On Tue, May 22, 2012 at 7:44 PM, Tres Seaver <tseaver@palladion.com> wrote:
On 05/22/2012 04:32 AM, Patrick Gerken wrote:
[2] FAILED ZTK 1.0dev / Python2.4.6 Linux 64bit https://mail.zope.org/pipermail/zope-tests/2012-May/063350.html
[3] FAILED ZTK 1.0dev / Python2.5.5 Linux 64bit https://mail.zope.org/pipermail/zope-tests/2012-May/063352.html
[4] FAILED ZTK 1.0dev / Python2.6.7 Linux 64bit https://mail.zope.org/pipermail/zope-tests/2012-May/063351.html
zope.app.twisted does not depend on twisted and does not have it as a dependency.
I made a stab at fixing it, but couldn't figure out which version of Twisted was compatible (none of the more recent versions work). I have no idea why tests passed before I copied the release branch.
[5] FAILED winbot / zope.app.appsetup_py_265_32 https://mail.zope.org/pipermail/zope-tests/2012-May/063340.html
Same error as last time. Does it make sense to test zope.app.appsetup? ztk does not depend on it it is not part of zopeapp.cfg of ztk 1.1 Grok from 1.3 onward does not use it. Bluebream... Well, bluebream uses zopeapp 1.0, but its last release is also nearly 1.5 years ago. There seems to be a successor(?) named ztfy, which uses ztk 1.1 so is also not dependent on zope.app.appsetup. But I am not sure for how long already.
Well, the question is, is there still a need for ztk 1.0 tests? If so, who are the consumers and for how long should we run the tests for 1.0? If anybody replies, please adjust the topic. I guess many people ignore the summarizer thread.
I think we can safely stop testing the deprecated pacakgaes individually.
Unfortunately somehow Plone still depends on it.
[6] FAILED winbot / zope.configuration_py_265_32 https://mail.zope.org/pipermail/zope-tests/2012-May/063339.html
Tres, could it be that this is related to the merge you did?
This is a Windows-only failure which I have tried several times to whack.
I THINK I fixed it. The commit message explains the failure if anybody wants to know. http://zope3.pov.lt/trac/changeset/126448