public inbox for git@vger.kernel.org 
 help / color / mirror / Atom feed
From: Jeff King <peff@peff•net>
To: Junio C Hamano <gitster@pobox•com>
Cc: "René Scharfe" <l.s.r@web•de>,
	phillip.wood@dunelm•org.uk, Cheng <prophecheng@stu•pku.edu.cn>,
	git@vger•kernel.org
Subject: Re: [PATCH 3/5] describe: catch unborn branch in describe_blob()
Date: Mon, 18 Aug 2025 19:07:17 -0400	[thread overview]
Message-ID: <20250818230717.GA1027509@coredump.intra.peff.net> (raw)
In-Reply-To: <xmqqy0rggwep.fsf@gitster.g>

On Mon, Aug 18, 2025 at 02:19:42PM -0700, Junio C Hamano wrote:

> Jeff King <peff@peff•net> writes:
> 
> > --- a/builtin/describe.c
> > +++ b/builtin/describe.c
> > @@ -518,10 +518,16 @@ static void describe_blob(const struct object_id *oid, struct strbuf *dst)
> >  {
> >  	struct rev_info revs;
> >  	struct strvec args = STRVEC_INIT;
> > +	struct object_id head_oid;
> >  	struct process_commit_data pcd = { *null_oid(the_hash_algo), oid, dst, &revs};
> 
> An unrelated tangent, but it seems that we are copying the object
> name for the first member of this struct, even though 1/5 changed
> the second one.

Yep, agreed. It goes away in the final patch, though.

> > +	if (repo_get_oid(the_repository, "HEAD", &head_oid))
> > +		die(_("cannot search for blob '%s' on an unborn branch"),
> > +		    oid_to_hex(oid));
> 
> Makes sense.  I briefly wondered if there is really a point in doing
> the traversal only from HEAD, but this topic is not about enhancing
> and making the "describe <blob>" more useful, but it is a strict
> improvement.

Yeah. We do document the traversal from HEAD, so I wondered if anybody
would be upset at being more inclusive. So it might need a new option.
Somebody who is more interested in the option is welcome to pick up that
topic. :)

> When you first mentioned "resolve HEAD ourselves", I somehow expected
> you to ask the ref subsystem to resolve HEAD, but this should do fine,
> thanks to ref_rev_parse_rules[].

I hadn't really thought about the distinction. If you grep for '"HEAD"',
looks like we have a mix of both types. I don't think it should matter
much in practice.

Some even call lookup_commit_reference_by_name(). So we could perhaps do
that, in which case the segfault fix in patch 4 just happens naturally,
because we know we are always traversing from at least one commit. I
don't see any difference between the approaches compelling enough to
switch, though.

-Peff

  reply	other threads:[~2025-08-18 23:07 UTC|newest]

Thread overview: 28+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-08-13  0:23 Potential Null Pointer Dereference detected by static analysis tool Cheng
2025-08-13 13:19 ` Phillip Wood
2025-08-14 23:26   ` Jeff King
2025-08-15 15:49     ` Phillip Wood
2025-08-17  9:27     ` René Scharfe
2025-08-18  4:48       ` Jeff King
2025-08-18  5:05         ` Jeff King
2025-08-18 19:56           ` René Scharfe
2025-08-18 20:21             ` Jeff King
2025-08-18 20:56               ` Jeff King
2025-08-18 20:58               ` [PATCH 0/5] fix segfault and other oddities describing blobs Jeff King
2025-08-18 20:59                 ` [PATCH 1/5] describe: pass oid struct by const pointer Jeff King
2025-08-18 21:05                   ` Junio C Hamano
2025-08-18 21:01                 ` [PATCH 2/5] describe: error if blob not found Jeff King
2025-08-18 21:12                   ` Junio C Hamano
2025-08-19  8:05                     ` Patrick Steinhardt
2025-08-19 18:32                   ` René Scharfe
2025-08-18 21:01                 ` [PATCH 3/5] describe: catch unborn branch in describe_blob() Jeff King
2025-08-18 21:19                   ` Junio C Hamano
2025-08-18 23:07                     ` Jeff King [this message]
2025-08-18 21:03                 ` [PATCH 4/5] describe: handle blob traversal with no commits Jeff King
2025-08-19  8:05                   ` Patrick Steinhardt
2025-08-19 16:59                     ` Jeff King
2025-08-20  4:34                       ` Patrick Steinhardt
2025-08-20  6:30                         ` [replacement PATCH " Jeff King
2025-08-18 21:04                 ` [PATCH 5/5] describe: pass commit to describe_commit() Jeff King
2025-08-19  8:05                   ` Patrick Steinhardt
2025-08-19 17:02                     ` Jeff King

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

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20250818230717.GA1027509@coredump.intra.peff.net \
    --to=peff@peff$(echo .)net \
    --cc=git@vger$(echo .)kernel.org \
    --cc=gitster@pobox$(echo .)com \
    --cc=l.s.r@web$(echo .)de \
    --cc=phillip.wood@dunelm$(echo .)org.uk \
    --cc=prophecheng@stu$(echo .)pku.edu.cn \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox