Re: test error
from
"Michael A Rowley, MD"
on Oct 23, 2006 03:03 PM
ahh, that is very useful information. I will work on this and let
you know what I get. would make the test run much faster I imagine
also. Thanks much Rob for your time.
Michael
On Oct 23, 2006, at 2:13 PM, Rob Miller wrote:
> Michael A Rowley, MD wrote:
>> Hello again.
>> Just looking for a little insite. Have been trying to get my
>> tests to run since moving to remember and membrane. Has been
>> rather an adventure.
>> again, here is what I am running.
>> Python 2.4.3
>> Zope 2.9.5-final.
>> Plone 2.5.1
>> membrane svn checkout 21Oct2006
>> remember svn checkout 21Oct2006
>> here is the wierdness for today....
>> I am running the tests using the following command:
>> sudo /Applications/Plone-2.5.1/Instance/bin/zopectl test -vvvv -s
>> Products.TORCH2
>> this only runs one test file in my tests directory, which does not
>> inherit from the remember test class RememberTestBase, then I get
>> the error:
>
> i'm not 100% sure why you're seeing the problem you're seeing, but
> i do have a suggestion for how you might get around it.
>
>> class torchTestCase(RememberTestBase):
>> """ base class for Torch integration testing """
>> def afterSetUp(self):
>> RememberTestBase.afterSetUp(self)
>> # now add our Product I guess.
>> qi = self.portal.portal_quickinstaller
>> #qi.installProducts(['AlphaFlow',], stoponerror=1)
>> qi.installProducts(['TORCH2',], stoponerror=1)
>> self.mdata = getToolByName(self.portal, 'portal_memberdata')
>> self.addProvider()
>> self.addPatient()
>
> the remember test setup has been refactored to use layers. this is
> more efficient that using afterSetUp, b/c you can have your custom
> initialization stuff only run once for the entire test suite,
> instead of rerunning it for every test.
>
> i'd recommend you create your own layer class, subclassing
> Products.remember.tests.base.RememberProfileLayer. create a setUp
> class method (use the base class as a model), and do your setup in
> there, committing the transaction when it's done. then specify a
> "layer=YourCustomLayerClass" in your torchTestCase.
>
> try that, let us know if it improves things for you.
>
> -r
>
> --
> To unsubscribe send an email with subject unsubscribe to
> remember@....
> Please contact robm@... for questions.