[bug#27542] [PATCH] gnu: gcr: Enable working tests.

[ludo@gnu.org (Ludovic Courtès)]

Kei Kebreau <kei@openmailbox.org> skribis:

> ludo@gnu.org (Ludovic Courtès) writes:
>
>> Kei Kebreau <kei@openmailbox.org> skribis:
>>
>>> ludo@gnu.org (Ludovic Courtès) writes:
>>>
>>>> Hello,
>>>>
>>>> Kei Kebreau <kei@openmailbox.org> skribis:
>>>>
>>>>> * gnu/packages/patches/gcr-disable-failing-tests.patch,
>>>>> gnu/packages/patches/gcr-fix-collection-tests-to-work-with-gpg-21.patch:
>>>>> New files.
>>>>> * gnu/local.mk (dist_patch_DATA): Add patches.
>>>>> * gnu/packages/gnome.scm (gcr)[source]: Use patches.
>>>>> [arguments]: Enable tests.
>>>>
>>>> [...]
>>>>
>>>>> +++ b/gnu/packages/patches/gcr-disable-failing-tests.patch
>>>>> @@ -0,0 +1,127 @@
>>>>> +This patch removes a group of tests that fail because
>>>>> /var/lib/dbus/machine-id
>>>>> +is not present in the build environment.
>>>>
>>>> We usually work around these with:
>>>>
>>>>   (setenv "DBUS_FATAL_WARNINGS" "0")
>>>>
>>>> Is it ineffective here?
>>>>
>>>
>>> That's good to know, but it doesn't seem to work in this case. I get 18
>>> test failures all from the same group mentioning the missing
>>> 'machine-id' file.
>>
>> Are they really errors and not just warnings (the errors being something
>> else)?
>>
>> Also, could it be that some Makefile or script forcefully sets
>> DBUS_FATAL_WARNINGS to 1, as is the case in ‘python-dbus’?
>>
>
> I see no such setting.

OK.  In that case your patch looks like the right thing to me.

Thank you, and sorry for introducing more delay!

Ludo’.