git.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Michael Haggerty <mhagger@alum.mit.edu>
To: git discussion list <git@vger.kernel.org>
Cc: Junio C Hamano <gitster@pobox.com>
Subject: Tag peeling peculiarities
Date: Wed, 13 Mar 2013 15:59:05 +0100	[thread overview]
Message-ID: <51409439.5090001@alum.mit.edu> (raw)

I have been working on the pack-refs code [1] and noticed what looks
like a problem with the handling of peeled refs in the packed-refs file
and in the reference cache.  In particular, the peeled versions of tags
outside of refs/tags are *not* stored in packed-refs, but after the
packed-refs file is read it is assumed that such tags cannot be peeled.

It is clear that annotated tags want to live under refs/tags, but there
are some ways to create them in other places (see below).  It is not
clear to me whether the prohibition of tags outside of refs/tags should
be made more airtight or whether the peeling of tags outside of
refs/tags should be fixed.

Example:

~/tmp$ git init foo
Initialized empty Git repository in /home/mhagger/tmp/foo/.git/
~/tmp$ cd foo
~/tmp/foo$ git config user.name 'Lou User'
~/tmp/foo$ git config user.email 'luser@exaple.com'
~/tmp/foo$
~/tmp/foo$ git commit --allow-empty -m "Initial commit"
[master (root-commit) 7e80ddd] Initial commit
~/tmp/foo$ git tag -m footag footag
~/tmp/foo$
~/tmp/foo$ # This is prohibited:
~/tmp/foo$ git update-ref refs/heads/foobranch $(git rev-parse footag)
error: Trying to write non-commit object
d9cdc84dd156ff83799f5226794711fbb2c8273a to branch refs/heads/foobranch
fatal: Cannot update the ref 'refs/heads/foobranch'.
~/tmp/foo$
~/tmp/foo$ # But this is allowed:
~/tmp/foo$ git update-ref refs/remotes/foo/bar $(git rev-parse footag)
~/tmp/foo$
~/tmp/foo$ # So is this:
~/tmp/foo$ git update-ref refs/yak/foobranch $(git rev-parse footag)
~/tmp/foo$
~/tmp/foo$ # Before packing, all tags are available in peel versions:
~/tmp/foo$ git show-ref -d
7e80ddd68f0225a0ea221f7cddbacf050be5a265 refs/heads/master
d9cdc84dd156ff83799f5226794711fbb2c8273a refs/remotes/foo/bar
7e80ddd68f0225a0ea221f7cddbacf050be5a265 refs/remotes/foo/bar^{}
d9cdc84dd156ff83799f5226794711fbb2c8273a refs/tags/footag
7e80ddd68f0225a0ea221f7cddbacf050be5a265 refs/tags/footag^{}
d9cdc84dd156ff83799f5226794711fbb2c8273a refs/yak/foobranch
7e80ddd68f0225a0ea221f7cddbacf050be5a265 refs/yak/foobranch^{}
~/tmp/foo$
~/tmp/foo$ git pack-refs --all
~/tmp/foo$
~/tmp/foo$ # After packing, tags outside of refs/tags are not peeled any
more:
~/tmp/foo$ git show-ref -d
7e80ddd68f0225a0ea221f7cddbacf050be5a265 refs/heads/master
d9cdc84dd156ff83799f5226794711fbb2c8273a refs/remotes/foo/bar
d9cdc84dd156ff83799f5226794711fbb2c8273a refs/tags/footag
7e80ddd68f0225a0ea221f7cddbacf050be5a265 refs/tags/footag^{}
d9cdc84dd156ff83799f5226794711fbb2c8273a refs/yak/foobranch
~/tmp/foo$
~/tmp/foo$ # Peeling the tags via "tag^0" works even after packing:
~/tmp/foo$ git rev-parse refs/yak/foobranch^0
7e80ddd68f0225a0ea221f7cddbacf050be5a265
~/tmp/foo$
~/tmp/foo$ # Here is another way to create a tag outside of refs/tags:
~/tmp/foo$ cd ..
~/tmp$ git clone foo foo-clone
Cloning into 'foo-clone'...
done.
~/tmp$ cd foo-clone
~/tmp/foo-clone$ git config --add remote.origin.fetch
'+refs/tags/*:refs/remotes/origin/tags/*'
~/tmp/foo-clone$ git fetch
From /home/mhagger/tmp/foo
 * [new tag]         footag     -> origin/tags/footag
~/tmp/foo-clone$
~/tmp/foo-clone$ # Again, the tag outside of refs/tags are not peeled
correctly after packing:
~/tmp/foo-clone$ git pack-refs --all
~/tmp/foo-clone$ git show-ref -d
7e80ddd68f0225a0ea221f7cddbacf050be5a265 refs/heads/master
7e80ddd68f0225a0ea221f7cddbacf050be5a265 refs/remotes/origin/HEAD
7e80ddd68f0225a0ea221f7cddbacf050be5a265 refs/remotes/origin/master
d9cdc84dd156ff83799f5226794711fbb2c8273a refs/remotes/origin/tags/footag
d9cdc84dd156ff83799f5226794711fbb2c8273a refs/tags/footag
7e80ddd68f0225a0ea221f7cddbacf050be5a265 refs/tags/footag^{}

Michael

[1] I am trying to fix the problem that peeled refs are lost whenever a
packed reference is deleted.

-- 
Michael Haggerty
mhagger@alum.mit.edu
http://softwareswirl.blogspot.com/

             reply	other threads:[~2013-03-13 14:59 UTC|newest]

Thread overview: 17+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2013-03-13 14:59 Michael Haggerty [this message]
2013-03-13 15:34 ` Tag peeling peculiarities Michael Haggerty
2013-03-13 17:29 ` Junio C Hamano
2013-03-13 21:58   ` Jeff King
2013-03-14  4:41     ` Michael Haggerty
2013-03-14  5:24       ` Jeff King
2013-03-14  5:32         ` Jeff King
2013-03-14 15:14           ` Junio C Hamano
2013-03-14 11:28         ` Michael Haggerty
2013-03-14 13:40           ` Jeff King
2013-03-15  5:12             ` Michael Haggerty
2013-03-15 16:28               ` Junio C Hamano
2013-03-16  8:48             ` Michael Haggerty
2013-03-16  9:34               ` Jeff King
2013-03-16 13:38                 ` Michael Haggerty
2013-03-18  3:17                   ` Michael Haggerty
2013-03-22 17:42                   ` 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=51409439.5090001@alum.mit.edu \
    --to=mhagger@alum.mit.edu \
    --cc=git@vger.kernel.org \
    --cc=gitster@pobox.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.
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 NNTP newsgroup(s).