[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[bug#51890] Vigra: disable more failing tests on core-updates-frozen
From: |
Maxim Cournoyer |
Subject: |
[bug#51890] Vigra: disable more failing tests on core-updates-frozen |
Date: |
Thu, 18 Nov 2021 08:18:38 -0500 |
User-agent: |
Gnus/5.13 (Gnus v5.13) Emacs/27.2 (gnu/linux) |
Hello,
Efraim Flashner <efraim@flashner.co.il> writes:
> On November 18, 2021 9:09:50 AM UTC, "Ludovic Courtès" <ludo@gnu.org> wrote:
>>Hi,
>>
>>Vivien Kraus <vivien@planete-kraus.eu> skribis:
>>
>>> From 4311c36dcd280ba385a3d0f33800428102bef0b8 Mon Sep 17 00:00:00 2001
>>> From: Vivien Kraus <vivien@planete-kraus.eu>
>>> Date: Tue, 16 Nov 2021 08:18:42 +0000
>>> Subject: [PATCH] gnu: vigra: Disable more failing tests.
>>>
>>> * gnu/packages/image.scm (vigra) [phase disable-broken-tests]: Also disable
>>> test2, test3, test_arraytypes and test_multidef.
>>
>>Looks like it was worked around by disabling NumPy support in Vigra, in
>>commit 94b76dc85ca3098bf4ceb0f5b884b5708b9e8134.
>>
>>I’m not sure how NumPy is used; is there a preference between these two
>>options (disabling flaky tests and removing the dependency on NumPy)?
>>
>>Thanks,
>>Ludo’.
>
> It looked to me like vigra was limping along and numpy compatibility
> was being dragged along by patches here and there. When I built the
> dependencies of vigra nothing failed to build. I'm definitely not
> against reinstating support for numpy but I'm not sure just disabling
> the tests is the right way.
I know nothing about vigra, but judging from its description "[...] It
is an image processing and analysis library that puts its main emphasis
on customizable algorithms and data structures. [...]" it seems it'd
probably make good use of Numpy? Not sure what are the consequences of
doing without; perhaps efficiency.
Thanks,
Maxim