Re: [PATCH v2] t0051: use "skip_all" under !MINGW in single-test file

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



Hi Ævar,

On Fri, 4 Feb 2022, Ævar Arnfjörð Bjarmason wrote:

> Have this file added in 06ba9d03e34 (t0051: test GIT_TRACE to a
> windows named pipe, 2018-09-11) use the same "skip_all" pattern as an
> existing Windows-only test added in 0e218f91c29 (mingw: unset PERL5LIB
> by default, 2018-10-30) uses.

This is not a nit, even if I won't insist on changing it in this instance:
This sentence is unnecessarily convoluted and therefore harder to read
than it has to be. Please pay attention to readability next time you craft
a commit message.

The rest of the commit message as well as the diff look good to me.

Thanks,
Johannes

>
> This way TAP consumers like "prove" will show a nice summary when the
> test is skipped. Instead of:
>
>     $ prove t0051-windows-named-pipe.sh
>     [...]
>     t0051-windows-named-pipe.sh .. ok
>     [...]
>
> We will prominently show a "skipped" notice:
>
>     $ prove t0051-windows-named-pipe.sh
>     [...]
>     t0051-windows-named-pipe.sh ... skipped: skipping Windows-specific tests
>     [...]
>
> This is because we are now making use of the right TAP-y way to
> communicate this to the consumer. I.e. skipping the whole test file,
> v.s. skipping individual tests (in this case there's only one test).
>
> Signed-off-by: Ævar Arnfjörð Bjarmason <avarab@xxxxxxxxx>
> ---
> Range-diff against v1:
> 1:  34ff968dcb8 ! 1:  1bc93bcba4b t0051: use "skip_all" under !MINGW in single-test file
>     @@ Commit message
>          by default, 2018-10-30) uses.
>
>          This way TAP consumers like "prove" will show a nice summary when the
>     -    test is skipped, e.g.:
>     +    test is skipped. Instead of:
>     +
>     +        $ prove t0051-windows-named-pipe.sh
>     +        [...]
>     +        t0051-windows-named-pipe.sh .. ok
>     +        [...]
>     +
>     +    We will prominently show a "skipped" notice:
>
>              $ prove t0051-windows-named-pipe.sh
>              [...]
>              t0051-windows-named-pipe.sh ... skipped: skipping Windows-specific tests
>              [...]
>
>     +    This is because we are now making use of the right TAP-y way to
>     +    communicate this to the consumer. I.e. skipping the whole test file,
>     +    v.s. skipping individual tests (in this case there's only one test).
>     +
>          Signed-off-by: Ævar Arnfjörð Bjarmason <avarab@xxxxxxxxx>
>
>       ## t/t0051-windows-named-pipe.sh ##
>
>  t/t0051-windows-named-pipe.sh | 7 ++++++-
>  1 file changed, 6 insertions(+), 1 deletion(-)
>
> diff --git a/t/t0051-windows-named-pipe.sh b/t/t0051-windows-named-pipe.sh
> index 10ac92d2250..412f413360d 100755
> --- a/t/t0051-windows-named-pipe.sh
> +++ b/t/t0051-windows-named-pipe.sh
> @@ -3,8 +3,13 @@
>  test_description='Windows named pipes'
>
>  . ./test-lib.sh
> +if ! test_have_prereq MINGW
> +then
> +	skip_all='skipping Windows-specific tests'
> +	test_done
> +fi
>
> -test_expect_success MINGW 'o_append write to named pipe' '
> +test_expect_success 'o_append write to named pipe' '
>  	GIT_TRACE="$(pwd)/expect" git status >/dev/null 2>&1 &&
>  	{ test-tool windows-named-pipe t0051 >actual 2>&1 & } &&
>  	pid=$! &&
> --
> 2.35.1.940.ge7a5b4b05f2
>
>

[Index of Archives]     [Linux Kernel Development]     [Gcc Help]     [IETF Annouce]     [DCCP]     [Netdev]     [Networking]     [Security]     [V4L]     [Bugtraq]     [Yosemite]     [MIPS Linux]     [ARM Linux]     [Linux Security]     [Linux RAID]     [Linux SCSI]     [Fedora Users]

  Powered by Linux