Re: [pygame] Automated Builds

["Brian Fisher" <brian@xxxxxxxxxxxxxxxxxxx>]

sadly, the test failures haven't all been fixed...
The problem with the automated build page was that it wasn't smart at parsing test output (we never had so many failures before :)

It's smarter now:
http://thorbrian.com/pygame/builds.php

... looks like windows still has the gl screen save crash, all python versions have the test_get_init and the test_subsurface_lock problems, and python 2.4 has 2 color test failures and a mixer quit test failure.

On Mon, Jun 9, 2008 at 6:30 PM, Nicholas Dudfield <ndudfield@xxxxxxxxx> wrote:

The call to pygame.image.save(gl_surface) segfaults on windows(and only
windows) for a reason unbeknownst to me. I wrapped it in an os.system() call
in another process so hopefully it shouldn't crash the builds.

So there are problems now with failing tests not showing up?

They haven't all been fixed over night have they? Fingers crossed.

-----Original Message-----
From: owner-pygame-users@xxxxxxxx [mailto:owner-pygame-users@xxxxxxxx] On
Behalf Of Lenard Lindstrom
Sent: Tuesday, June 10, 2008 5:12 AM
To: pygame-users@xxxxxxxx
Subject: Re: [pygame] Automated Builds

I see 'The Spectacularly Adequate Automated Pygame Build Page' shows
revision 1293 as passing all tests for Python 2.4 on Windows. Yet I get one
error and four fails on the unit tests. The GL segfault is still there. The
results are included as an attachment. This is using the installer from that
page after uninstalling all previous Pygame versions. The error and fails
are also present for my own build from a fresh SVN copy. I ran the tests on
Windows 98.

Lenard


Lenard Lindstrom wrote:
> Ah, so that error isn't just showing up on my computer.
>
> Lenard
>
>
> Brian Fisher wrote:
>> yeah, one of the unit tests crashes on windows, and it stops the builds.
>>
>> I haven't figure out a good way to deal with that yet. I think I need
>> to implement a timeout on the guy that calls the unit tests.
>>
>> I haven't figured out why they crash yet.
>>
>> So why is there a segfault on the GL save?
>>
>> On Sun, Jun 8, 2008 at 11:54 PM, Nicholas Dudfield
>> <ndudfield@xxxxxxxxx <mailto:ndudfield@xxxxxxxxx>> wrote:
>>
>>     Brian,
>>
>>     Just a quick note to let you know that the windows builds are
>>     lagging behind
>>     the osx.
>>
>>     At time of writing;
>>
>>            Most recent Automated build results:
>>
>>            OSX:            r1293
>>            Windows:        r1272
>>
>>     Is it possible the segmentation fault
>> (pygame.image.save(gl_surface))
>>     stopped the builds?
>>
>>