[PATCH] tests: drop some unnecessary globs for Windows

[Yuya Nishihara]

On Sat, 2 Feb 2019 21:06:54 -0500, Matt Harbison wrote:
> 
> > On Feb 2, 2019, at 8:28 PM, Yuya Nishihara <yuya at tcha.org> wrote:
> > 
> >> On Fri, 01 Feb 2019 22:49:59 -0500, Matt Harbison wrote:
> >> # HG changeset patch
> >> # User Matt Harbison <matt_harbison at yahoo.com>
> >> # Date 1549077873 18000
> >> #      Fri Feb 01 22:24:33 2019 -0500
> >> # Node ID 2cdacfe6002918eddc7fb75d47e42f1b26214703
> >> # Parent  1bf9a0827395c593144ffd1fd4cd0713bf0b62e6
> >> tests: drop some unnecessary globs for Windows
> >> 
> >> This changed in 1d6eae696cf1.
> >> 
> >> diff --git a/tests/test-status-color.t b/tests/test-status-color.t
> >> --- a/tests/test-status-color.t
> >> +++ b/tests/test-status-color.t
> >> @@ -31,19 +31,19 @@ hg status in repo root:
> >>   [status.unknown|? ][status.unknown|in_root]
> >> HGPLAIN disables color
> >>   $ HGPLAIN=1 hg status --color=debug
> >> -  ? a/1/in_a_1 (glob)
> >> -  ? a/in_a (glob)
> >> -  ? b/1/in_b_1 (glob)
> >> -  ? b/2/in_b_2 (glob)
> >> -  ? b/in_b (glob)
> >> +  ? a/1/in_a_1
> >> +  ? a/in_a
> >> +  ? b/1/in_b_1
> >> +  ? b/2/in_b_2
> >> +  ? b/in_b
> > 
> > Is this '\' vs '/' thing? Sounds like a bug of 1d6eae696cf1.
> 
> I believe so. Past experience says that the test harness will leave stray globs on non Windows platforms, but will remove them if there are no ‘/‘ on Windows.  And now the test harness can make that conversion without marking a glob.  I figured we didn’t care about the minor style change, because it’s already fairly inconsistent on Windows, depending on the command and options given.

We don't care minor change in error/warning messages, but I think we should
preserve the existing behavior because paths are the most important data
"hg status" provides.