On Fri, 2011-04-29 at 22:18 +0200, Baptiste Lepilleur wrote: > 2011/4/29 Ronny Pfannschmidt <ronny.pfannschm...@gmx.de> > > [...] > > E Failed: funcarg 'property' not used in > this > > function. > [...] > it seems the function is test_property_validation > the current check for the metafunc funcarg dict does not > consider > arguments with default values funcargs > thus it complains since you pass in something that has a > default at the > function level > > > It was indeed the cause. Changing: > def test_property_validation( validation_scenario_iter, property=None, > value=None, good_value=None ): > > > to: > > > def test_property_validation( validation_scenario_iter, property, > value, good_value ): > > > fixed the issue. This is weird as it used to work in 1.3.4... > > > > > Source of the test module can be found > > there: http://pastebin.com/1D2VArGt > > > > > i would like to suggest splitting the scenario types into > multiple test > functions, (good/bad values/types) > > > > I'm not sure how I would go about that: it would means going through > the VALIDATION_SCENARIOS test data for each test function, and > duplicating some setup code of the test function...
unless i'm misstaken, only the creation of the class is shared which means that could actually be something like a funcarg or a param to a funcarg -- Ronny > > > also giving names/id's to the scenarios > I did not know it was possible to give name. I added this and it > greatly improve the output! > > > Thanks a lot for helping figuring this out, > Baptiste. > > > _______________________________________________ > py-dev mailing list > py-dev@codespeak.net > http://codespeak.net/mailman/listinfo/py-dev
signature.asc
Description: This is a digitally signed message part
_______________________________________________ py-dev mailing list py-dev@codespeak.net http://codespeak.net/mailman/listinfo/py-dev
