FYI: I solved it. As usual the solution was dead simple once it struck me. I made a custom test command for setuptools, that simply create a temporary script modeled after the bin/test script that buildout would create, and call that. Problem solved! On Mon, Dec 7, 2009 at 16:24, Lennart Regebro <regebro@gmail.com> wrote:
On Mon, Dec 7, 2009 at 15:46, Lennart Regebro <regebro@gmail.com> wrote:
That doesn't seem to work. If I use discover withing setup.py, I get the same problems as using the standard setuptools testrunner. If I have a custom script that I run to run the tests, the tests seem to get stuck in some sort of infinite regression and never end.
It's really strange that running tests with setuptools fails. I think the best way forward is to try to figure out why, but I have no idea when or if I'm able to do that.
The problem here is that the tests for the test runner assumes that the tests are run via the test script created by the buildout, and must call that script to test itself. This needs to be changed, somehow. I'll think about it, and see if I can come up with a solution.
-- Lennart Regebro: Python, Zope, Plone, Grok http://regebro.wordpress.com/ +33 661 58 14 64