From: Glen Choo <chooglen@google.com>
To: Heather Lapointe via GitGitGadget <gitgitgadget@gmail.com>,
git@vger.kernel.org
Cc: "René Scharfe" <l.s.r@web.de>,
"Heather Lapointe" <alpha@alphaservcomputing.solutions>
Subject: Re: [PATCH v3 4/9] tree: handle submodule case for read_tree_at properly
Date: Wed, 26 Oct 2022 15:48:12 -0700 [thread overview]
Message-ID: <kl6lk04mqk0j.fsf@chooglen-macbookpro.roam.corp.google.com> (raw)
In-Reply-To: <d3d1738e670d5dbf1378fc5c3209b2e98234a771.1665973401.git.gitgitgadget@gmail.com>
"Heather Lapointe via GitGitGadget" <gitgitgadget@gmail.com> writes:
> @@ -47,40 +48,73 @@ int read_tree_at(struct repository *r,
> return -1;
> }
>
> - if (S_ISDIR(entry.mode))
> + if (S_ISDIR(entry.mode)) {
> oidcpy(&oid, &entry.oid);
> - else if (S_ISGITLINK(entry.mode)) {
> - struct commit *commit;
>
> - commit = lookup_commit(r, &entry.oid);
> + len = tree_entry_len(&entry);
> + strbuf_add(base, entry.path, len);
> + strbuf_addch(base, '/');
> + retval = read_tree_at(r, lookup_tree(r, &oid),
> + base, pathspec,
> + fn, context);
> + strbuf_setlen(base, oldlen);
> + if (retval)
> + return -1;
> + } else if (pathspec->recurse_submodules && S_ISGITLINK(entry.mode)) {
> + struct commit *commit;
> + struct repository subrepo;
> + struct repository* subrepo_p = &subrepo;
> + struct tree* submodule_tree;
> + char *submodule_rel_path;
> + int name_base_len = 0;
> +
> + len = tree_entry_len(&entry);
> + strbuf_add(base, entry.path, len);
> + submodule_rel_path = base->buf;
> + // repo_submodule_init expects a path relative to submodule_prefix
> + if (r->submodule_prefix) {
> + name_base_len = strlen(r->submodule_prefix);
> + // we should always expect to start with submodule_prefix
> + assert(!strncmp(submodule_rel_path, r->submodule_prefix, name_base_len));
> + // strip the prefix
> + submodule_rel_path += name_base_len;
> + // if submodule_prefix doesn't end with a /, we want to get rid of that too
> + if (is_dir_sep(submodule_rel_path[0])) {
> + submodule_rel_path++;
> + }
> + }
> +
> + if (repo_submodule_init(subrepo_p, r, submodule_rel_path, null_oid()))
> + die("couldn't init submodule %s", base->buf);
> +
> + if (repo_read_index(subrepo_p) < 0)
> + die("index file corrupt");
> +
> + commit = lookup_commit(subrepo_p, &entry.oid);
> if (!commit)
> - die("Commit %s in submodule path %s%s not found",
> + die("Commit %s in submodule path %s not found",
> oid_to_hex(&entry.oid),
> - base->buf, entry.path);
> -
> - // FIXME: This is the wrong repo instance (it refers to the superproject)
> - // it will always fail as is (will fix in later patch)
> - // This current codepath isn't executed by any existing callbacks
> - // so it wouldn't show up as an issue at this time.
> - if (repo_parse_commit(r, commit))
> - die("Invalid commit %s in submodule path %s%s",
> + base->buf);
> +
> + if (repo_parse_commit(subrepo_p, commit))
> + die("Invalid commit %s in submodule path %s",
> oid_to_hex(&entry.oid),
> - base->buf, entry.path);
> + base->buf);
>
> - oidcpy(&oid, get_commit_tree_oid(commit));
> - }
> - else
> - continue;
> + submodule_tree = repo_get_commit_tree(subrepo_p, commit);
> + oidcpy(&oid, submodule_tree ? &submodule_tree->object.oid : NULL);
>
> - len = tree_entry_len(&entry);
> - strbuf_add(base, entry.path, len);
> - strbuf_addch(base, '/');
> - retval = read_tree_at(r, lookup_tree(r, &oid),
> - base, pathspec,
> - fn, context);
> - strbuf_setlen(base, oldlen);
> - if (retval)
> - return -1;
> + strbuf_addch(base, '/');
> +
> + retval = read_tree_at(subrepo_p, lookup_tree(subrepo_p, &oid),
> + base, pathspec,
> + fn, context);
> + if (retval)
> + die("failed to read tree for %s", base->buf);
> + strbuf_setlen(base, oldlen);
> + repo_clear(subrepo_p);
> + }
> + // else, this is a file (or a submodule, but no pathspec->recurse_submodules)
In this patch, we say that we can ignore a submodule when
pathspec->recurse_submodules is 0, but unless I'm missing something, I
don't think that's the case. The preimage is:
else if (S_ISGITLINK(entry.mode)) {
struct commit *commit;
commit = lookup_commit(r, &entry.oid);
if (!commit)
die("Commit %s in submodule path %s%s not found",
oid_to_hex(&entry.oid),
base->buf, entry.path);
/* ... */
if (repo_parse_commit(r, commit))
die("Invalid commit %s in submodule path %s%s",
oid_to_hex(&entry.oid),
base->buf, entry.path);
oidcpy(&oid, get_commit_tree_oid(commit));
}
else
continue;
len = tree_entry_len(&entry);
strbuf_add(base, entry.path, len);
strbuf_addch(base, '/');
retval = read_tree_at(r, lookup_tree(r, &oid),
base, pathspec,
fn, context);
which isn't a no-op since we actually do recurse into the gitlink. I
don't know whether the subsequent call actually succeeds though (e.g.
maybe it always failed and it was just a de facto no-op?), but that's
much harder to prove. Since this function has callers outside of "git
archive", it would be better to be conservative and keep the original
behavior in the S_ISGITLINK(entry.mode) && !pathspec->recurse_submodules
case.
> }
> return 0;
> }
> --
> gitgitgadget
next prev parent reply other threads:[~2022-10-26 22:49 UTC|newest]
Thread overview: 48+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-10-12 17:52 [PATCH] archive: add --recurse-submodules to git-archive command Heather Lapointe via GitGitGadget
2022-10-13 11:35 ` [PATCH v2 0/2] archive: Add " Heather Lapointe via GitGitGadget
2022-10-13 11:35 ` [PATCH v2 1/2] archive: add " Alphadelta14 via GitGitGadget
2022-10-13 17:53 ` René Scharfe
2022-10-13 21:37 ` Heather Lapointe
2022-10-13 11:36 ` [PATCH v2 2/2] archive: fix a case of submodule in submodule traversal Alphadelta14 via GitGitGadget
2022-10-13 17:53 ` [PATCH v2 0/2] archive: Add --recurse-submodules to git-archive command René Scharfe
2022-10-13 21:23 ` Heather Lapointe
2022-10-14 9:47 ` René Scharfe
2022-10-17 2:23 ` [PATCH v3 0/9] " Heather Lapointe via GitGitGadget
2022-10-17 2:23 ` [PATCH v3 1/9] tree: do not use the_repository for tree traversal methods Alphadelta14 via GitGitGadget
2022-10-17 13:26 ` Junio C Hamano
2022-10-26 22:33 ` Glen Choo
2022-10-27 18:09 ` Jonathan Tan
2022-10-27 18:50 ` Junio C Hamano
2022-10-17 2:23 ` [PATCH v3 2/9] tree: update cases to use repo_ tree methods Heather Lapointe via GitGitGadget
2022-10-17 2:23 ` [PATCH v3 3/9] tree: increase test coverage for tree.c Heather Lapointe via GitGitGadget
2022-10-17 13:34 ` Phillip Wood
2022-10-17 13:36 ` Junio C Hamano
2022-10-27 18:28 ` Jonathan Tan
2022-10-17 2:23 ` [PATCH v3 4/9] tree: handle submodule case for read_tree_at properly Heather Lapointe via GitGitGadget
2022-10-17 13:48 ` Phillip Wood
2022-10-17 13:56 ` Junio C Hamano
2022-10-26 22:48 ` Glen Choo [this message]
2022-10-27 18:43 ` Jonathan Tan
2022-10-17 2:23 ` [PATCH v3 5/9] tree: add repository parameter to read_tree_fn_t Heather Lapointe via GitGitGadget
2022-10-17 2:23 ` [PATCH v3 6/9] archive: pass repo objects to write_archive handlers Heather Lapointe via GitGitGadget
2022-10-17 13:50 ` Phillip Wood
2022-10-17 2:23 ` [PATCH v3 7/9] archive: remove global repository from archive_args Heather Lapointe via GitGitGadget
2022-10-17 2:23 ` [PATCH v3 8/9] archive: add --recurse-submodules to git-archive command Heather Lapointe via GitGitGadget
2022-10-26 23:34 ` Glen Choo
2022-10-27 7:09 ` René Scharfe
2022-10-27 17:29 ` Glen Choo
2022-10-27 17:30 ` Glen Choo
2022-10-27 17:33 ` Glen Choo
2022-10-17 2:23 ` [PATCH v3 9/9] archive: add tests for git archive --recurse-submodules Heather Lapointe via GitGitGadget
2022-10-27 18:54 ` Jonathan Tan
2022-10-27 23:30 ` Glen Choo
2022-10-28 0:17 ` Ævar Arnfjörð Bjarmason
2022-10-17 13:57 ` [PATCH v3 0/9] archive: Add --recurse-submodules to git-archive command Phillip Wood
2022-10-18 18:34 ` Junio C Hamano
2022-10-18 18:48 ` Heather Lapointe
2022-10-19 16:16 ` Junio C Hamano
2022-10-19 20:44 ` Junio C Hamano
2022-10-20 1:21 ` Junio C Hamano
2022-10-21 1:43 ` Junio C Hamano
2022-10-26 22:14 ` Glen Choo
2022-10-28 18:18 ` Heather Lapointe
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
List information: http://vger.kernel.org/majordomo-info.html
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=kl6lk04mqk0j.fsf@chooglen-macbookpro.roam.corp.google.com \
--to=chooglen@google.com \
--cc=alpha@alphaservcomputing.solutions \
--cc=git@vger.kernel.org \
--cc=gitgitgadget@gmail.com \
--cc=l.s.r@web.de \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
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).