On Jan 12, 2007, at 12:25 PM, Christoph Zwerschke wrote:
> > Alberto Valverde wrote: >> AFAIK and IIRC TG releases have always passed all tests.... ;) I >> think I'll tweak (though I wouldn't mind someone doing it for >> me... ;) #1234 to use lower() as I believe those tests only fail with >> Kid 0.9.4.... Hmmm, I've just upgraded to 0.9.4 and I'm seeing plenty >> of "TypeError: generate_content() takes exactly 1 argument (2 >> given)". I think the safest solution for this would be to make >> TurboKid require Kid < 0.9.4 and don't patch TG's tests with >> something that will make them fail under 0.9.4. > > Alberto, the issues with Kid 0.9.4 here are the following: > > 1) Kid 0.9.4 output for the 'html' method used in the tests has now > lowercase tags instead of uppercase. If you want uppercase, use the > 'HTML' output method. I was aware of that.... > > So you have to either update your tests to check against lowercase > tags > or use 'HTML' as Kid output method in the test configuration. Maybe > the > tests should better use 'xhtml' instead of 'html' as output method > anyway. > > 2) Compiled Kid templates (.pyc files in template directories) of Kid > 0.9.3 are not compatible with Kid 0.9.4, they need to be recompiled. Aha! That explain the weird TypeError. > > The best solution here would be to automatically remove all .pyc files > from all Kid template directories before running the tests. This also > ensures the test is really running and uses the templates built > with the > current version of TurboGears. In Kid version 0.9.5, recompilation of > templates with lower version number happens automatically, but > removing > compiled templates before running the tests makes sense anyway. Hmm, I'm nt sure if it's a good idea to automatically remove pycs before running the tests. No testing framework that I'm aware of does it and I think that kind of clean up is better left to the person running the tests (a quick "find . -name '*pyc' | xargs rm"). However, I might be wrong and I really have nothing agais automatic cleanup so it's up to you... Reagrding the case issue I'd rather user "lower" on the output (or use xhtml) than manually changing the case like the current patch does because that will make sure tests also pass for kid < 0.9.4) > > I can try submitting a patch for these issues over the weekend if you > want me to. That would be great! Thanks :) Alberto --~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google Groups "TurboGears" group. To post to this group, send email to [email protected] To unsubscribe from this group, send email to [EMAIL PROTECTED] For more options, visit this group at http://groups.google.com/group/turbogears?hl=en -~----------~----~----~----~------~----~------~--~---
