Skip Menu |

This queue is for tickets about the Firefox-Marionette CPAN distribution.

Report information
The Basics
Id: 124076
Status: resolved
Priority: 0/
Queue: Firefox-Marionette

People
Owner: Nobody in particular
Requestors: SREZIC [...] cpan.org
Cc:
AdminCc:

Bug Information
Severity: (no value)
Broken in: (no value)
Fixed in: (no value)



Subject: Can't locate ./inc/Test/NeedsDisplay.pm (0.05+0.06)
The test suite fails, probably because Test/NeedsDisplay.pm is not included in inc: Can't locate ./inc/Test/NeedsDisplay.pm in @INC (you may need to install the .::inc::Test::NeedsDisplay module) (@INC contains: /home/eserte/.cpan/build/2018011409/Firefox-Marionette-0.06-IXHKSd/blib/lib /home/eserte/.cpan/build/2018011409/Firefox-Marionette-0.06-IXHKSd/blib/arch /home/eserte/.cpan/build/2018011409/Test-NeedsDisplay-1.07-KLEixG/blib/arch /home/eserte/.cpan/build/2018011409/Test-NeedsDisplay-1.07-KLEixG/blib/lib /home/eserte/.cpan/build/2018011409/Test-NeedsDisplay-1.07-KLEixG/blib/arch /home/eserte/.cpan/build/2018011409/Test-NeedsDisplay-1.07-KLEixG/blib/lib /opt/perl-5.18.4/lib/site_perl/5.18.4/x86_64-linux /opt/perl-5.18.4/lib/site_perl/5.18.4 /opt/perl-5.18.4/lib/5.18.4/x86_64-linux /opt/perl-5.18.4/lib/5.18.4 .) at t/01-marionette.t line 7. BEGIN failed--compilation aborted at t/01-marionette.t line 10. t/01-marionette.t .. Dubious, test returned 2 (wstat 512, 0x200) No subtests run And actually, is there a need for bundling a patched Test::NeedsDisplay?
On Sun Jan 14 19:47:50 2018, SREZIC wrote: Show quoted text
> The test suite fails, probably because Test/NeedsDisplay.pm is not > included in inc: > > Can't locate ./inc/Test/NeedsDisplay.pm in @INC (you may need to > install the .::inc::Test::NeedsDisplay module) (@INC contains: > /home/eserte/.cpan/build/2018011409/Firefox-Marionette-0.06- > IXHKSd/blib/lib /home/eserte/.cpan/build/2018011409/Firefox- > Marionette-0.06-IXHKSd/blib/arch > /home/eserte/.cpan/build/2018011409/Test-NeedsDisplay-1.07- > KLEixG/blib/arch /home/eserte/.cpan/build/2018011409/Test- > NeedsDisplay-1.07-KLEixG/blib/lib > /home/eserte/.cpan/build/2018011409/Test-NeedsDisplay-1.07- > KLEixG/blib/arch /home/eserte/.cpan/build/2018011409/Test- > NeedsDisplay-1.07-KLEixG/blib/lib /opt/perl- > 5.18.4/lib/site_perl/5.18.4/x86_64-linux /opt/perl- > 5.18.4/lib/site_perl/5.18.4 /opt/perl-5.18.4/lib/5.18.4/x86_64-linux > /opt/perl-5.18.4/lib/5.18.4 .) at t/01-marionette.t line 7. > BEGIN failed--compilation aborted at t/01-marionette.t line 10. > t/01-marionette.t .. > Dubious, test returned 2 (wstat 512, 0x200) > No subtests run > > > And actually, is there a need for bundling a patched > Test::NeedsDisplay?
Included the bundled Test::NeedsDisplay in 0.07. I'm trying to debug the reason behind the "ABORT: X_GetImage: BadMatch (invalid parameter attributes)" crashes in cpan testers results such as http://www.cpantesters.org/cpan/report/bac4c28a-f883-11e7-9eec-c8dc58f925d1. One theory is that Test::NeedsDisplay should be using 24 bit colours as detailed in https://bugs.chromium.org/p/chromedriver/issues/detail?id=394 which is the change i have made to the locally packaged Test::NeedsDisplay. If this theory is supported by cpan testers, i'll supply a patch to Test::NeedsDisplay. Thanks for all your cpan testing work!
On 2018-01-14 05:11:38, DDICK wrote: Show quoted text
> On Sun Jan 14 19:47:50 2018, SREZIC wrote:
> > The test suite fails, probably because Test/NeedsDisplay.pm is not > > included in inc: > > > > Can't locate ./inc/Test/NeedsDisplay.pm in @INC (you may need to > > install the .::inc::Test::NeedsDisplay module) (@INC contains: > > /home/eserte/.cpan/build/2018011409/Firefox-Marionette-0.06- > > IXHKSd/blib/lib /home/eserte/.cpan/build/2018011409/Firefox- > > Marionette-0.06-IXHKSd/blib/arch > > /home/eserte/.cpan/build/2018011409/Test-NeedsDisplay-1.07- > > KLEixG/blib/arch /home/eserte/.cpan/build/2018011409/Test- > > NeedsDisplay-1.07-KLEixG/blib/lib > > /home/eserte/.cpan/build/2018011409/Test-NeedsDisplay-1.07- > > KLEixG/blib/arch /home/eserte/.cpan/build/2018011409/Test- > > NeedsDisplay-1.07-KLEixG/blib/lib /opt/perl- > > 5.18.4/lib/site_perl/5.18.4/x86_64-linux /opt/perl- > > 5.18.4/lib/site_perl/5.18.4 /opt/perl-5.18.4/lib/5.18.4/x86_64-linux > > /opt/perl-5.18.4/lib/5.18.4 .) at t/01-marionette.t line 7. > > BEGIN failed--compilation aborted at t/01-marionette.t line 10. > > t/01-marionette.t .. > > Dubious, test returned 2 (wstat 512, 0x200) > > No subtests run > > > > > > And actually, is there a need for bundling a patched > > Test::NeedsDisplay?
> > Included the bundled Test::NeedsDisplay in 0.07. > > I'm trying to debug the reason behind the "ABORT: X_GetImage: BadMatch > (invalid parameter attributes)" crashes in cpan testers results such > as http://www.cpantesters.org/cpan/report/bac4c28a-f883-11e7-9eec- > c8dc58f925d1. > > One theory is that Test::NeedsDisplay should be using 24 bit colours > as detailed in > https://bugs.chromium.org/p/chromedriver/issues/detail?id=394 which is > the change i have made to the locally packaged Test::NeedsDisplay. > > If this theory is supported by cpan testers, i'll supply a patch to > Test::NeedsDisplay.
All fail reports come from my Debian/jessie workstation, where the tests run in a vnc session (so the X11 windows popping up by the Test::NeedsDisplay and Firefox::Marionette test suites don't disturb my normal work). However, the test suite passes for most perl versions installed on this system. And running the test suite again for the failing ones makes it pass again. So it's a random test failure, which does not seem to be related to the X11 environment.
On Sun Jan 14 21:39:31 2018, SREZIC wrote: Show quoted text
> On 2018-01-14 05:11:38, DDICK wrote:
> > On Sun Jan 14 19:47:50 2018, SREZIC wrote:
> > > The test suite fails, probably because Test/NeedsDisplay.pm is not > > > included in inc: > > > > > > Can't locate ./inc/Test/NeedsDisplay.pm in @INC (you may need to > > > install the .::inc::Test::NeedsDisplay module) (@INC contains: > > > /home/eserte/.cpan/build/2018011409/Firefox-Marionette-0.06- > > > IXHKSd/blib/lib /home/eserte/.cpan/build/2018011409/Firefox- > > > Marionette-0.06-IXHKSd/blib/arch > > > /home/eserte/.cpan/build/2018011409/Test-NeedsDisplay-1.07- > > > KLEixG/blib/arch /home/eserte/.cpan/build/2018011409/Test- > > > NeedsDisplay-1.07-KLEixG/blib/lib > > > /home/eserte/.cpan/build/2018011409/Test-NeedsDisplay-1.07- > > > KLEixG/blib/arch /home/eserte/.cpan/build/2018011409/Test- > > > NeedsDisplay-1.07-KLEixG/blib/lib /opt/perl- > > > 5.18.4/lib/site_perl/5.18.4/x86_64-linux /opt/perl- > > > 5.18.4/lib/site_perl/5.18.4 /opt/perl-5.18.4/lib/5.18.4/x86_64- > > > linux > > > /opt/perl-5.18.4/lib/5.18.4 .) at t/01-marionette.t line 7. > > > BEGIN failed--compilation aborted at t/01-marionette.t line 10. > > > t/01-marionette.t .. > > > Dubious, test returned 2 (wstat 512, 0x200) > > > No subtests run > > > > > > > > > And actually, is there a need for bundling a patched > > > Test::NeedsDisplay?
> > > > Included the bundled Test::NeedsDisplay in 0.07. > > > > I'm trying to debug the reason behind the "ABORT: X_GetImage: > > BadMatch > > (invalid parameter attributes)" crashes in cpan testers results such > > as http://www.cpantesters.org/cpan/report/bac4c28a-f883-11e7-9eec- > > c8dc58f925d1. > > > > One theory is that Test::NeedsDisplay should be using 24 bit colours > > as detailed in > > https://bugs.chromium.org/p/chromedriver/issues/detail?id=394 which > > is > > the change i have made to the locally packaged Test::NeedsDisplay. > > > > If this theory is supported by cpan testers, i'll supply a patch to > > Test::NeedsDisplay.
> > All fail reports come from my Debian/jessie workstation, where the > tests run in a vnc session (so the X11 windows popping up by the > Test::NeedsDisplay and Firefox::Marionette test suites don't disturb > my normal work). However, the test suite passes for most perl versions > installed on this system. And running the test suite again for the > failing ones makes it pass again. So it's a random test failure, which > does not seem to be related to the X11 environment.
Thanks for that information, that's very useful to know. Extremely annoying information as well, but your assistance is most appreciated. I'll try to code around random test failures.