Re: [PATCH 16/19] mingw: avoid absolute path in t0008

From: Johannes Schindelin <johannes.schindelin@gmx.de>
Date: Jan 26, 2016, 08:00 UTC
Message-ID: <alpine.DEB.2.20.1601260758070.2964@virtualbox>
In-Reply-To: <xmqq1t95fgig.fsf@gitster.mtv.corp.google.com>

Hi Junio,

On Mon, 25 Jan 2016, Junio C Hamano wrote:

Show 17 quoted lines

> Johannes Schindelin <Johannes.Schindelin@gmx.de> writes:
> 
> >> Is the fact that $global_excludes is specified using an absolute
> >> path significant to the correctness of this test script?
> >
> > Apparently not. So I followed your suggestion and made this independent
> > of the OS:
> >
> > 	https://github.com/dscho/git/commit/0b9eb308
> 
> I see many "cd" in the script, some of which come before the global
> exclude file is enabled, but others ("--stdin from subdirectory",
> for example) do run from a subdirectory.  It turns out that this
> file is used as the value of core.excludesfile.  By the time that
> file is read in setup_standard_excludes(), presumably we have
> already done setup_git_directory() and moved back to the root of the
> working tree, so I guess such a change would be harmless.

Yeah, it appears that after the entire test ran, there is only one global-excludes file, and none of the `rm` commands removes any global-excludes file.

> It somehow makes me feel dirty, though.  But that is not a specific
> problem to mingw port.

Well, I found the entire file a little hard to read ;-) But yeah, I do not think that is specific to the Windows port.

Ciao, Dscho