From: Tarmigan <tarmigan+git@gmail.com>
Date: Thu, 17 Apr 2008 16:31:52 GMT
Subject: Re: [RFH] cleaning up "add across symlinks"
Message-ID: <905315640804170931v23df8633xf51276a5858eabc7@mail.gmail.com>
In-Reply-To: <905315640804161426l442c2812k3c36c1bafe484157@mail.gmail.com>

On Wed, Apr 16, 2008 at 2:26 PM, Tarmigan <tarmigan+git@gmail.com> wrote:
> On Wed, Apr 16, 2008 at 1:53 PM, Junio C Hamano <gitster@pobox.com> wrote:
>  > If you have this structure in your work tree:
>  >
>  >         lrwxrwxrwx a -> c
>  >         drwxrwxrwx c
>  >         -rw-rw-rw- c/b
>  >
>  >  and let million monkeys give random paths to "git-update-index --add" or
>  >  "git add", you should end up with the index with two entries, a symlink
>  >  "a" and file "c/b".
>  >
>  >  Not so.  If an unfortunate monkey says "git add a/b", we happily add it to
>  >  the index, because we notice lstat("a/b") succeeds and assume that there
>  >  is such a path.  There isn't, as far as git is concerned, because we track
>  >  symbolic links.
>
>  Thanks Junio, I'll try to do some testing with it later.

I tested with all the cases I could think of and it works for me.

>  That's almost the case I used.  The exact test to add to these,
>  without the '*' after a/, would be something like this (warning: cut
>  and paste):
>
>  +test_expect_success 'add confusion (4)' '
>  +
>  +       test_must_fail git add "a/" &&

Currently, this will succeed, so if you decide to add this case, the
test_must_fail should go.

Thanks,
Tarmigan