Closed ebo closed 4 years ago
I am not sure what the problem is with nose
but all automated tests are passing (please refer to CI logs/configuration for 5.x-dev
branch for details). I use pytest
(docs). For the branch information to work with, please refer to the wiki. Current stable version is 5.x series.
Just in case, I've dropped support for Python v3.4 and pypy3.5 for 5.x-dev
branch and Python 2.7, Python 3.4 and pypy3.5 for 6.x-dev
branch.
Edit: I've developed the numba port of the evaluators almost 2 years ago but never released it. I am planning to add numba.jit
and numba.cuda.jit
evaluators in 6.x
branch when I find time to write tests for them :)
I have a little time at the moment to test the numba.jit support if you can release them now -- it would be less work for me writing tests than starting a branch from scratch...
As a note, my comment about the tests broken turned out to be the pypy version issue mentioned in the other bug report. On my install using pytests also breaks with the error:
=================================================================================== ERRORS ====================================================================================
____ ERROR collecting tests/testvisualization.py ____
ImportError while importing test module '/home/ebo/save_all/gtd/projects/cnc/RPi/ref/NURBS-Python/tests/test_visualization.py'.
Hint: make sure your test modules/packages have valid Python names.
Traceback:
tests/test_visualization.py:15: in
This is still a system issue, but broken all the same. I will see if I can either get pypy to work, or look into getting pytest to finish testing without a critical fail.
While evaluating NURBS-Python as a starting point for a numba project, I ran nosetest to start from the examples. I consistently found from 100 to over 200 broken tests in the test suite (depending on the branch).
What is the most stable branch that I should start from?
Are the automated test suite maintained?
Steps to reproduce:
results in 100's of broken tests.