From: Nguyen Thai Ngoc Duy <pclouds@gmail.com>
Date: Sun, 28 Mar 2010 07:11:21 GMT
Subject: Re: [PATCH] builtins: reset startup_info->have_run_setup_gitdir when unsetting up repository
Message-ID: <fcaeb9bf1003280011r4c25a076t22d8759de31c3e3f@mail.gmail.com>
In-Reply-To: <20100327223813.GA5809@progeny.tock>

2010/3/27 Jonathan Nieder <jrnieder@gmail.com>:
> Hi again,
>
> Nguyễn Thái Ngọc Duy wrote:
>
>> While at it, also reset repository_format_version to zero. When
>> omitted, the format should be understood as the last supported
>> version, i.e. zero. This is probably only used by "git init" or "git
>> clone".
> [...]
>>       /* Initialized in check_repository_format_version() */
>> -     repository_format_version = 0xFF;
>> +     repository_format_version = 0;
>
> Good change, but wrong justification in my opinion.
>
> v0.99.9l^2~54 (init-db: check template and repository format.,
> 2005-11-25) taught ‘git init-db’ to make sure that
>
>  repository_format_version <= GIT_REPO_VERSION
>
> before initializing a new repository.  repository_format_version was
> being explicitly initialized globally to 0 at the time, presumably to
> ensure tests like this always succeed when no repository format version
> is declared (or in other words, the repository format for repositories
> initialized before git v0.99.9l^2~56 is zero by convention).  As a happy
> side effect, that default takes care of the “no pre-existing repository”
> case here.
>
> The explicit initialization was removed in commit v1.4.3-rc1~230 (remove
> unnecessary initializations, 2006-08-15), since according to ANSI C it
> is redundant.
>
> So I think the convention is not “if in doubt, the repository has
> version GIT_REPO_VERSION” but “if the repository lacks a
> core.repositoryversion setting, it must be really old”.

Makes sense. Thanks. Will leave a note at the initialization to reduce
history digging time for other people.
-- 
Duy