On 12/13, Junio C Hamano wrote:
> Johannes Schindelin <Johannes.Schindelin@xxxxxx> writes:
>
> > ok 13 - grep tree and more pathspecs
> >
> > expecting success:
> > git init parent &&
> > test_when_finished "rm -rf parent" &&
> > echo "foobar" >"parent/fi:le" &&
> > git -C parent add "fi:le" &&
> > git -C parent commit -m "add fi:le" &&
> > ...
> > test_cmp expect actual
> >
> > ++ git init parent
> > Initialized empty Git repository in C:/git-sdk-64/usr/src/git/wip3/t/trash
> > directory.t7814-grep-recurse-submodules/parent/.git/
> > ++ test_when_finished 'rm -rf parent'
> > ++ test 0 = 0
> > ++ test_cleanup='{ rm -rf parent
> > } && (exit "$eval_ret"); eval_ret=$?; :'
> > ++ echo foobar
> > ++ git -C parent add fi:le
> > fatal: pathspec 'fi:le' did not match any files
>
> I think !MINGW prereq is missing?
Yes the !MINGW prereq is missing on this one test since the test uses a
filename with a ":" which isn't supported on windows. I have that
change made in my local grep branch but I haven't sent out a reroll of
the series yet due to the underlying race-condition that existed (due to
the way realpath was being calculated). I'll send out a reroll of the
series once the discussion on bw/realpath-wo-chdir has concluded (as
the grep series is now dependent on it).
--
Brandon Williams