From: Baz <brian.ewins@gmail.com>
Date: Tue, 16 Jan 2007 13:50:30 GMT
Subject: Re: sha-1 mismatch using command line sha1sum
Message-ID: <2faad3050701160550g2d3fdf56p64b8ae1ea4d4197e@mail.gmail.com>
In-Reply-To: <200701161342.56431.andyparkins@gmail.com>

On 16/01/07, Andy Parkins <andyparkins@gmail.com> wrote:
> Hello,
>
> I came across this message:
>
> http://www.gelato.unsw.edu.au/archives/git/0504/0212.html
>
> So I thought, "hey that's interesting, I'll try it".  I was surprised to find
> this:
>
> $ sha1sum .git/objects/f4/b6c6b90fdce12d69e4ad80ff6082405ec8cfb8
> ac4c6f1d36d6e0341486789b32f86ae6f506df35  .git/objects/f4/b6c6b90fdce12d69e4ad80ff6082405ec8cfb8
>
> This is in the latest git repository.
>
> Is this correct, or should we all begin screaming and crying that our
> repositories are woefully corrupt? :-)

That changed some time ago. From the git(7) manpage:

"It's worth noting that the SHA1 hash that is used to name the object
is the hash of the original data plus this header, so sha1sum file
does not match the object name for file. (Historical note: in the dawn
of the age of git the hash was the sha1 of the compressed object.)"

Cheers,
Baz