I have an emergent bug that I've got to track down tomorrow. I know a previous hg revision which was good so I'm thinking about using hg bisect.
However, I'm on Windows and don't want to get into DOS scripting.
Ideally, I'd be able to write a Python unit test and have hg bisect use that. This is my first attempt.
bisector.py
#!/usr/bin/env python
import sys
import unittest
class TestCase(unittest.TestCase):
def test(self):
#raise Exception('Exception for testing.')
#self.fail("Failure for testing.")
pass
def main():
suite = unittest.defaultTestLoader.loadTestsFromTestCase(TestCase)
result = unittest.TestResult()
suite.run(result)
if result.errors:
# Skip the revision
return 125
if result.wasSuccessful():
return 0
else:
return 1
if '__main__' == __name__:
sys.exit(main())
Perhaps I could then run:
hg bisect --reset
hg bisect --bad
hg bisect --good -r 1
hg bisect --command=bisector.py
Is there a better way of doing it? Thanks for any advice.
Thanks to all, especially to Will McCutchen. The solution that worked best is below.
bisector.py
#!/usr/bin/env python
import unittest
class TestCase(unittest.TestCase):
def test(self):
# Raise an assertion error to mark the revision as bad
pass
if '__main__' == __name__:
unittest.main()
The hard part was getting the hg bisect commands right:
hg update tip
hg bisect --reset
hg bisect --bad
hg bisect --good 0
hg bisect --command ./bisector.py
or on Windows, the last command is:
hg bisect --command bisector.py
I think you can remove your main()
function and use the following block to run the tests:
if __name__ == '__main__':
unittest.main()
The call to unittest.main()
will run the tests it finds in this file and exit with an appropriate status code depending on whether all the tests pass or fail.
If you love us? You can donate to us via Paypal or buy me a coffee so we can maintain and grow! Thank you!
Donate Us With