All of lore.kernel.org
 help / color / mirror / Atom feed
From: "Ævar Arnfjörð Bjarmason" <avarab@gmail.com>
To: Junio C Hamano <gitster@pobox.com>
Cc: "René Scharfe" <l.s.r@web.de>,
	"Eli Schwartz" <eschwartz@archlinux.org>,
	git@vger.kernel.org
Subject: Re: [PATCH 1/2] pretty: add %(describe)
Date: Wed, 17 Feb 2021 01:47:52 +0100	[thread overview]
Message-ID: <87mtw3a5af.fsf@evledraar.gmail.com> (raw)
In-Reply-To: <xmqqft1vrgxa.fsf@gitster.c.googlers.com>


On Tue, Feb 16 2021, Junio C Hamano wrote:

> Ævar Arnfjörð Bjarmason <avarab@gmail.com> writes:
>
>> On Sun, Feb 14 2021, René Scharfe wrote:
>>
>>> Add a format placeholder for describe output.  Implement it by actually
>>> calling git describe, which is simple and guarantees correctness.  It's
>>> intended to be used with $Format:...$ in files with the attribute
>>> export-subst and git archive. 
>>
>> Does it really guarantee correctness though? In "builtin/describe.c" we
>> first walk over the refs and use that to format all N items we're asked
>> about.
>>
>> Under "git log" this is presumably in a race where refs added/deleted
>> during the run of "git log" will change the describe output to be
>> inconsistent with earlier lines.
>
> Yes, but it is not a news that the describe for a given commit will
> not stay the constant while you add more objects to the repository
> or you change the tags, whether the "describe" is driven internally
> by "git log" or by the end-user, so I am not sure how that becomes
> an issue.  If the output is inconsistent in a quiescent repository,
> that would be a problem, though.
>
> Puzzled.

Usually something shelling out has going for it is that even if it's
slow it's at least known to be bug-free, after all we use the command
like that already.

I just wanted to point out this edge case, for "git describe <commits>"
we do the ref list once at the beginning so our N list will be
consistent within itself.

Whereas one might expect "git log" to also format such a list, but due
to the internal implementation the semantics are different.

  reply	other threads:[~2021-02-17  0:51 UTC|newest]

Thread overview: 27+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-01-25  0:32 gitattributes export-subst and software versioning Eli Schwartz
2021-02-08 19:46 ` René Scharfe
2021-02-08 22:41   ` Junio C Hamano
2021-02-09  0:19   ` Eli Schwartz
2021-02-09 20:42     ` Junio C Hamano
2021-02-14 10:04       ` René Scharfe
2021-02-14 10:04     ` René Scharfe
2021-02-14 10:04 ` [PATCH 1/2] pretty: add %(describe) René Scharfe
2021-02-14 10:10   ` [PATCH 2/2] pretty: add merge and exclude options to %(describe) René Scharfe.
2021-02-17 18:31     ` Jeff King
2021-02-28 11:22       ` René Scharfe.
2021-02-28 15:41         ` Ævar Arnfjörð Bjarmason
2021-03-02 16:00           ` René Scharfe.
2021-03-06 16:18             ` René Scharfe.
     [not found]         ` <xmqqy2f6rc8f.fsf@gitster.c.googlers.com>
2021-03-02 16:00           ` René Scharfe.
     [not found]     ` <xmqqsg5uletz.fsf@gitster.g>
2021-02-28 11:22       ` René Scharfe.
2021-02-16  5:04   ` [PATCH 1/2] pretty: add %(describe) Eli Schwartz
2021-02-16 13:00   ` Ævar Arnfjörð Bjarmason
2021-02-16 17:13     ` René Scharfe.
2021-02-16 18:44     ` Junio C Hamano
2021-02-17  0:47       ` Ævar Arnfjörð Bjarmason [this message]
2021-02-28 11:22         ` René Scharfe.
     [not found]           ` <xmqq35xesqzk.fsf@gitster.c.googlers.com>
2021-03-02 16:00             ` René Scharfe.
2021-02-17  0:58   ` Ævar Arnfjörð Bjarmason
2021-02-17 18:12     ` Junio C Hamano
2021-02-28 11:22     ` René Scharfe.
     [not found]       ` <xmqq7dmqsr72.fsf@gitster.c.googlers.com>
2021-03-02 16:00         ` René Scharfe.

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=87mtw3a5af.fsf@evledraar.gmail.com \
    --to=avarab@gmail.com \
    --cc=eschwartz@archlinux.org \
    --cc=git@vger.kernel.org \
    --cc=gitster@pobox.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.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.