I'm not sure there really is much dh-python can do here. Note that there
really are no tests in the upstream package. Cloning upstream's hg you find:
% python3.4 -m unittest discover -v
----------------------------------------------------------------------
Ran 0 tests in 0.000s
OK
and
% python3.5 -m unittest discover -v
patchqueue (unittest.loader._FailedTest) ... ERROR
======================================================================
ERROR: patchqueue (unittest.loader._FailedTest)
----------------------------------------------------------------------
ImportError: Failed to import test module: patchqueue
Traceback (most recent call last):
File "/usr/lib/python3.5/unittest/loader.py", line 462, in _find_test_path
package = self._get_module_from_name(name)
File "/usr/lib/python3.5/unittest/loader.py", line 369, in
_get_module_from_name
__import__(name)
File
"/home/barry/projects/debian/sphinxpatchq/upstream/patchqueue/__init__.py",
line 11, in <module>
import mercurial.patch
ImportError: No module named 'mercurial'
----------------------------------------------------------------------
Ran 1 test in 0.000s
FAILED (errors=1)
So without dh-python getting involved, the bug is reproducible. I'm not sure
it's a bug in Python 3.5 either really, just a new feature which turns out in
this case to be useful for finding the error early. I guess.
