[FFmpeg-devel] Who is against GIT now?

[Måns Rullgård]

Michael Niedermayer <michaelni at gmx.at> writes:

> On Mon, Apr 06, 2009 at 03:53:05PM +0100, M?ns Rullg?rd wrote:
>> Luca Barbato <lu_zero at gentoo.org> writes:
>> 
>> > Mike Melanson wrote:
>> >
>> >> (actual hashes from my FATE code repository). So what should the
>> >> main page of FATE report regarding the state of FFmpeg code? That
>> >> version 6e6fc12c61446bfe8dff7ce3b50ebfe0b0835bc8 passed all tests
>> >> but that 8e1f4d881f813d7d2c67656868f39b46a98d320d had problems?
>> >>
>> >
>> > That would work otherwise you may use git-describe to have a reference
>> > as N commits since the last tag.
>> 
>> Note that the commit count is informative only, not unique; two
>> branches diverging after the tag could have the same number of
>> commits.  
>
>> The hash at the end is unique.
>
> I wanted to stay out of the discussion but, this one is not strictly
> true, given that there are more possible trees than hashes it cant be
> of course this matters little in practice and you could say its nitpicking
> but still, its just very likely unique ...

It has to be unique since a tree cannot contain duplicate hashes.  I
don't know how that unlikely situation would be resolved though.

-- 
M?ns Rullg?rd
mans at mansr.com