git@vger.kernel.org mailing list mirror (one of many)
 help / color / mirror / code / Atom feed
From: Michael Haggerty <mhagger@alum.mit.edu>
To: David Turner <dturner@twopensource.com>, git@vger.kernel.org
Cc: Jeff King <peff@peff.net>
Subject: Re: [PATCH v2 40/43] refs: allow ref backend to be set for clone
Date: Mon, 05 Oct 2015 11:32:14 +0200	[thread overview]
Message-ID: <5612439E.4080200@alum.mit.edu> (raw)
In-Reply-To: <1443477738-32023-41-git-send-email-dturner@twopensource.com>

On 09/29/2015 12:02 AM, David Turner wrote:
> Add a new option, --refs-backend-type, to allow the ref backend type to
> be set on new clones.
> 
> Submodules must use the same ref backend as the parent repository, so
> we also pass the --refs-backend-type option option when cloning
> submodules.

If I'm reading this correctly, you propose to add a new configuration
setting,

    core.refs-backend-type

to record which reference backend the repository is using.

(By the way, please document any new config settings in
Documentation/config.txt.)

This does not provide adequate backwards compatibility. Suppose I create
a repository with a DB backend, then later try to access the repository
using an older version of Git (or JGit or libgit2 or ...) that doesn't
know about this configuration setting. It will think it is a normal
repository that only has a couple of references (HEAD, stash, etc). If I
run `git gc` using the older version of git, *poof* all of the objects
referred to by other branches will be garbage-collected.

Therefore, I don't think this can be merged without a bump to
core.repositoryformatversion. Such a bump will tell well-behaved older
Git clients keep their hands off the repository. (Of course repositories
that use the files backend can continue using
core.repositoryformatversion 0.)

I thought Peff proposed a way to do such a bump, including a way to
extend repositories one by one with new features. But that was something
that we were chatting about off-list.

I haven't reviewed the actual code in this patch yet but I wanted to get
the above comment on your radar.

Michael

-- 
Michael Haggerty
mhagger@alum.mit.edu

  reply	other threads:[~2015-10-05  9:32 UTC|newest]

Thread overview: 90+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-09-28 22:01 [PATCH v2 00/43] lmdb ref backend David Turner
2015-09-28 22:01 ` [PATCH v2 01/43] refs.c: create a public version of verify_refname_available David Turner
2015-10-03  5:02   ` Torsten Bögershausen
2015-10-03 16:50     ` David Turner
2015-10-03 21:07       ` Torsten Bögershausen
2015-10-04  6:07       ` Torsten Bögershausen
2015-10-05  4:29   ` Michael Haggerty
2015-10-05 20:23     ` David Turner
2015-09-28 22:01 ` [PATCH v2 02/43] refs: make repack_without_refs and is_branch public David Turner
2015-10-05  4:34   ` Michael Haggerty
2015-10-05 20:26     ` David Turner
2015-09-28 22:01 ` [PATCH v2 03/43] refs-be-files.c: rename refs to refs-be-files David Turner
2015-09-28 22:01 ` [PATCH v2 04/43] refs.c: add a new refs.c file to hold all common refs code David Turner
2015-09-28 22:01 ` [PATCH v2 05/43] refs.c: move update_ref to refs.c David Turner
2015-09-28 22:01 ` [PATCH v2 06/43] refs.c: move delete_ref and delete_refs to the common code David Turner
2015-09-28 22:01 ` [PATCH v2 07/43] refs.c: move read_ref_at to the common refs file David Turner
2015-09-28 22:01 ` [PATCH v2 08/43] refs.c: move the hidden refs functions to the common code David Turner
2015-09-28 22:01 ` [PATCH v2 09/43] refs.c: move dwim and friend functions to the common refs code David Turner
2015-09-28 22:01 ` [PATCH v2 10/43] refs.c: move warn_if_dangling_symref* to the common code David Turner
2015-09-28 22:01 ` [PATCH v2 11/43] refs.c: move read_ref, read_ref_full and ref_exists " David Turner
2015-09-28 22:01 ` [PATCH v2 12/43] refs.c: move resolve_refdup to common David Turner
2015-09-28 22:01 ` [PATCH v2 13/43] refs.c: move check_refname_format to the common code David Turner
2015-09-28 22:01 ` [PATCH v2 14/43] refs.c: move is_branch " David Turner
2015-09-28 22:01 ` [PATCH v2 15/43] refs.c: move prettify_refname " David Turner
2015-09-28 22:01 ` [PATCH v2 16/43] refs.c: move ref iterators " David Turner
2015-09-28 22:01 ` [PATCH v2 17/43] refs.c: move head_ref_namespaced " David Turner
2015-09-28 22:01 ` [PATCH v2 18/43] refs-be-files.c: add a backend method structure with transaction functions David Turner
2015-10-05  8:03   ` Michael Haggerty
2015-10-05 17:25     ` Junio C Hamano
2015-10-06  0:20       ` David Turner
2015-09-28 22:01 ` [PATCH v2 19/43] refs-be-files.c: add methods for misc ref operations David Turner
2015-09-28 22:01 ` [PATCH v2 20/43] refs-be-files.c: add methods for the ref iterators David Turner
2015-09-28 22:01 ` [PATCH v2 21/43] refs-be-files.c: add method for for_each_reftype_ David Turner
2015-09-28 22:01 ` [PATCH v2 22/43] refs-be-files.c: add do_for_each_per_worktree_ref David Turner
2015-10-05  8:19   ` Michael Haggerty
2015-10-05 20:14     ` David Turner
2015-09-28 22:01 ` [PATCH v2 23/43] refs.c: move refname_is_safe to the common code David Turner
2015-09-28 22:01 ` [PATCH v2 24/43] refs.h: document make refname_is_safe and add it to header David Turner
2015-09-28 22:02 ` [PATCH v2 25/43] refs.c: move copy_msg to the common code David Turner
2015-09-28 22:02 ` [PATCH v2 26/43] refs.c: move peel_object " David Turner
2015-09-28 22:02 ` [PATCH v2 27/43] refs.c: move should_autocreate_reflog to " David Turner
2015-10-02 20:57   ` Junio C Hamano
2015-09-28 22:02 ` [PATCH v2 28/43] refs.c: add ref backend init function David Turner
2015-10-05  8:37   ` Michael Haggerty
2015-10-05 20:37     ` David Turner
2015-09-28 22:02 ` [PATCH v2 29/43] refs.c: add methods for reflog David Turner
2015-09-28 22:02 ` [PATCH v2 30/43] refs-be-files.c: add method to expire reflogs David Turner
2015-10-05  8:41   ` Michael Haggerty
2015-09-28 22:02 ` [PATCH v2 31/43] refs.c: add method for initial ref transaction commit David Turner
2015-09-28 22:02 ` [PATCH v2 32/43] initdb: move safe_create_dir into common code David Turner
2015-09-28 22:02 ` [PATCH v2 33/43] refs.c: add method for initializing refs db David Turner
2015-09-28 22:02 ` [PATCH v2 34/43] refs.c: make struct ref_transaction generic David Turner
2015-10-06 17:43   ` Michael Blume
2015-10-06 17:53     ` David Turner
2015-09-28 22:02 ` [PATCH v2 35/43] refs-be-files.c: add method to rename refs David Turner
2015-09-28 22:02 ` [PATCH v2 36/43] run-command: track total number of commands run David Turner
2015-09-28 22:02 ` [PATCH v2 37/43] refs: move some defines from refs-be-files.c to refs.h David Turner
2015-10-05  8:57   ` Michael Haggerty
2015-09-28 22:02 ` [PATCH v2 38/43] refs: make some files backend functions public David Turner
2015-10-05  9:03   ` Michael Haggerty
2015-10-06  1:24     ` David Turner
2015-10-07  1:25     ` David Turner
2015-10-07 16:00       ` Michael Haggerty
2015-10-07 17:20         ` Junio C Hamano
2015-10-07 20:55         ` David Turner
2015-10-07 22:47           ` Michael Haggerty
2015-09-28 22:02 ` [PATCH v2 39/43] refs: break out a ref conflict check David Turner
2015-10-05  9:06   ` Michael Haggerty
2015-10-06  0:28     ` David Turner
2015-09-28 22:02 ` [PATCH v2 40/43] refs: allow ref backend to be set for clone David Turner
2015-10-05  9:32   ` Michael Haggerty [this message]
2015-10-06  1:29     ` David Turner
2015-10-06  1:58       ` Jeff King
2015-10-06 18:09         ` David Turner
2015-10-12  9:00           ` Carlos Martín Nieto
2015-10-05 11:55   ` Michael Haggerty
2015-10-06  1:16     ` David Turner
2015-09-28 22:02 ` [PATCH v2 41/43] refs: add register_refs_backend David Turner
2015-09-28 22:02 ` [PATCH v2 42/43] refs: add LMDB refs backend David Turner
2015-10-02 21:35   ` Junio C Hamano
2015-10-05 15:47   ` Michael Haggerty
2015-10-07  1:51     ` David Turner
2015-10-07 18:31       ` Michael Haggerty
2015-10-07 19:08         ` Junio C Hamano
2015-10-07 19:20         ` David Turner
2015-10-07 22:13           ` Michael Haggerty
2015-09-28 22:02 ` [PATCH v2 43/43] refs: tests for db backend David Turner
2015-10-03 17:39   ` Dennis Kaarsemaker
2015-10-05 16:56     ` Junio C Hamano
2015-10-06  0:20       ` David Turner

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=5612439E.4080200@alum.mit.edu \
    --to=mhagger@alum.mit.edu \
    --cc=dturner@twopensource.com \
    --cc=git@vger.kernel.org \
    --cc=peff@peff.net \
    /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).