Le samedi 26 mai 2012 à 13:13 -0600, Aaron Meurer a écrit : > Hi. > > I recently downloaded the latest alpha of Python 3.3, and there are a > lot of test failures. See > http://code.google.com/p/sympy/issues/detail?id=3272 for details. > > Any help in debugging these would be appreciated. Right now, the > doctests completely crash because of some metaclass issue, and the > regular tests hang on sympy/core/tests/test_expand.py (see the issue > for more details).
My uninformed guess is that the metaclass issue is a bug of 3.3, given that it works in 3.2 and that the traceback message is wrong. > > Python 3.3 enables hash randomization, which randomizes the hash of > objects, so likely most if not all of the failures are due to that. > If you get different results on different runs, this is likely the > reason. Note that hash randomization can be enabled in 2.7.3 and 3.2.3 with the -R option. > > We also need to figure out if it's possible to get the seed for the > hash randomization and if it can be inputted manually. > > To test Python 3.3, you'll need to download it from > http://python.org/download/releases/3.3.0/. There are binaries for > Mac OS X and Windows, but for Linux you'll have to compile it (don't > worry, it's not hard, and doesn't take long). Then, run (from the > sympy directory) > > ./bin/use2to3 > /path/to/the/python3.3/executable/python.exe py3k-sympy/bin/test > /path/to/the/python3.3/executable/python.exe py3k-sympy/bin/doctest > > You can't run setup.py test yet for the same reason you can't in > Python 3.2, because of bug in Python that causes a stack overflow > (when pull request #1304 is merged, this will no longer be an issue). > > Aaron Meurer > -- You received this message because you are subscribed to the Google Groups "sympy" 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/sympy?hl=en.
