From: John Keeping <john@keeping.me.uk>
To: Johannes Schindelin <johannes.schindelin@gmx.de>
Cc: Joey Hess <id@joeyh.name>, git@vger.kernel.org
Subject: Re: undocumented core.sharedRepository=2 set by git init --shared=world
Date: Tue, 7 Jul 2015 12:52:05 +0100 [thread overview]
Message-ID: <20150707115205.GD18226@serenity.lan> (raw)
In-Reply-To: <b1139f60dde3533e12d13e4b0f6584f2@www.dscho.org>
On Tue, Jul 07, 2015 at 01:32:13PM +0200, Johannes Schindelin wrote:
> Hi John,
>
> On 2015-07-07 11:10, John Keeping wrote:
> > On Tue, Jul 07, 2015 at 10:14:28AM +0200, Johannes Schindelin wrote:
> >> Hi Joey,
> >>
> >> On 2015-07-06 21:25, Joey Hess wrote:
> >> > joey@darkstar:~/tmp>git init --shared=world testrepo
> >> > Initialized empty shared Git repository in /home/joey/tmp/testrepo/.git/
> >> > joey@darkstar:~/tmp>grep shared testrepo/.git/config
> >> > sharedrepository = 2
> >> >
> >> > This magic value of 2 seems to be undocumented, as is the magic value of 1
> >> > that's equvilant to "group".
> >> >
> >> > I think it would be better to have git init put in "world" or "group" and not
> >> > these magic values. Anyway, I suppose they ought to be documented too.
> >>
> >> The rationale can be found here:
> >> https://github.com/git/git/blob/v2.4.5/builtin/init-db.c#L413-L418
> >>
> >> /* We do not spell "group" and such, so that
> >> * the configuration can be read by older version
> >> * of git. Note, we use octal numbers for new share modes,
> >> * and compatibility values for PERM_GROUP and
> >> * PERM_EVERYBODY.
> >> */
> >>
> >> I am sympathetic to your wish, of course, and I am sure that you
> >> understand why we cannot simply break other people's setups to satisfy
> >> it.
> >
> > That comment was added in 94df250 (shared repository: optionally allow
> > reading to "others"., 2006-06-09) which was in 1.4.1. I suspect that is
> > now sufficiently old that it no longer matters.
>
> I understand your point of view. With my maintainer hat on I have to
> say, though, that things like that require a major version change.
> Users tend to appreciate such a careful maintenance.
However, there has been a major version since the new syntax was
introduced (in the same commit mentioned above), so this only affects
users who initialize a repository with (say) 2.6.0 or later and then try
to use 1.4.0 or earlier to operate on it.
That means using two versions of Git released more than 9 years apart to
operate on the same repository. IMHO even careful maintenance can
declare that an unsupported configuration.
prev parent reply other threads:[~2015-07-07 11:52 UTC|newest]
Thread overview: 6+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-07-06 19:25 undocumented core.sharedRepository=2 set by git init --shared=world Joey Hess
2015-07-07 6:44 ` Torsten Bögershausen
2015-07-07 8:14 ` Johannes Schindelin
2015-07-07 9:10 ` John Keeping
2015-07-07 11:32 ` Johannes Schindelin
2015-07-07 11:52 ` John Keeping [this message]
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=20150707115205.GD18226@serenity.lan \
--to=john@keeping.me.uk \
--cc=git@vger.kernel.org \
--cc=id@joeyh.name \
--cc=johannes.schindelin@gmx.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 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).