quite a few tests that predict times or durations failed in spurious
fashion in the last few days.
The branch at hand introduces a BuildQueue._now() method that is used by the
BuildQueue code under test while enabling the unit tests to "monkey-patch"
that method and have it return a constant time in order to facilitate the
testing process.
Hello,
quite a few tests that predict times or durations failed in spurious
fashion in the last few days.
The branch at hand introduces a BuildQueue._now() method that is used by the
BuildQueue code under test while enabling the unit tests to "monkey-patch"
that method and have it return a constant time in order to facilitate the
testing process.
This is based on an idea of Julian's.
Tests to run:
bin/test -vv -t test_buildqueue
No "make lint" errors or warnings.