All of lore.kernel.org
 help / color / mirror / Atom feed
From: Nguyen Thai Ngoc Duy <pclouds@gmail.com>
To: Marc Branchaud <marcnarc@xiplink.com>
Cc: git@vger.kernel.org
Subject: Re: [1.8.0] Tag namespaces
Date: Tue, 1 Feb 2011 22:21:32 +0700	[thread overview]
Message-ID: <AANLkTi=kwU+-UxfnCnDj0dD2NLPovTTsCPbSf-RQ3L5P@mail.gmail.com> (raw)
In-Reply-To: <4D481EA0.9090802@xiplink.com>

On Tue, Feb 1, 2011 at 9:54 PM, Marc Branchaud <marcnarc@xiplink.com> wrote:
> On 11-01-31 10:20 PM, Nguyen Thai Ngoc Duy wrote:
>> On Tue, Feb 1, 2011 at 12:05 AM, Junio C Hamano <gitster@pobox.com> wrote:
>>> Now the 1.7.4 release is out, I'd like people to help thinking about the
>>> next cycle(s).
>>>
>>> As a discussion-starter, here are my random wishes.  Even though this does
>>> not attempt to be exhaustive, keeping the number of goals manageably small
>>> may help us focus.
>>
>> Another random wish, which does not come with a proposal. How about
>> tag namespace (ie. tags from a remote stay in remote namespace)?
>
> I had just started writing up such a proposal yesterday.  What I have so far
> is pretty preliminary:

Thanks. I wrote another proposal (should we have more or less standard
"Git Enhancement Proposal" process like Python's PEP or Scheme's SRFI
to keep track of these over time?) but it's good to see others look at
it too.

> Proposal:
>
> Change tag refspecs to distinguish between remote and local tags.  An
> unadorned tag "foo" could point to different commits in different
> repositories.  A remote could move/edit it's "foo" tag and have that update
> smoothly propagated to clones.
>
> I believe this was last brought up in November while discussing the refs base
> for notes:
>
> http://thread.gmane.org/gmane.comp.version-control.git/160503/focus=160655

One problem with the proposed ref layout is that it breaks current
layout (remotes/<remote>/head -> remotes/<remote>/heads/head). Some
sort of migration support is needed. On the other hand, new layout is
cleaner than my proposal (remote-tags/<remote>/tag).

> Risks:
>
> I think the main risk lies in breaking plain <tagname> refs, as they would
> become "origin/<tagname>" refs instead.  But I think that can be mitagated
> against (see below).
> ...
> To help mitigate the risk of breaking plain "<tagname>" refs, "git rev-parse"
> can look for plain names (i.e. ones without a /) in the remote tags location.

Hmm I thought "a-ref" would check "refs/tags/a-ref" and
"refs/remotes/*/a-ref". But you are right. Maybe "tags.relative" can
take three values instead of boolean (names TBD):

 - deprecated (current behavior)
 - migrating (fetch tags to refs/remotes, but tag lookup will look in
refs/tags as well as refs/remotes/*/tags)
 - migrated (fetch tags to refs/remotes, look up in order)

We may slowly turn default value step by step until it becomes "migrated".
-- 
Duy

  reply	other threads:[~2011-02-01 15:22 UTC|newest]

Thread overview: 126+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2011-01-31  5:53 What's cooking in git.git (Jan 2011, #06; Sun, 30) Junio C Hamano
2011-01-31 15:08 ` Sverre Rabbelier
2011-02-08 17:48   ` Sverre Rabbelier
2011-02-08 19:27     ` Junio C Hamano
2011-01-31 17:05 ` Planning for 1.7.5 and 1.8.0 Junio C Hamano
2011-01-31 17:06   ` [1.8.0] default "git merge" without argument to "git merge @{u}" Junio C Hamano
2011-01-31 20:14     ` Jeff King
2011-01-31 20:17       ` Junio C Hamano
2011-01-31 20:32         ` Felipe Contreras
2011-01-31 20:50     ` [1.8.0] (v2) " Junio C Hamano
2011-01-31 22:55       ` Jeff King
2011-02-01  0:01         ` Thomas Adam
2011-02-01 18:34         ` Scott Chacon
2011-02-01 20:11           ` moving to a git-backed wiki Jeff King
2011-02-01 22:36             ` Jay Soffian
2011-02-01 22:48               ` J.H.
2011-02-02  9:55                 ` Vincent Hanquez
2011-02-02 10:53                   ` Felipe Contreras
2011-02-02 11:14                     ` Jakub Narebski
2011-02-03  2:24                   ` J.H.
2011-02-03 17:45                     ` Jeff King
2011-02-03 19:06                       ` Sverre Rabbelier
2011-02-04  6:03                         ` Jeff King
2011-02-03 20:34                       ` Felipe Contreras
2011-02-04  6:16                         ` Jeff King
2011-02-04 17:50                           ` Felipe Contreras
2011-02-04 14:34                       ` Joey Hess
2011-02-05  7:00                         ` david
2011-02-04  7:31         ` [1.8.0] (v2) default "git merge" without argument to "git merge @{u}" Thomas Hochstein
2011-02-04 23:01     ` [PATCH/RFC] Add support for merging from upstream by default Jared Hance
2011-01-31 17:07   ` [1.8.0] Unify "pathspec" semantics Junio C Hamano
2011-02-01 14:56     ` Nguyen Thai Ngoc Duy
2011-01-31 20:28   ` [1.8.0] reorganize the mess that the source tree has become Nicolas Pitre
2011-01-31 20:57     ` Junio C Hamano
2011-01-31 21:08       ` Matthieu Moy
2011-01-31 21:33         ` Nicolas Pitre
2011-01-31 21:19       ` Nicolas Pitre
2011-01-31 21:00     ` Jeff King
2011-01-31 21:28       ` Nicolas Pitre
2011-01-31 22:17         ` Junio C Hamano
2011-01-31 22:36           ` João P. Sampaio
2011-01-31 22:37           ` Nicolas Pitre
2011-01-31 23:12         ` Jeff King
2011-02-01  0:29           ` Nicolas Pitre
2011-02-01  1:48             ` Jeff King
2011-02-01  4:05               ` Nicolas Pitre
2011-02-01 12:42                 ` Thomas Rast
2011-02-01 11:14                   ` Jonathan Nieder
2011-02-01 11:22                     ` Jonathan Nieder
2011-02-01 13:08                   ` Nicolas Pitre
2011-02-01 16:02                   ` Nguyen Thai Ngoc Duy
2011-02-01 21:53               ` Junio C Hamano
2011-02-01  0:35           ` Erik Faye-Lund
2011-02-01  1:53             ` Jeff King
2011-02-01  1:00           ` Sverre Rabbelier
2011-02-01  1:57             ` Jeff King
2011-02-01  7:24           ` Jay Soffian
2011-02-01 14:42         ` Andreas Ericsson
2011-01-31 21:59     ` [1.8.0] 't/' is standard name for directory with tests Jakub Narebski
2011-01-31 22:32       ` Nicolas Pitre
2011-02-01  0:12         ` Alex Budovski
2011-02-01  0:33           ` Nicolas Pitre
2011-02-01  0:58             ` Jakub Narebski
2011-02-01  1:15             ` Junio C Hamano
2011-02-02 23:55               ` Sam Vilain
2011-02-01 18:26     ` [1.8.0] split largest remaining scripts, gitk and gitweb Jakub Narebski
2011-02-01 22:15       ` Junio C Hamano
2011-02-01 23:20         ` Jakub Narebski
2011-02-05  3:21     ` [1.8.0] reorganize the mess that the source tree has become Martin von Zweigbergk
2011-01-31 21:44   ` [1.8.0] make two-argument fetch update remote branches Thomas Rast
2011-01-31 22:18     ` Matthieu Moy
2011-01-31 22:24     ` Junio C Hamano
2011-01-31 22:27     ` Eugene Sajine
2011-01-31 23:06       ` Junio C Hamano
2011-01-31 23:39         ` Eugene Sajine
2011-02-01  1:13           ` Junio C Hamano
2011-01-31 23:22     ` Jeff King
2011-02-01  7:04     ` Jay Soffian
2011-02-01 15:58       ` Nguyen Thai Ngoc Duy
2011-02-01 22:09       ` Junio C Hamano
2011-02-01 21:05     ` A Large Angry SCM
2011-02-01 22:39       ` Thomas Rast
2011-02-01 23:25         ` A Large Angry SCM
2011-01-31 21:55   ` [1.8.0] forbid full fetchspecs in git-pull Thomas Rast
2011-01-31 22:38     ` Junio C Hamano
2011-01-31 23:15     ` Dmitry Potapov
2011-02-01 15:14       ` Thomas Rast
2011-02-01 20:23         ` Dmitry Potapov
2011-02-01  3:20   ` Planning for 1.7.5 and 1.8.0 Nguyen Thai Ngoc Duy
2011-02-01  4:16     ` Nicolas Pitre
2011-02-01 14:54     ` [1.8.0] Tag namespaces Marc Branchaud
2011-02-01 15:21       ` Nguyen Thai Ngoc Duy [this message]
2011-02-01 18:37   ` [1.8.0] Remove deprecated commands René Scharfe
2011-02-01 22:16     ` Junio C Hamano
2011-02-02  0:57     ` Jonathan Nieder
2011-02-10 19:42       ` René Scharfe
2011-02-10 20:56         ` Jonathan Nieder
2011-02-10 21:08           ` Junio C Hamano
2011-02-12 13:24           ` René Scharfe
2011-02-12 21:04             ` Jonathan Nieder
2011-02-13 23:14             ` Junio C Hamano
2011-02-01 21:41   ` [1.8.0] Handle submodule config options consistently in diff plumbing Jens Lehmann
2011-02-02 11:56   ` [1.8.0] Tracking empty directories Jakub Narebski
2011-02-02 23:23     ` Jay Soffian
2011-02-02 23:33       ` David Aguilar
2011-02-02 23:52         ` Jakub Narebski
2011-02-03  2:21         ` Wesley J. Landaker
2011-02-03  5:53           ` Jonathan Nieder
2011-02-03 10:07           ` Matthieu Moy
2011-02-05  7:43         ` Pete Harlan
2011-02-05 18:31     ` Thomas Koch
2011-02-05 19:00       ` Sverre Rabbelier
2011-02-05 22:37         ` Jared Hance
2011-02-06 20:41         ` Junio C Hamano
2011-02-06 20:46           ` Sverre Rabbelier
2011-02-06  4:42       ` Nguyen Thai Ngoc Duy
2011-02-02 17:23   ` [1.8.0] git-stash invocation changes Thomas Rast
2011-02-02 17:35     ` Shawn Pearce
2011-02-02 18:15       ` Matthieu Moy
2011-02-02 18:51         ` Thomas Rast
2011-02-09 14:35           ` Pat Notz
2011-02-23 19:54   ` [1.8.0] Don't copy "submodule.<name>.update" to .git/config on submodule init Jens Lehmann
2011-02-23 20:28     ` Junio C Hamano
2011-02-23 22:43       ` Jens Lehmann
2011-02-24  0:34         ` Junio C Hamano
2011-02-24 23:44           ` Jens Lehmann

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='AANLkTi=kwU+-UxfnCnDj0dD2NLPovTTsCPbSf-RQ3L5P@mail.gmail.com' \
    --to=pclouds@gmail.com \
    --cc=git@vger.kernel.org \
    --cc=marcnarc@xiplink.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 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.