Unit-testing for a collection of jupyter notebooks. testipynb
relies on nbconvert to run the notebooks and catches errors so that they are output (with syntax highlighting!) when unit-tests are run.
- If you want to share your notebooks and be confident that they _should_ work on someone else's machine
- If you are using notebooks to generate figures in a publication and want to ensure they are reproducible (powerful when connected with cron jobs on travis-ci)
pip install testipynb
import testipynb
NBDIR = '../notebooks'
Test = testipynb.TestNotebooks(directory=NBDIR)
Test.assertTrue(Test.run_tests())
or in a unit-test file:
import testipynb
import unittest
NBDIR = '../notebooks'
Test = testipynb.TestNotebooks(directory=NBDIR, timeout=2100)
TestNotebooks = Test.get_tests()
if __name__ == "__main__":
unittest.main()
testipynb
is used in:
If you use testipynb
in one of your repositories and would like it listed, please edit this file