing to run up to 1 tasks
00:28:32.565231 run-command.c:1585 run_processes_parallel: done
00:28:32.566411 git.c:415 trace: built-in: git submodule--helper name qtscxml
00:28:32.567548 git.c:415 trace: built-in: git config submodule.qtscxml.update
00:28:32.568436 git.c:415 trace: built-in: git submodule--helper relative-path qtscxml
00:28:32.569265 git.c:415 trace: built-in: git rev-parse --local-env-vars
00:28:32.569916 git.c:415 trace: built-in: git rev-parse --verify HEAD
00:28:32.570831 git.c:415 trace: built-in: git rev-parse --git-common-dir
00:28:32.571669 git.c:415 trace: built-in: git config -f .git/modules/qtscxml/config core.worktree
00:28:32.572460 git.c:415 trace: built-in: git submodule--helper connect-gitdir-workingtree qtscxml qtscxml
fatal: could not open 'qtscxml/.git' for writing: Is a directory
00:28:32.573413 git.c:415 trace: built-in: git submodule--helper relative-path qtscxml/
00:28:32.574148 git.c:415 trace: built-in: git rev-parse --local-env-vars
00:28:32.575136 git.c:415 trace: built-in: git submodule--helper update-clone --recursive-prefix qtscxml/
00:28:32.575381 run-command.c:1553 run_processes_parallel: preparing to run up to 1 tasks
00:28:32.575389 run-command.c:1585 run_processes_parallel: done
00:28:32.576437 git.c:415 trace: built-in: git submodule--helper name qtsensors
00:28:32.577525 git.c:415 trace: built-in: git config submodule.qtsensors.update
00:28:32.578342 git.c:415 trace: built-in: git submodule--helper relative-path qtsensors
00:28:32.579090 git.c:415 trace: built-in: git rev-parse --local-env-vars
00:28:32.579769 git.c:415 trace: built-in: git rev-parse --verify HEAD
00:28:32.580707 git.c:415 trace: built-in: git rev-parse --git-common-dir
00:28:32.581594 git.c:415 trace: built-in: git config -f .git/modules/qtsensors/config core.worktree
00:28:32.582376 git.c:415 trace: built-in: git submodule--helper connect-gitdir-workingtree qtsensors qtsensors
fatal: could not open 'qtsensors/.git' for writing: Is a directory
00:28:32.583481 git.c:415 trace: built-in: git submodule--helper relative-path qtsensors/
00:28:32.584231 git.c:415 trace: built-in: git rev-parse --local-env-vars
00:28:32.585326 git.c:415 trace: built-in: git submodule--helper update-clone --recursive-prefix qtsensors/
00:28:32.585508 run-command.c:1553 run_processes_parallel: preparing to run up to 1 tasks
00:28:32.585534 run-command.c:1585 run_processes_parallel: done
00:28:32.586897 git.c:415 trace: built-in: git submodule--helper name qtserialbus
00:28:32.588337 git.c:415 trace: built-in: git config submodule.qtserialbus.update
00:28:32.589209 git.c:415 trace: built-in: git submodule--helper relative-path qtserialbus
00:28:32.590098 git.c:415 trace: built-in: git rev-parse --local-env-vars
00:28:32.590756 git.c:415 trace: built-in: git rev-parse --verify HEAD
00:28:32.591667 git.c:415 trace: built-in: git rev-parse --git-common-dir
00:28:32.592607 git.c:415 trace: built-in: git config -f .git/modules/qtserialbus/config core.worktree
00:28:32.593610 git.c:415 trace: built-in: git submodule--helper connect-gitdir-workingtree qtserialbus qtserialbus
fatal: could not open 'qtserialbus/.git' for writing: Is a directory
00:28:32.594737 git.c:415 trace: built-in: git submodule--helper relative-path qtserialbus/
00:28:32.595530 git.c:415 trace: built-in: git rev-parse --local-env-vars
00:28:32.596573 git.c:415 trace: built-in: git submodule--helper update-clone --recursive-prefix qtserialbus/
00:28:32.596704 run-command.c:1553 run_processes_parallel: preparing to run up to 1 tasks
00:28:32.596713 run-command.c:1585 run_processes_parallel: done
00:28:32.597783 git.c:415 trace: built-in: git submodule--helper name qtserialport
00:28:32.598892 git.c:415 trace: built-in: git config submodule.qtserialport.update
00:28:32.599733 git.c:415 trace: built-in: git submodule--helper relative-path qtserialport
00:28:32.600596 git.c:415 trace: built-in: git rev-parse --local-env-vars
00:28:32.601270 git.c:415 trace: built-in: git rev-parse --verify HEAD
00:28:32.602151 git.c:415 trace: built-in: git rev-parse --git-common-dir
00:28:32.603026 git.c:415 trace: built-in: git config -f .git/modules/qtserialport/config core.worktree
00:28:32.603836 git.c:415 trace: built-in: git submodule--helper connect-gitdir-workingtree qtserialport qtserialport
fatal: could not open 'qtserialport/.git' for writing: Is a directory
00:28:32.604865 git.c:415 trace: built-in: git submodule--helper relative-path qtserialport/
00:28:32.605611 git.c:415 trace: built-in: git rev-parse --local-env-vars
00:28:32.606636 git.c:415 trace: built-in: git submodule--helper update-clone --recursive-prefix qtserialport/
00:28:32.606810 run-command.c:1553 run_processes_parallel: preparing to run up to 1 tasks
00:28:32.606839 run-command.c:1585 run_processes_parallel: done
00:28:32.608349 git.c:415 trace: built-in: git submodule--helper name qtspeech
00:28:32.609706 git.c:415 trace: built-in: git config submodule.qtspeech.update
00:28:32.610607 git.c:415 trace: built-in: git submodule--helper relative-path qtspeech
00:28:32.611487 git.c:415 trace: built-in: git rev-parse --local-env-vars
00:28:32.612204 git.c:415 trace: built-in: git rev-parse --verify HEAD
00:28:32.613232 git.c:415 trace: built-in: git rev-parse --git-common-dir
00:28:32.614108 git.c:415 trace: built-in: git config -f .git/modules/qtspeech/config core.worktree
00:28:32.614933 git.c:415 trace: built-in: git submodule--helper connect-gitdir-workingtree qtspeech qtspeech
fatal: could not open 'qtspeech/.git' for writing: Is a directory
00:28:32.616036 git.c:415 trace: built-in: git submodule--helper relative-path qtspeech/
00:28:32.616776 git.c:415 trace: built-in: git rev-parse --local-env-vars
00:28:32.617739 git.c:415 trace: built-in: git submodule--helper update-clone --recursive-prefix qtspeech/
00:28:32.617837 run-command.c:1553 run_processes_parallel: preparing to run up to 1 tasks
00:28:32.617848 run-command.c:1585 run_processes_parallel: done
00:28:32.619008 git.c:415 trace: built-in: git submodule--helper name qtsvg
00:28:32.620270 git.c:415 trace: built-in: git config submodule.qtsvg.update
00:28:32.621128 git.c:415 trace: built-in: git submodule--helper relative-path qtsvg
00:28:32.622013 git.c:415 trace: built-in: git rev-parse --local-env-vars
00:28:32.622629 git.c:415 trace: built-in: git rev-parse --verify HEAD
00:28:32.623561 git.c:415 trace: built-in: git rev-parse --git-common-dir
00:28:32.624515 git.c:415 trace: built-in: git config -f .git/modules/qtsvg/config core.worktree
00:28:32.625298 git.c:415 trace: built-in: git submodule--helper connect-gitdir-workingtree qtsvg qtsvg
fatal: could not open 'qtsvg/.git' for writing: Is a directory
00:28:32.626287 git.c:415 trace: built-in: git submodule--helper relative-path qtsvg/
00:28:32.627145 git.c:415 trace: built-in: git rev-parse --local-env-vars
00:28:32.628338 git.c:415 trace: built-in: git submodule--helper update-clone --recursive-prefix qtsvg/
00:28:32.628495 run-command.c:1553 run_processes_parallel: preparing to run up to 1 tasks
00:28:32.628506 run-command.c:1585 run_processes_parallel: done
00:28:32.629807 git.c:415 trace: built-in: git submodule--helper name qtsystems
00:28:32.631235 git.c:415 trace: built-in: git config submodule.qtsystems.update
00:28:32.632216 git.c:415 trace: built-in: git submodule--helper relative-path qtsystems
00:28:32.633090 git.c:415 trace: built-in: git rev-parse --local-env-vars
00:28:32.633793 git.c:415 trace: built-in: git rev-parse --verify HEAD
00:28:32.634746 git.c:415 trace: built-in: git rev-parse --git-common-dir
00:28:32.635733 git.c:415 trace: built-in: git config -f .git/modules/qtsystems/config core.worktree
00:28:32.636868 git.c:415 trace: built-in: git submodule--helper connect-gitdir-workingtree qtsystems qtsystems
00:28:32.638098 git.c:415 trace: built-in: git submodule--helper relative-path qtsystems/
00:28:32.638884 git.c:415 trace: built-in: git rev-parse --local-env-vars
00:28:32.640013 git.c:415 trace: built-in: git submodule--helper update-clone --recursive-prefix qtsystems/
00:28:32.640214 run-command.c:1553 run_processes_parallel: preparing to run up to 1 tasks
00:28:32.640223 run-command.c:1585 run_processes_parallel: done
00:28:32.641300 git.c:415 trace: built-in: git submodule--helper name qttools
00:28:32.642395 git.c:415 trace: built-in: git config submodule.qttools.update
00:28:32.643236 git.c:415 trace: built-in: git submodule--helper relative-path qttools
00:28:32.644077 git.c:415 trace: built-in: git rev-parse --local-env-vars
00:28:32.644769 git.c:415 trace: built-in: git rev-parse --verify HEAD
00:28:32.645713 git.c:415 trace: built-in: git rev-parse --git-common-dir
00:28:32.646548 git.c:415 trace: built-in: git config -f .git/modules/qttools/config core.worktree
00:28:32.647334 git.c:415 trace: built-in: git submodule--helper connect-gitdir-workingtree qttools qttools
fatal: could not open 'qttools/.git' for writing: Is a directory
00:28:32.648349 git.c:415 trace: built-in: git submodule--helper relative-path qttools/
00:28:32.649095 git.c:415 trace: built-in: git rev-parse --local-env-vars
00:28:32.650073 git.c:415 trace: built-in: git submodule--helper update-clone --recursive-prefix qttools/
00:28:32.650421 run-command.c:1553 run_processes_parallel: preparing to run up to 1 tasks
00:28:32.650430 run-command.c:1585 run_processes_parallel: done
00:28:32.651506 git.c:415 trace: built-in: git submodule--helper name qttranslations
00:28:32.652684 git.c:415 trace: built-in: git config submodule.qttranslations.update
00:28:32.653475 git.c:415 trace: built-in: git submodule--helper relative-path qttranslations
00:28:32.654350 git.c:415 trace: built-in: git rev-parse --local-env-vars
00:28:32.654959 git.c:415 trace: built-in: git rev-parse --verify HEAD
00:28:32.656013 git.c:415 trace: built-in: git rev-parse --git-common-dir
00:28:32.656890 git.c:415 trace: built-in: git config -f .git/modules/qttranslations/config core.worktree
00:28:32.657870 git.c:415 trace: built-in: git submodule--helper connect-gitdir-workingtree qttranslations qttranslations
fatal: could not open 'qttranslations/.git' for writing: Is a directory
00:28:32.659252 git.c:415 trace: built-in: git submodule--helper relative-path qttranslations/
00:28:32.660213 git.c:415 trace: built-in: git rev-parse --local-env-vars
00:28:32.661187 git.c:415 trace: built-in: git submodule--helper update-clone --recursive-prefix qttranslations/
00:28:32.661332 run-command.c:1553 run_processes_parallel: preparing to run up to 1 tasks
00:28:32.661361 run-command.c:1585 run_processes_parallel: done
00:28:32.662521 git.c:415 trace: built-in: git submodule--helper name qtvirtualkeyboard
00:28:32.663699 git.c:415 trace: built-in: git config submodule.qtvirtualkeyboard.update
00:28:32.664602 git.c:415 trace: built-in: git submodule--helper relative-path qtvirtualkeyboard
00:28:32.665572 git.c:415 trace: built-in: git rev-parse --local-env-vars
00:28:32.666259 git.c:415 trace: built-in: git rev-parse --verify HEAD
00:28:32.667181 git.c:415 trace: built-in: git rev-parse --git-common-dir
00:28:32.668119 git.c:415 trace: built-in: git config -f .git/modules/qtvirtualkeyboard/config core.worktree
00:28:32.669040 git.c:415 trace: built-in: git submodule--helper connect-gitdir-workingtree qtvirtualkeyboard qtvirtualkeyboard
fatal: could not open 'qtvirtualkeyboard/.git' for writing: Is a directory
00:28:32.670111 git.c:415 trace: built-in: git submodule--helper relative-path qtvirtualkeyboard/
00:28:32.671037 git.c:415 trace: built-in: git rev-parse --local-env-vars
00:28:32.672260 git.c:415 trace: built-in: git submodule--helper update-clone --recursive-prefix qtvirtualkeyboard/
00:28:32.672510 run-command.c:1553 run_processes_parallel: preparing to run up to 1 tasks
00:28:32.672522 run-command.c:1585 run_processes_parallel: done
00:28:32.673824 git.c:415 trace: built-in: git submodule--helper name qtwayland
00:28:32.675254 git.c:415 trace: built-in: git config submodule.qtwayland.update
00:28:32.676254 git.c:415 trace: built-in: git submodule--helper relative-path qtwayland
00:28:32.677149 git.c:415 trace: built-in: git rev-parse --local-env-vars
00:28:32.677815 git.c:415 trace: built-in: git rev-parse --verify HEAD
00:28:32.678705 git.c:415 trace: built-in: git rev-parse --git-common-dir
00:28:32.679515 git.c:415 trace: built-in: git config -f .git/modules/qtwayland/config core.worktree
00:28:32.680378 git.c:415 trace: built-in: git submodule--helper connect-gitdir-workingtree qtwayland qtwayland
fatal: could not open 'qtwayland/.git' for writing: Is a directory
00:28:32.681349 git.c:415 trace: built-in: git submodule--helper relative-path qtwayland/
00:28:32.682068 git.c:415 trace: built-in: git rev-parse --local-env-vars
00:28:32.683020 git.c:415 trace: built-in: git submodule--helper update-clone --recursive-prefix qtwayland/
00:28:32.683211 run-command.c:1553 run_processes_parallel: preparing to run up to 1 tasks
00:28:32.683220 run-command.c:1585 run_processes_parallel: done
00:28:32.684288 git.c:415 trace: built-in: git submodule--helper name qtwebchannel
00:28:32.685365 git.c:415 trace: built-in: git config submodule.qtwebchannel.update
00:28:32.686197 git.c:415 trace: built-in: git submodule--helper relative-path qtwebchannel
00:28:32.687250 git.c:415 trace: built-in: git rev-parse --local-env-vars
00:28:32.688053 git.c:415 trace: built-in: git rev-parse --verify HEAD
00:28:32.689027 git.c:415 trace: built-in: git rev-parse --git-common-dir
00:28:32.689828 git.c:415 trace: built-in: git config -f .git/modules/qtwebchannel/config core.worktree
00:28:32.690649 git.c:415 trace: built-in: git submodule--helper connect-gitdir-workingtree qtwebchannel qtwebchannel
fatal: could not open 'qtwebchannel/.git' for writing: Is a directory
00:28:32.691673 git.c:415 trace: built-in: git submodule--helper relative-path qtwebchannel/
00:28:32.692395 git.c:415 trace: built-in: git rev-parse --local-env-vars
00:28:32.693337 git.c:415 trace: built-in: git submodule--helper update-clone --recursive-prefix qtwebchannel/
00:28:32.693440 run-command.c:1553 run_processes_parallel: preparing to run up to 1 tasks
00:28:32.693450 run-command.c:1585 run_processes_parallel: done
00:28:32.694513 git.c:415 trace: built-in: git submodule--helper name qtwebengine
00:28:32.695558 git.c:415 trace: built-in: git config submodule.qtwebengine.update
00:28:32.696394 git.c:415 trace: built-in: git submodule--helper relative-path qtwebengine
00:28:32.697128 git.c:415 trace: built-in: git rev-parse --local-env-vars
00:28:32.697813 git.c:415 trace: built-in: git rev-parse --verify HEAD
00:28:32.698713 git.c:415 trace: built-in: git rev-parse --git-common-dir
00:28:32.699550 git.c:415 trace: built-in: git config -f .git/modules/qtwebengine/config core.worktree
00:28:32.700365 git.c:415 trace: built-in: git submodule--helper connect-gitdir-workingtree qtwebengine qtwebengine
fatal: could not open 'qtwebengine/.git' for writing: Is a directory
00:28:32.701368 git.c:415 trace: built-in: git submodule--helper relative-path qtwebengine/
00:28:32.702128 git.c:415 trace: built-in: git rev-parse --local-env-vars
00:28:32.703122 git.c:415 trace: built-in: git submodule--helper update-clone --recursive-prefix qtwebengine/
00:28:32.703383 run-command.c:1553 run_processes_parallel: preparing to run up to 1 tasks
00:28:32.703414 run-command.c:1585 run_processes_parallel: done
00:28:32.704307 git.c:415 trace: built-in: git submodule--helper name src/3rdparty
00:28:32.705228 git.c:415 trace: built-in: git config submodule.src/3rdparty.update
00:28:32.706067 git.c:415 trace: built-in: git submodule--helper relative-path qtwebengine/src/3rdparty
00:28:32.707040 git.c:415 trace: built-in: git rev-parse --local-env-vars
00:28:32.707919 git.c:415 trace: built-in: git rev-parse --verify HEAD
00:28:32.709130 git.c:415 trace: built-in: git rev-parse --git-common-dir
00:28:32.709996 git.c:415 trace: built-in: git config -f .git/modules/src/3rdparty/config core.worktree
00:28:32.710800 git.c:415 trace: built-in: git submodule--helper connect-gitdir-workingtree src/3rdparty src/3rdparty
fatal: could not open 'src/3rdparty/.git' for writing: Is a directory
00:28:32.711881 git.c:415 trace: built-in: git submodule--helper relative-path qtwebengine/src/3rdparty/
00:28:32.712760 git.c:415 trace: built-in: git rev-parse --local-env-vars
00:28:32.713700 git.c:415 trace: built-in: git submodule--helper update-clone --recursive-prefix qtwebengine/src/3rdparty/
00:28:32.725343 run-command.c:1553 run_processes_parallel: preparing to run up to 1 tasks
00:28:32.725361 run-command.c:1585 run_processes_parallel: done
00:28:32.727259 git.c:415 trace: built-in: git submodule--helper name qtwebglplugin
00:28:32.728403 git.c:415 trace: built-in: git config submodule.qtwebglplugin.update
00:28:32.729247 git.c:415 trace: built-in: git submodule--helper relative-path qtwebglplugin
00:28:32.730030 git.c:415 trace: built-in: git rev-parse --local-env-vars
00:28:32.730647 git.c:415 trace: built-in: git rev-parse --verify HEAD
00:28:32.731663 git.c:415 trace: built-in: git rev-parse --git-common-dir
00:28:32.732627 git.c:415 trace: built-in: git config -f .git/modules/qtwebglplugin/config core.worktree
00:28:32.733597 git.c:415 trace: built-in: git submodule--helper connect-gitdir-workingtree qtwebglplugin qtwebglplugin
00:28:32.734749 git.c:415 trace: built-in: git submodule--helper relative-path qtwebglplugin/
00:28:32.735528 git.c:415 trace: built-in: git rev-parse --local-env-vars
00:28:32.736522 git.c:415 trace: built-in: git submodule--helper update-clone --recursive-prefix qtwebglplugin/
00:28:32.736647 run-command.c:1553 run_processes_parallel: preparing to run up to 1 tasks
00:28:32.736675 run-command.c:1585 run_processes_parallel: done
00:28:32.737777 git.c:415 trace: built-in: git submodule--helper name qtwebsockets
00:28:32.738846 git.c:415 trace: built-in: git config submodule.qtwebsockets.update
00:28:32.739635 git.c:415 trace: built-in: git submodule--helper relative-path qtwebsockets
00:28:32.740495 git.c:415 trace: built-in: git rev-parse --local-env-vars
00:28:32.741177 git.c:415 trace: built-in: git rev-parse --verify HEAD
00:28:32.742182 git.c:415 trace: built-in: git rev-parse --git-common-dir
00:28:32.742995 git.c:415 trace: built-in: git config -f .git/modules/qtwebsockets/config core.worktree
00:28:32.743886 git.c:415 trace: built-in: git submodule--helper connect-gitdir-workingtree qtwebsockets qtwebsockets
fatal: could not open 'qtwebsockets/.git' for writing: Is a directory
00:28:32.744957 git.c:415 trace: built-in: git submodule--helper relative-path qtwebsockets/
00:28:32.745717 git.c:415 trace: built-in: git rev-parse --local-env-vars
00:28:32.746692 git.c:415 trace: built-in: git submodule--helper update-clone --recursive-prefix qtwebsockets/
00:28:32.746840 run-command.c:1553 run_processes_parallel: preparing to run up to 1 tasks
00:28:32.746866 run-command.c:1585 run_processes_parallel: done
00:28:32.748007 git.c:415 trace: built-in: git submodule--helper name qtwebview
00:28:32.749097 git.c:415 trace: built-in: git config submodule.qtwebview.update
00:28:32.749954 git.c:415 trace: built-in: git submodule--helper relative-path qtwebview
00:28:32.750826 git.c:415 trace: built-in: git rev-parse --local-env-vars
00:28:32.751483 git.c:415 trace: built-in: git rev-parse --verify HEAD
00:28:32.752413 git.c:415 trace: built-in: git rev-parse --git-common-dir
00:28:32.753361 git.c:415 trace: built-in: git config -f .git/modules/qtwebview/config core.worktree
00:28:32.754127 git.c:415 trace: built-in: git submodule--helper connect-gitdir-workingtree qtwebview qtwebview
fatal: could not open 'qtwebview/.git' for writing: Is a directory
00:28:32.755139 git.c:415 trace: built-in: git submodule--helper relative-path qtwebview/
00:28:32.755984 git.c:415 trace: built-in: git rev-parse --local-env-vars
00:28:32.757138 git.c:415 trace: built-in: git submodule--helper update-clone --recursive-prefix qtwebview/
00:28:32.757282 run-command.c:1553 run_processes_parallel: preparing to run up to 1 tasks
00:28:32.757296 run-command.c:1585 run_processes_parallel: done
00:28:32.758804 git.c:415 trace: built-in: git submodule--helper name qtwinextras
00:28:32.760424 git.c:415 trace: built-in: git config submodule.qtwinextras.update
00:28:32.761271 git.c:415 trace: built-in: git submodule--helper relative-path qtwinextras
00:28:32.762207 git.c:415 trace: built-in: git rev-parse --local-env-vars
00:28:32.762894 git.c:415 trace: built-in: git rev-parse --verify HEAD
00:28:32.763791 git.c:415 trace: built-in: git rev-parse --git-common-dir
00:28:32.764976 git.c:415 trace: built-in: git config -f .git/modules/qtwinextras/config core.worktree
00:28:32.765725 git.c:415 trace: built-in: git submodule--helper connect-gitdir-workingtree qtwinextras qtwinextras
fatal: could not open 'qtwinextras/.git' for writing: Is a directory
00:28:32.766885 git.c:415 trace: built-in: git submodule--helper relative-path qtwinextras/
00:28:32.767774 git.c:415 trace: built-in: git rev-parse --local-env-vars
00:28:32.768941 git.c:415 trace: built-in: git submodule--helper update-clone --recursive-prefix qtwinextras/
00:28:32.769102 run-command.c:1553 run_processes_parallel: preparing to run up to 1 tasks
00:28:32.769131 run-command.c:1585 run_processes_parallel: done
00:28:32.770438 git.c:415 trace: built-in: git submodule--helper name qtx11extras
00:28:32.771952 git.c:415 trace: built-in: git config submodule.qtx11extras.update
00:28:32.772819 git.c:415 trace: built-in: git submodule--helper relative-path qtx11extras
00:28:32.773589 git.c:415 trace: built-in: git rev-parse --local-env-vars
00:28:32.774251 git.c:415 trace: built-in: git rev-parse --verify HEAD
00:28:32.775162 git.c:415 trace: built-in: git rev-parse --git-common-dir
00:28:32.776071 git.c:415 trace: built-in: git config -f .git/modules/qtx11extras/config core.worktree
00:28:32.776842 git.c:415 trace: built-in: git submodule--helper connect-gitdir-workingtree qtx11extras qtx11extras
fatal: could not open 'qtx11extras/.git' for writing: Is a directory
00:28:32.777843 git.c:415 trace: built-in: git submodule--helper relative-path qtx11extras/
00:28:32.778635 git.c:415 trace: built-in: git rev-parse --local-env-vars
00:28:32.779692 git.c:415 trace: built-in: git submodule--helper update-clone --recursive-prefix qtx11extras/
00:28:32.779869 run-command.c:1553 run_processes_parallel: preparing to run up to 1 tasks
00:28:32.779879 run-command.c:1585 run_processes_parallel: done
00:28:32.780894 git.c:415 trace: built-in: git submodule--helper name qtxmlpatterns
00:28:32.781947 git.c:415 trace: built-in: git config submodule.qtxmlpatterns.update
00:28:32.782800 git.c:415 trace: built-in: git submodule--helper relative-path qtxmlpatterns
00:28:32.784034 git.c:415 trace: built-in: git rev-parse --local-env-vars
00:28:32.784739 git.c:415 trace: built-in: git rev-parse --verify HEAD
00:28:32.785617 git.c:415 trace: built-in: git rev-parse --git-common-dir
00:28:32.786458 git.c:415 trace: built-in: git config -f .git/modules/qtxmlpatterns/config core.worktree
00:28:32.787263 git.c:415 trace: built-in: git submodule--helper connect-gitdir-workingtree qtxmlpatterns qtxmlpatterns
fatal: could not open 'qtxmlpatterns/.git' for writing: Is a directory
00:28:32.788317 git.c:415 trace: built-in: git submodule--helper relative-path qtxmlpatterns/
00:28:32.789011 git.c:415 trace: built-in: git rev-parse --local-env-vars
00:28:32.789985 git.c:415 trace: built-in: git submodule--helper update-clone --recursive-prefix qtxmlpatterns/
00:28:32.790227 run-command.c:1553 run_processes_parallel: preparing to run up to 1 tasks
00:28:32.790256 run-command.c:1585 run_processes_parallel: done
00:28:32.791071 git.c:415 trace: built-in: git submodule--helper name tests/auto/3rdparty/testsuites
00:28:32.792081 git.c:415 trace: built-in: git config submodule.testsuites.update
00:28:32.792875 git.c:415 trace: built-in: git submodule--helper relative-path qtxmlpatterns/tests/auto/3rdparty/testsuites
00:28:32.793713 git.c:415 trace: built-in: git rev-parse --local-env-vars
00:28:32.794363 git.c:415 trace: built-in: git rev-parse --verify HEAD
00:28:32.795249 git.c:415 trace: built-in: git rev-parse --git-common-dir
00:28:32.796074 git.c:415 trace: built-in: git config -f .git/modules/testsuites/config core.worktree
00:28:32.796861 git.c:415 trace: built-in: git submodule--helper connect-gitdir-workingtree testsuites tests/auto/3rdparty/testsuites
fatal: could not open 'tests/auto/3rdparty/testsuites/.git' for writing: Is a directory
00:28:32.798041 git.c:415 trace: built-in: git submodule--helper relative-path qtxmlpatterns/tests/auto/3rdparty/testsuites/
00:28:32.798785 git.c:415 trace: built-in: git rev-parse --local-env-vars
00:28:32.799766 git.c:415 trace: built-in: git submodule--helper update-clone --recursive-prefix qtxmlpatterns/tests/auto/3rdparty/testsuites/
00:28:32.805569 run-command.c:1553 run_processes_parallel: preparing to run up to 1 tasks
00:28:32.805586 run-command.c:1585 run_processes_parallel: done
^ permalink raw reply [relevance 3%]
* Re: Old submodules broken in 2.19rc1 and 2.19rc2
2018-09-07 17:08 7% ` Stefan Beller
2018-09-07 20:20 6% ` Jonathan Nieder
2018-09-07 22:33 3% ` Allan Sandfeld Jensen
@ 2018-09-07 22:35 7% ` Jonathan Nieder
2018-09-07 22:45 7% ` Stefan Beller
2 siblings, 1 reply; 200+ results
From: Jonathan Nieder @ 2018-09-07 22:35 UTC (permalink / raw)
To: Stefan Beller; +Cc: allan.jensen, git
Stefan Beller wrote:
> On Fri, Sep 7, 2018 at 2:53 AM Allan Sandfeld Jensen <allan.jensen@qt.io> wrote:
>> Submodules checked out with older versions of git not longer works in the
>> latest 2.19 releases. A "git submodule update --recursive" command wil fail
>> for each submodule with a line saying "fatal: could not open
>> '<submodule>/.git' for writing> Is a directory.
[...]
> I have the suspicion that e98317508c0 (submodule:
> ensure core.worktree is set after update, 2018-06-18)
> might be the offender.
I still was not able to reproduce it, but after a bit of staring at
the code, I'm pretty sure I just did something wrong in the
reproduction process. That commit is indeed the offender.
It introduces the following code (rewrapped for clarity) in
git-submodule.sh:
if ! $(
git config -f \
"$(git rev-parse --git-common-dir)/modules/$name/config" \
core.worktree
) 2>/dev/null
then
git submodule--helper connect-gitdir-workingtree "$name" "$sm_path"
fi
Staring at it for a while, you can see one problem: the 'if ! $(git
config)' should be simply 'if ! git config'. This ends up trying to
run the core.worktree value as a command, which would usually fail.
That brings us into connect_work_tree_and_git_dir, which does
/* Prepare .git file */
strbuf_addf(&gitfile_sb, "%s/.git", work_tree_);
if (safe_create_leading_directories_const(gitfile_sb.buf))
die(_("could not create directories for %s"), gitfile_sb.buf);
/* Prepare config file */
strbuf_addf(&cfg_sb, "%s/config", git_dir_);
if (safe_create_leading_directories_const(cfg_sb.buf))
die(_("could not create directories for %s"), cfg_sb.buf);
git_dir = real_pathdup(git_dir_, 1);
work_tree = real_pathdup(work_tree_, 1);
/* Write .git file */
write_file(gitfile_sb.buf, "gitdir: %s",
relative_path(git_dir, work_tree, &rel_path));
The write_file runs into .git already existing as a directory, failing
with the message Allan saw.
This would happen in at least two cases:
- if the submodule exists both in .git/modules/ *and* in the worktree
(due to flipping between Git versions and branches with and without
the submodule), the above will happen
- likewise if the submodule exists only in the worktree, like for Allan.
In "next" there is 74d4731d (submodule--helper: replace
connect-gitdir-workingtree by ensure-core-worktree, 2018-08-13) which
uses robust helpers in C that handle this much better. I think we
should revert e98317508c0 in "master" (for 2.19) and keep making use
of that 'second try' in "next" (for 2.20).
I'll try to pin down a reproduction case and send a revert + testsuite
patch.
Thanks again,
Jonathan
^ permalink raw reply [relevance 7%]
* Re: Old submodules broken in 2.19rc1 and 2.19rc2
2018-09-07 17:08 7% ` Stefan Beller
@ 2018-09-07 20:20 6% ` Jonathan Nieder
2018-09-07 22:33 3% ` Allan Sandfeld Jensen
2018-09-07 22:35 7% ` Jonathan Nieder
2 siblings, 0 replies; 200+ results
From: Jonathan Nieder @ 2018-09-07 20:20 UTC (permalink / raw)
To: Stefan Beller; +Cc: allan.jensen, git
Stefan Beller wrote:
> On Fri, Sep 7, 2018 at 2:53 AM Allan Sandfeld Jensen <allan.jensen@qt.io> wrote:
>> A "git submodule update --recursive" command wil fail
>> for each submodule with a line saying "fatal: could not open
>> '<submodule>/.git' for writing> Is a directory.
[...]
> I have the suspicion that e98317508c0 (submodule:
> ensure core.worktree is set after update, 2018-06-18)
> might be the offender.
Oh! That seems likely.
Allan, output from "strace -f git submodule update --init" would also
be interesting.
Jonathan
^ permalink raw reply [relevance 6%]
* Re: [PATCH 2/2] submodule.c: warn about missing submodule git directories
2018-09-07 18:49 5% ` Stefan Beller
@ 2018-09-07 19:53 8% ` Jonathan Nieder
0 siblings, 0 replies; 200+ results
From: Jonathan Nieder @ 2018-09-07 19:53 UTC (permalink / raw)
To: Stefan Beller; +Cc: git
Hi,
Stefan Beller wrote:
> On Wed, Sep 5, 2018 at 12:18 PM Jonathan Nieder <jrnieder@gmail.com> wrote:
>> Stefan Beller wrote:
>>> This is the continuation of f2d48994dc1 (submodule.c: submodule_move_head
>>> works with broken submodules, 2017-04-18), which tones down the case of
>>> "broken submodule" in case of a missing git directory of the submodule to
>>> be only a warning.
[...]
>> I don't understand what workflow this is a part of.
>>
>> If the submodule is missing, shouldn't we make it non-missing instead
>> of producing a partial checkout that doesn't build?
>
> No. checkout and friends do not want to touch the network
> (unless we are in a partial clone world; that is the user is fully
> aware that commands can use the network at totally unexpected
> times)
>
> So for that, all we can do is better error messages.
Thanks. This patch doesn't just improve error messages, though, but
it makes the operation report success instead of failing.
Isn't that likely to produce more confusion when I run additional
commands afterward? In other words, instead of
$ git checkout --recurse-submodules -B master origin/new-fancy-branch
Branch 'master' set up to track remote branch 'new-fancy-branch' from 'origin'.
Switched to a new branch 'master'
warning: Submodule 'new-fancy-submodule' is missing
$ git status
[some unclean state]
I would prefer to experience
$ git checkout --recurse-submodules -B master origin/new-fancy-branch
fatal: missing submodule 'new-fancy-submodule'
hint: run "git fetch --recurse-submodules" to fetch it
$ git status
[clean state]
$ git fetch --recurse-submodules
[...]
$ git checkout --recurse-submodules -B master origin/new-fancy-branch
Branch 'master' set up to track remote branch 'new-fancy-branch' from 'origin'.
Switched to a new branch 'master'
$ git status
[clean state]
Thanks,
Jonathan
^ permalink raw reply [relevance 8%]
* Re: [PATCH 2/2] submodule.c: warn about missing submodule git directories
@ 2018-09-07 18:49 5% ` Stefan Beller
2018-09-07 19:53 8% ` Jonathan Nieder
0 siblings, 1 reply; 200+ results
From: Stefan Beller @ 2018-09-07 18:49 UTC (permalink / raw)
To: Jonathan Nieder; +Cc: git
On Wed, Sep 5, 2018 at 12:18 PM Jonathan Nieder <jrnieder@gmail.com> wrote:
>
> Hi,
>
> Stefan Beller wrote:
>
> > This is the continuation of f2d48994dc1 (submodule.c: submodule_move_head
> > works with broken submodules, 2017-04-18), which tones down the case of
> > "broken submodule" in case of a missing git directory of the submodule to
> > be only a warning.
> >
> > Signed-off-by: Stefan Beller <sbeller@google.com>
> > ---
> > submodule.c | 16 ++++++++++++++++
> > t/t2013-checkout-submodule.sh | 2 +-
> > 2 files changed, 17 insertions(+), 1 deletion(-)
>
> I don't understand what workflow this is a part of.
>
> If the submodule is missing, shouldn't we make it non-missing instead
> of producing a partial checkout that doesn't build?
No. checkout and friends do not want to touch the network
(unless we are in a partial clone world; that is the user is fully
aware that commands can use the network at totally unexpected
times)
So for that, all we can do is better error messages.
Stefan
^ permalink raw reply [relevance 5%]
* Re: Old submodules broken in 2.19rc1 and 2.19rc2
@ 2018-09-07 17:08 7% ` Stefan Beller
2018-09-07 20:20 6% ` Jonathan Nieder
` (2 more replies)
0 siblings, 3 replies; 200+ results
From: Stefan Beller @ 2018-09-07 17:08 UTC (permalink / raw)
To: allan.jensen; +Cc: git
On Fri, Sep 7, 2018 at 2:53 AM Allan Sandfeld Jensen <allan.jensen@qt.io> wrote:
>
> Submodules checked out with older versions of git not longer works in the
> latest 2.19 releases. A "git submodule update --recursive" command wil fail
> for each submodule with a line saying "fatal: could not open
> '<submodule>/.git' for writing> Is a directory.
Can you run the update again with
GIT_TRACE=1 git submodule update ....
and post the output?
I have the suspicion that e98317508c0 (submodule:
ensure core.worktree is set after update, 2018-06-18)
might be the offender.
Could you try reverting that commit and check as well?
git clone https://github.com/git/git && cd git
git revert e98317508c0
make install # installs to you home dir at ~/bin
and then try again, as well?
(though bisection may be more fruitful if this doesn't pan out)
^ permalink raw reply [relevance 7%]
* Re: [PATCH] submodule.sh update --remote: default to oid instead of master
@ 2018-09-06 22:54 7% ` Jonathan Nieder
2018-10-02 0:17 26% ` [PATCH] submodule update --remote: introduce pinning Stefan Beller
0 siblings, 1 reply; 200+ results
From: Jonathan Nieder @ 2018-09-06 22:54 UTC (permalink / raw)
To: Stefan Beller; +Cc: git
Stefan Beller wrote:
> On Wed, Sep 5, 2018 at 4:10 PM Jonathan Nieder <jrnieder@gmail.com> wrote:
>> Broader comment: do you think people will be surprised by this new
>> behavior? Is there anything special we'd need to do to call it out
>> (e.g., print a warning or put something in release notes)?
>
> I guess. Not sure how to approach this best. Maybe we can
> extend the output of 'submodule update' to print that branch names
> instead of hashes for the configured case and keep printing hashes
> only for this case. Although that would not help someone who relies
> on the default solely.
Thinking more out loud: often the simplest migration path involves
multiple steps:
1. Warn in the case that is going to change, with no behavior change
yet.
2. Treat the case that will change as an error. This should
help flush out cases where people were relying on the old behavior.
3. Introduce the new behavior. Warn that old versions of Git don't
support it yet.
4. Eliminate the warning. You're all clear now.
Sometimes some of these steps can be combined.
Another possible approach is to measure. For example, is there some
way to find out how many people are relying in this "git submodule
update --remote" defaulting behavior? One example of this approach is
to make the change (all in one step) in "next" and deploy to some
relevant subpopulation and see if anyone screams. By making the
change in "next" instead of something with more stability guarantees,
you get the ability to roll back quickly.
There are other tools at our disposal --- e.g. command-line flags,
config, other kinds of research.
Here my first instinct would be to say this should be a command-line
flag. To start out, we can keep the historical behavior as a default,
but introduce a command-line option for the new behavior. This way,
people can pass the negation of that command-line option if they want
the older behavior, throughout the transition.
For example (please ignore names):
Step 0: introduce
git submodule update --remote --default-to-master; # current behavior
git submodule update --remote --no-default-to-master; # new behavior
and treat plain "git submodule update --remote" as --default-to-master.
Step 1: when neither --default-to-master nor --no-default-to-master
has been passed, warn when encountering a submodule with no branch
and treat it as "master".
Step 2: when neither --default-to-master nor --no-default-to-master
has been passed, error out when encountering a submodule with no
branch.
Step 3: when neither --default-to-master nor --no-default-to-master
has been passed, warn when encountering a submodule with no branch
and treat it as pinned.
Step 4: eliminate the warning.
What do you think?
Thanks,
Jonathan
^ permalink raw reply [relevance 7%]
Results 201-400 of ~6000 next (older) | prev (newer) | reverse | options above
-- pct% links below jump to the message on this page, permalinks otherwise --
2018-03-17 3:57 Confusing behavior with ignored submodules and `git commit -a` Michael Forney
2018-10-25 18:03 ` Michael Forney
2018-10-25 18:26 7% ` Stefan Beller
2018-08-26 10:02 [PATCH 00/21] Kill the_index part 4 Nguyễn Thái Ngọc Duy
2018-09-03 18:09 ` [PATCH v2 00/24] " Nguyễn Thái Ngọc Duy
2018-09-03 18:09 ` [PATCH v2 19/24] submodule.c: remove implicit dependency on the_index Nguyễn Thái Ngọc Duy
2018-10-19 16:20 ` Jeff King
2018-10-19 16:33 ` Duy Nguyen
2018-10-19 16:42 ` Jeff King
2018-10-19 16:57 4% ` Stefan Beller
2018-08-27 22:12 [PATCH 1/2] t2013: add test for missing but active submodule Stefan Beller
2018-08-27 22:12 ` [PATCH 2/2] submodule.c: warn about missing submodule git directories Stefan Beller
2018-09-05 19:18 ` Jonathan Nieder
2018-09-07 18:49 5% ` Stefan Beller
2018-09-07 19:53 8% ` Jonathan Nieder
2018-09-04 23:01 [PATCH 00/11] fetch: make sure submodule oids are fetched Stefan Beller
2018-09-04 23:01 ` [PATCH 06/11] submodule.c: sort changed_submodule_names before searching it Stefan Beller
2018-09-06 18:03 ` Junio C Hamano
2018-09-11 18:31 7% ` Stefan Beller
2018-09-05 22:48 [PATCH] submodule.sh update --remote: default to oid instead of master Stefan Beller
2018-09-05 23:10 ` Jonathan Nieder
2018-09-06 18:06 ` Stefan Beller
2018-09-06 22:54 7% ` Jonathan Nieder
2018-10-02 0:17 26% ` [PATCH] submodule update --remote: introduce pinning Stefan Beller
2018-09-06 20:14 sb/submodule-move-nested breaks t7411 under GIT_FSMONITOR_TEST Stefan Beller
2018-09-06 20:34 ` [PATCH] git-mv: allow submodules and fsmonitor to work together Stefan Beller
2018-09-10 15:58 10% ` Ben Peart
2018-09-10 16:29 11% ` [PATCH v1] " Ben Peart
2018-09-10 17:07 6% ` Stefan Beller
2018-09-10 19:38 4% ` Ben Peart
2018-09-07 9:52 Old submodules broken in 2.19rc1 and 2.19rc2 Allan Sandfeld Jensen
2018-09-07 17:08 7% ` Stefan Beller
2018-09-07 20:20 6% ` Jonathan Nieder
2018-09-07 22:33 3% ` Allan Sandfeld Jensen
2018-09-07 22:35 7% ` Jonathan Nieder
2018-09-07 22:45 7% ` Stefan Beller
2018-09-08 0:09 22% ` [PATCH] Revert "Merge branch 'sb/submodule-core-worktree'" (was Re: Old submodules broken in 2.19rc1 and 2.19rc2) Jonathan Nieder
2018-09-08 2:04 2% ` Junio C Hamano
2018-09-08 18:39 10% ` Johannes Sixt
2018-09-10 17:11 5% ` Stefan Beller
2018-09-07 12:31 2.19.0.rc2.windows.1: stash fails with dirty submodule Thomas Braun
2018-09-08 0:35 ` Jonathan Nieder
2018-09-11 22:24 4% ` Stefan Beller
2018-09-10 20:11 3% [ANNOUNCE] Git v2.19.0 Junio C Hamano
2018-09-11 23:49 10% [PATCH 0/9] fetch: make sure submodule oids are fetched Stefan Beller
2018-09-11 23:49 24% ` [PATCH 3/9] submodule.c: fix indentation Stefan Beller
2018-09-12 18:02 4% ` Junio C Hamano
2018-09-11 23:49 17% ` [PATCH 4/9] submodule.c: sort changed_submodule_names before searching it Stefan Beller
2018-09-12 18:18 6% ` Junio C Hamano
2018-09-12 19:06 4% ` Stefan Beller
2018-09-11 23:49 16% ` [PATCH 5/9] submodule: move global changed_submodule_names into fetch submodule struct Stefan Beller
2018-09-11 23:49 17% ` [PATCH 6/9] submodule.c: do not copy around submodule list Stefan Beller
2018-09-12 2:25 5% ` Ramsay Jones
2018-09-11 23:49 26% ` [PATCH 7/9] submodule: fetch in submodules git directory instead of in worktree Stefan Beller
2018-09-12 18:36 7% ` Junio C Hamano
2018-09-13 19:29 8% ` Stefan Beller
2018-09-11 23:49 18% ` [PATCH 8/9] fetch: retry fetching submodules if sha1 were not fetched Stefan Beller
2018-09-12 19:03 6% ` Junio C Hamano
2018-09-11 23:49 23% ` [PATCH 9/9] builtin/fetch: check for submodule updates for non branch fetches Stefan Beller
2018-09-12 19:20 6% ` Junio C Hamano
2018-09-17 21:35 9% ` [PATCHv2 0/9] fetch: make sure submodule oids are fetched Stefan Beller
2018-09-17 21:35 24% ` [PATCH 3/9] submodule.c: fix indentation Stefan Beller
2018-09-17 21:35 17% ` [PATCH 4/9] submodule.c: sort changed_submodule_names before searching it Stefan Beller
2018-09-17 21:35 16% ` [PATCH 5/9] submodule: move global changed_submodule_names into fetch submodule struct Stefan Beller
2018-09-17 21:35 17% ` [PATCH 6/9] submodule.c: do not copy around submodule list Stefan Beller
2018-09-17 21:35 26% ` [PATCH 7/9] submodule: fetch in submodules git directory instead of in worktree Stefan Beller
2018-09-17 21:35 18% ` [PATCH 8/9] fetch: retry fetching submodules if needed objects were not fetched Stefan Beller
2018-09-17 21:35 21% ` [PATCH 9/9] builtin/fetch: check for submodule updates for non branch fetches Stefan Beller
2018-09-12 16:45 [PATCH 0/1] contrib: Add script to show uncovered "new" lines Derrick Stolee via GitGitGadget
2018-09-12 19:14 1% ` Derrick Stolee
2018-09-17 14:09 [PATCH v5 0/9] Make submodules work if .gitmodules is not checked out Antonio Ospite
2018-09-17 14:09 ` [PATCH v5 1/9] submodule: add a print_config_from_gitmodules() helper Antonio Ospite
2018-09-24 10:25 ` Antonio Ospite
2018-09-24 23:06 4% ` Stefan Beller
2018-09-17 14:09 ` [PATCH v5 9/9] submodule: support reading .gitmodules when it's not in the working tree Antonio Ospite
2018-09-18 17:12 ` SZEDER Gábor
2018-09-20 15:35 ` Antonio Ospite
2018-09-21 16:19 ` Junio C Hamano
2018-09-27 14:49 4% ` Antonio Ospite
2018-09-24 10:20 ` Antonio Ospite
2018-09-24 21:00 7% ` Stefan Beller
2018-09-27 14:44 7% ` Antonio Ospite
2018-09-27 18:00 7% ` Stefan Beller
2018-10-01 15:45 7% ` Antonio Ospite
2018-10-01 19:42 7% ` Stefan Beller
2018-09-21 18:51 [PATCH] fetch: Ensure that fetch.recurseSubmodules overrides submodule.recurse Marc Branchaud
2018-09-21 19:22 7% ` Stefan Beller
2018-09-21 22:35 9% [PATCHv3 0/8] fetch: make sure submodule oids are fetched Stefan Beller
2018-09-21 22:35 ` [PATCH 1/8] sha1-array: provide oid_array_filter Stefan Beller
2018-09-22 12:58 ` Ævar Arnfjörð Bjarmason
2018-09-25 19:26 ` Stefan Beller
2018-09-26 4:15 ` Jeff King
2018-09-26 17:10 ` Junio C Hamano
2018-09-26 17:49 ` Ævar Arnfjörð Bjarmason
2018-09-26 18:27 ` Junio C Hamano
2018-09-26 18:43 ` Ævar Arnfjörð Bjarmason
2018-09-26 18:58 ` Jeff King
2018-09-26 20:44 ` On shipping more of our technical docs as manpages Ævar Arnfjörð Bjarmason
2018-09-26 23:21 4% ` Stefan Beller
2018-09-21 22:35 24% ` [PATCH 2/8] submodule.c: fix indentation Stefan Beller
2018-09-21 22:35 17% ` [PATCH 3/8] submodule.c: sort changed_submodule_names before searching it Stefan Beller
2018-09-21 22:35 16% ` [PATCH 4/8] submodule: move global changed_submodule_names into fetch submodule struct Stefan Beller
2018-09-21 22:35 17% ` [PATCH 5/8] submodule.c: do not copy around submodule list Stefan Beller
2018-09-21 22:35 26% ` [PATCH 6/8] submodule: fetch in submodules git directory instead of in worktree Stefan Beller
2018-09-21 22:35 23% ` [PATCH 7/8] fetch: retry fetching submodules if needed objects were not fetched Stefan Beller
2018-09-21 22:35 21% ` [PATCH 8/8] builtin/fetch: check for submodule updates for non branch fetches Stefan Beller
2018-09-22 18:04 [PATCH 0/8] fix per-worktree ref iteration in fsck/reflog expire Nguyễn Thái Ngọc Duy
2018-09-22 18:04 ` [PATCH 2/8] Add a place for (not) sharing stuff between worktrees Nguyễn Thái Ngọc Duy
2018-09-25 2:35 ` Stefan Beller
2018-09-25 15:36 ` Duy Nguyen
2018-09-25 16:24 5% ` Stefan Beller
2018-09-25 16:55 5% ` Duy Nguyen
2018-09-25 17:56 4% ` Stefan Beller
2018-09-24 17:59 "git rev-parse --show-superproject-working-tree" fails with "fatal: BUG: returned path string doesn't match cwd?" if supermodule has unmerged changes of the submodule reference Sam McKelvie
2018-09-24 19:15 4% ` Stefan Beller
2018-09-24 21:23 [PATCH] submodule.c: Make get_superproject_working_tree() work when " Sam McKelvie
2018-09-25 1:24 7% ` Stefan Beller
2018-09-25 18:42 4% Git Test Coverage Report (Tuesday, Sept 25) Derrick Stolee
2018-09-25 19:47 8% [PATCH v4 0/9] fetch: make sure submodule oids are fetched Stefan Beller
2018-09-25 19:47 24% ` [PATCH v4 2/9] submodule.c: fix indentation Stefan Beller
2018-09-26 22:18 4% ` Junio C Hamano
2018-09-25 19:47 17% ` [PATCH v4 3/9] submodule.c: sort changed_submodule_names before searching it Stefan Beller
2018-09-26 22:14 4% ` Junio C Hamano
2018-09-25 19:47 16% ` [PATCH v4 4/9] submodule: move global changed_submodule_names into fetch submodule struct Stefan Beller
2018-09-26 22:19 5% ` Junio C Hamano
2018-09-25 19:47 17% ` [PATCH v4 5/9] submodule.c: do not copy around submodule list Stefan Beller
2018-09-25 19:47 24% ` [PATCH v4 6/9] repository: repo_submodule_init to take a submodule struct Stefan Beller
2018-09-25 19:47 26% ` [PATCH v4 7/9] submodule: fetch in submodules git directory instead of in worktree Stefan Beller
2018-09-25 19:47 23% ` [PATCH v4 8/9] fetch: retry fetching submodules if needed objects were not fetched Stefan Beller
2018-09-25 19:47 21% ` [PATCH v4 9/9] builtin/fetch: check for submodule updates for non branch fetches Stefan Beller
2018-09-27 18:10 [PATCH] submodule: Alllow staged changes for get_superproject_working_tree Sam McKelvie
2018-09-27 19:42 7% ` Stefan Beller
2018-09-27 22:16 19% [PATCH] FYI / RFC: submodules: introduce repo-like workflow Stefan Beller
2018-09-27 22:27 7% ` Jonathan Nieder
2018-09-28 18:08 2% ` Junio C Hamano
2018-09-28 19:26 6% ` Ævar Arnfjörð Bjarmason
2018-09-28 20:23 6% ` Stefan Beller
2018-09-29 14:01 wishlist: git grep -r Christoph Berg
2018-09-29 14:55 ` [PATCH] grep: provide a noop --recursive option Ævar Arnfjörð Bjarmason
2018-10-01 19:23 5% ` Stefan Beller
2018-10-05 8:15 2% ` Christoph Berg
2018-10-05 8:19 5% ` Junio C Hamano
2018-10-05 13:05 2% ` Mischa POSLAWSKY
2018-10-05 19:17 5% ` Stefan Beller
2018-09-29 15:22 [BUG] Segfault in "git submodule" Raymond Jennings
2018-09-29 15:29 ` Ævar Arnfjörð Bjarmason
2018-09-29 15:33 ` Duy Nguyen
2018-09-29 16:43 ` Raymond Jennings
2018-10-01 19:19 7% ` Stefan Beller
2018-10-01 21:31 4% ` Raymond Jennings
2018-10-02 4:10 2% ` Raymond Jennings
2018-10-02 21:59 [PATCH 0/1] Limit client version advertisements Josh Steadmon
2018-10-02 21:59 ` [PATCH 1/1] protocol: limit max protocol version per service Josh Steadmon
2018-10-02 22:28 ` Stefan Beller
2018-10-03 21:33 ` Josh Steadmon
2018-10-03 22:47 3% ` Stefan Beller
2018-10-05 13:05 6% [PATCH v6 00/10] Make submodules work if .gitmodules is not checked out Antonio Ospite
2018-10-05 13:05 ` [PATCH v6 08/10] submodule: add a helper to check if it is safe to write to .gitmodules Antonio Ospite
2018-10-05 23:50 4% ` Stefan Beller
2018-10-06 9:19 4% ` Antonio Ospite
2018-10-06 23:44 4% ` Junio C Hamano
2018-10-08 12:37 2% ` Antonio Ospite
2018-10-05 13:06 ` [PATCH v6 09/10] submodule: support reading .gitmodules when it's not in the working tree Antonio Ospite
2018-10-08 22:19 4% ` Stefan Beller
2018-10-10 18:56 7% ` Antonio Ospite
2018-10-10 22:55 7% ` Stefan Beller
2018-10-05 22:45 16% [PATCH] builtin/grep.c: remote superflous submodule code Stefan Beller
2018-10-06 8:59 4% ` Antonio Ospite
2018-10-07 0:29 7% ` Junio C Hamano
2018-10-07 0:33 4% ` Junio C Hamano
2018-10-09 0:14 6% ` Stefan Beller
2018-10-08 15:17 [PATCH 0/3] Add GIT_TEST_MULTI_PACK_INDEX environment variable Derrick Stolee via GitGitGadget
2018-10-08 15:17 ` [PATCH 2/3] midx: close multi-pack-index on repack Derrick Stolee via GitGitGadget
2018-10-09 9:10 ` Junio C Hamano
2018-10-09 14:11 ` Derrick Stolee
2018-10-09 18:15 4% ` Stefan Beller
2018-10-08 21:56 [PATCH 00/14] Hash function transition part 15 brian m. carlson
2018-10-08 21:57 ` [PATCH 13/14] submodule: make zero-oid comparison hash function agnostic brian m. carlson
2018-10-08 23:10 4% ` Stefan Beller
2018-10-09 18:35 16% [PATCH] builtin/grep.c: remote superflous submodule code Stefan Beller
2018-10-10 0:10 ` Jonathan Tan
2018-10-10 22:49 7% ` Stefan Beller
2018-10-10 5:43 What's cooking in git.git (Oct 2018, #01; Wed, 10) Junio C Hamano
2018-10-10 18:55 4% ` Stefan Beller
2018-10-11 2:00 2% ` Junio C Hamano
2018-10-10 21:49 10% [PATCH v5 0/9] fetch: make sure submodule oids are fetched Stefan Beller
2018-10-10 21:50 24% ` [PATCH v5 2/9] submodule.c: fix indentation Stefan Beller
2018-10-10 21:50 17% ` [PATCH v5 3/9] submodule.c: sort changed_submodule_names before searching it Stefan Beller
2018-10-10 21:50 16% ` [PATCH v5 4/9] submodule.c: move global changed_submodule_names into fetch submodule struct Stefan Beller
2018-10-10 21:50 17% ` [PATCH v5 5/9] submodule.c: do not copy around submodule list Stefan Beller
2018-10-10 21:50 24% ` [PATCH v5 6/9] repository: repo_submodule_init to take a submodule struct Stefan Beller
2018-10-10 21:50 26% ` [PATCH v5 7/9] submodule: fetch in submodules git directory instead of in worktree Stefan Beller
2018-10-10 21:50 23% ` [PATCH v5 8/9] fetch: retry fetching submodules if needed objects were not fetched Stefan Beller
2018-10-10 21:50 21% ` [PATCH v5 9/9] builtin/fetch: check for submodule updates for non branch fetches Stefan Beller
2018-10-11 21:17 6% [RFC PATCH 00/19] Bring more repository handles into our code base Stefan Beller
2018-10-11 21:17 18% ` [PATCH 17/19] submodule: use submodule repos for object lookup Stefan Beller
2018-10-11 22:40 ` Jonathan Tan
2018-10-13 0:20 8% ` Stefan Beller
2018-10-16 19:30 8% ` Stefan Beller
2018-10-16 23:13 ` Jonathan Tan
2018-10-16 23:16 7% ` Stefan Beller
2018-10-11 21:17 14% ` [PATCH 18/19] submodule: don't add submodule as odb for push Stefan Beller
2018-10-11 23:00 ` Jonathan Tan
2018-10-11 23:09 8% ` Stefan Beller
2018-10-11 21:17 2% ` [PATCH 19/19] Apply semantic patches from previous patches Stefan Beller
2018-10-11 23:31 2% ` [RFC PATCH 00/19] Bring more repository handles into our code base Junio C Hamano
2018-10-12 12:59 2% Git Test Coverage Report (Friday, Oct 12) Derrick Stolee
2018-10-12 14:53 What's cooking in git.git (Oct 2018, #02; Sat, 13) Junio C Hamano
2018-10-12 19:44 2% ` Stefan Beller
2018-10-12 23:37 7% ` Stefan Beller
2018-10-13 1:03 5% ` Junio C Hamano
2018-10-15 17:32 5% ` Stefan Beller
2018-10-12 21:53 22% [PATCH] submodule helper: convert relative URL to absolute URL if needed Stefan Beller
2018-10-12 22:27 4% ` Jonathan Nieder
2018-10-16 0:19 21% ` Stefan Beller
2018-10-16 0:33 4% ` Jonathan Nieder
2018-10-16 5:15 2% ` Junio C Hamano
2018-10-16 17:27 21% ` Stefan Beller
2018-10-16 21:05 4% ` Jonathan Nieder
2018-10-15 0:01 5% [PATCH v2 00/15] Hash function transition part 15 brian m. carlson
2018-10-15 16:24 1% Git Test Coverage Report (Monday, Oct 15) Derrick Stolee
2018-10-16 18:13 9% [PATCH 0/9] Resending sb/submodule-recursive-fetch-gets-the-tip Stefan Beller
2018-10-16 18:13 24% ` [PATCH 2/9] submodule.c: fix indentation Stefan Beller
2018-10-16 18:13 17% ` [PATCH 3/9] submodule.c: sort changed_submodule_names before searching it Stefan Beller
2018-10-16 18:13 16% ` [PATCH 4/9] submodule.c: move global changed_submodule_names into fetch submodule struct Stefan Beller
2018-10-17 21:26 ` Jonathan Tan
2018-10-18 19:09 5% ` Stefan Beller
2018-10-16 18:13 16% ` [PATCH 5/9] submodule.c: do not copy around submodule list Stefan Beller
2018-10-16 18:13 24% ` [PATCH 6/9] repository: repo_submodule_init to take a submodule struct Stefan Beller
2018-10-16 18:13 26% ` [PATCH 7/9] submodule: fetch in submodules git directory instead of in worktree Stefan Beller
2018-10-17 22:58 ` Jonathan Tan
2018-10-23 18:26 7% ` Stefan Beller
2018-10-23 22:55 ` Jonathan Tan
2018-10-23 23:01 5% ` Stefan Beller
2018-10-16 18:13 23% ` [PATCH 8/9] fetch: retry fetching submodules if needed objects were not fetched Stefan Beller
2018-10-18 0:39 ` Jonathan Tan
2018-10-23 22:37 7% ` Stefan Beller
2018-10-23 23:37 ` Jonathan Tan
2018-10-25 21:42 7% ` Stefan Beller
2018-10-16 18:13 21% ` [PATCH 9/9] builtin/fetch: check for submodule updates for non branch fetches Stefan Beller
2018-10-18 2:30 4% ` [PATCH 0/9] Resending sb/submodule-recursive-fetch-gets-the-tip Junio C Hamano
2018-10-18 7:30 14% ` Junio C Hamano
2018-10-18 18:00 7% ` Stefan Beller
2018-10-16 23:35 7% [PATCH 00/19] Bring more repository handles into our code base Stefan Beller
2018-10-16 23:35 20% ` [PATCH 18/19] submodule: use submodule repos for object lookup Stefan Beller
2018-10-25 9:14 8% ` SZEDER Gábor
2018-10-16 23:35 14% ` [PATCH 19/19] submodule: don't add submodule as odb for push Stefan Beller
2018-10-19 20:39 2% ` Jonathan Tan
2018-10-17 12:41 4% ` [PATCH 00/19] Bring more repository handles into our code base Derrick Stolee
2018-10-17 17:53 5% ` Stefan Beller
2018-10-18 18:37 ` [RFC PATCH 0/2] Bring the_repository into cmd_foo Stefan Beller
2018-10-18 18:37 7% ` [RFC PATCH 1/2] repository: have get_the_repository() to remove the_repository dependency Stefan Beller
2018-10-22 17:39 ` New semantic patches vs. in-flight topics [was: Re: [PATCH 00/19] Bring more repository handles into our code base] SZEDER Gábor
2018-10-22 18:54 ` Stefan Beller
2018-10-25 1:59 ` SZEDER Gábor
2018-10-25 19:25 3% ` Stefan Beller
2018-10-16 23:45 17% [PATCH] builtin/submodule--helper: remove debugging leftover tracing Stefan Beller
2018-10-17 2:52 4% ` Jonathan Nieder
2018-10-19 12:33 Bug with "git mv" and submodules, and with "git submodule add something --force" Juergen Vogl
2018-10-19 20:40 7% ` Stefan Beller
2018-10-19 20:58 7% ` Jonathan Nieder
2018-10-22 2:52 5% ` Junio C Hamano
2018-10-22 21:52 12% ` Stefan Beller
2018-10-24 7:18 7% ` Junio C Hamano
2018-10-19 14:52 [PATCH/WIP 00/19] Kill the_index, final part Nguyễn Thái Ngọc Duy
2018-10-19 17:38 4% ` Stefan Beller
2018-10-19 16:55 [PATCH v2 19/24] submodule.c: remove implicit dependency on the_index Jeff King
2018-10-19 17:34 ` [PATCH] submodule.c: remove some of the_repository references Nguyễn Thái Ngọc Duy
2018-10-19 17:46 7% ` Stefan Beller
2018-10-23 21:04 git pull defaults for recursesubmodules Tommi Vainikainen
2018-10-23 21:57 5% ` Stefan Beller
2018-10-24 5:32 2% ` Tommi Vainikainen
2018-10-25 16:18 6% [PATCH v7 00/10] Make submodules work if .gitmodules is not checked out Antonio Ospite
2018-10-25 18:49 4% ` Stefan Beller
2018-10-26 1:59 4% ` Junio C Hamano
2018-10-26 18:43 4% ` Stefan Beller
2018-10-25 23:32 8% [PATCH 00/10] Resending sb/submodule-recursive-fetch-gets-the-tip Stefan Beller
2018-10-25 23:32 24% ` [PATCH 02/10] submodule.c: fix indentation Stefan Beller
2018-10-25 23:32 16% ` [PATCH 03/10] submodule.c: sort changed_submodule_names before searching it Stefan Beller
2018-10-25 23:32 16% ` [PATCH 04/10] submodule.c: tighten scope of changed_submodule_names struct Stefan Beller
2018-10-25 23:32 16% ` [PATCH 05/10] submodule: store OIDs in changed_submodule_names Stefan Beller
2018-10-25 23:32 25% ` [PATCH 06/10] repository: repo_submodule_init to take a submodule struct Stefan Beller
2018-10-26 19:15 ` Jonathan Tan
2018-10-26 22:01 4% ` Stefan Beller
2018-10-25 23:32 23% ` [PATCH 07/10] submodule: migrate get_next_submodule to use repository structs Stefan Beller
2018-10-25 23:32 17% ` [PATCH 08/10] submodule.c: fetch in submodules git directory instead of in worktree Stefan Beller
2018-10-25 23:32 24% ` [PATCH 09/10] fetch: try fetching submodules if needed objects were not fetched Stefan Beller
2018-10-25 23:32 17% ` [PATCH 10/10] builtin/fetch: check for submodule updates in any ref update Stefan Beller
Code repositories for project(s) associated with this public inbox
https://80x24.org/mirrors/git.git
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).