From: Jonathan Tan <jonathantanmy@google.com>
To: sbeller@google.com
Cc: jonathantanmy@google.com, gitster@pobox.com, git@vger.kernel.org
Subject: Re: [PATCH 8/9] fetch: retry fetching submodules if needed objects were not fetched
Date: Tue, 23 Oct 2018 16:37:56 -0700 [thread overview]
Message-ID: <20181023233756.190026-1-jonathantanmy@google.com> (raw)
In-Reply-To: <CAGZ79kZrK5G-EeGRzxyw0xW3ozo9_aPab4r8fn_Jc4hzKDOEwg@mail.gmail.com>
> > 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?
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).
> > > +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)
Ah, OK. I would call it a fake submodule then, and copy over the "No
entry in .gitmodules?" comment.
> > 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.
Ah, yes you're right.
next prev parent reply other threads:[~2018-10-23 23:49 UTC|newest]
Thread overview: 29+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-10-16 18:13 [PATCH 0/9] Resending sb/submodule-recursive-fetch-gets-the-tip Stefan Beller
2018-10-16 18:13 ` [PATCH 1/9] sha1-array: provide oid_array_filter Stefan Beller
2018-10-16 18:13 ` [PATCH 2/9] submodule.c: fix indentation Stefan Beller
2018-10-16 18:13 ` [PATCH 3/9] submodule.c: sort changed_submodule_names before searching it Stefan Beller
2018-10-17 21:21 ` Jonathan Tan
2018-10-16 18:13 ` [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 ` Stefan Beller
2018-10-16 18:13 ` [PATCH 5/9] submodule.c: do not copy around submodule list Stefan Beller
2018-10-17 21:45 ` Jonathan Tan
2018-10-18 2:35 ` Junio C Hamano
2018-10-16 18:13 ` [PATCH 6/9] repository: repo_submodule_init to take a submodule struct Stefan Beller
2018-10-17 21:55 ` Jonathan Tan
2018-10-16 18:13 ` [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 ` Stefan Beller
2018-10-23 22:55 ` Jonathan Tan
2018-10-23 23:01 ` Stefan Beller
2018-10-16 18:13 ` [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 ` Stefan Beller
2018-10-23 23:37 ` Jonathan Tan [this message]
2018-10-25 21:42 ` Stefan Beller
2018-10-16 18:13 ` [PATCH 9/9] builtin/fetch: check for submodule updates for non branch fetches Stefan Beller
2018-10-18 0:47 ` Jonathan Tan
2018-10-18 2:30 ` [PATCH 0/9] Resending sb/submodule-recursive-fetch-gets-the-tip Junio C Hamano
2018-10-18 7:30 ` Junio C Hamano
2018-10-18 18:00 ` Stefan Beller
-- strict thread matches above, loose matches on Subject: below --
2018-09-11 23:49 [PATCH 0/9] fetch: make sure submodule oids are fetched Stefan Beller
2018-09-17 21:35 ` [PATCHv2 " Stefan Beller
2018-09-17 21:35 ` [PATCH 8/9] fetch: retry fetching submodules if needed objects were not fetched Stefan Beller
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=20181023233756.190026-1-jonathantanmy@google.com \
--to=jonathantanmy@google.com \
--cc=git@vger.kernel.org \
--cc=gitster@pobox.com \
--cc=sbeller@google.com \
/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).