Running pytest locally fails

[Alphare]

While working on updating my PR, I wanted to run the test suite locally after adding a test, and was greeted with the following message:

$ pytest -v
ImportError while loading conftest '[...]/asv/test/conftest.py'.
__init__.py:7: in <module>
    from asv.console import log
../asv/__init__.py:32: in <module>
    from . import plugin_manager
../asv/plugin_manager.py:60: in <module>
    plugin_manager.load_plugins(commands)
../asv/plugin_manager.py:30: in load_plugins
    __import__(name)
asv/commands/compare_variant.py:17: in <module>
    ???
E   ImportError: cannot import name results_default_iter

I have no idea when that issue started. Here is the output of pip freeze:

alabaster==0.7.12
asv==0.3.1
atomicwrites==1.3.0
attrs==19.1.0
Babel==2.6.0
certifi==2019.3.9
chardet==3.0.4
docutils==0.14
funcsigs==1.0.2
idna==2.8
imagesize==1.1.0
Jinja2==2.10
MarkupSafe==1.1.1
more-itertools==5.0.0
packaging==19.0
pathlib2==2.3.3
pluggy==0.9.0
py==1.8.0
Pygments==2.3.1
pyparsing==2.3.1
pytest==4.4.0
pytz==2018.9
requests==2.21.0
scandir==1.10.0
six==1.12.0
snowballstemmer==1.2.1
Sphinx==1.8.5
sphinx-bootstrap-theme==0.7.1
sphinxcontrib-websupport==1.1.0
typing==3.6.6
urllib3==1.24.1

Note: asv is installed locally as editable.

I am using "Python 2.7.16+".

[pv]

Mysterious import errors are sometimes due to import cycles. (Didn't actually look at your code so just guessing)

…

On September 3, 2019 8:12:38 PM UTC, "Raphaël Gomès" ***@***.***> wrote: While working on updating my PR, I wanted to run the test suite locally after adding a test, and was greeted with the following message: ``` $ pytest -v ImportError while loading conftest '[...]/asv/test/conftest.py'. __init__.py:7: in <module> from asv.console import log ../asv/__init__.py:32: in <module> from . import plugin_manager ../asv/plugin_manager.py:60: in <module> plugin_manager.load_plugins(commands) ../asv/plugin_manager.py:30: in load_plugins __import__(name) asv/commands/compare_variant.py:17: in <module> ??? E ImportError: cannot import name results_default_iter ``` I have no idea when that issue started. Here is the output of `pip freeze`: ``` alabaster==0.7.12 asv==0.3.1 atomicwrites==1.3.0 attrs==19.1.0 Babel==2.6.0 certifi==2019.3.9 chardet==3.0.4 docutils==0.14 funcsigs==1.0.2 idna==2.8 imagesize==1.1.0 Jinja2==2.10 MarkupSafe==1.1.1 more-itertools==5.0.0 packaging==19.0 pathlib2==2.3.3 pluggy==0.9.0 py==1.8.0 Pygments==2.3.1 pyparsing==2.3.1 pytest==4.4.0 pytz==2018.9 requests==2.21.0 scandir==1.10.0 six==1.12.0 snowballstemmer==1.2.1 Sphinx==1.8.5 sphinx-bootstrap-theme==0.7.1 sphinxcontrib-websupport==1.1.0 typing==3.6.6 urllib3==1.24.1 ``` Note: asv is installed locally as editable. I am using "Python 2.7.16+".

[Alphare]

I've checked out the main master branch and still have the same issue. Am I following the right procedure to run tests?

[pv]

You can check what tox and travisci do.

…

On September 4, 2019 8:34:09 AM UTC, "Raphaël Gomès" ***@***.***> wrote: I've checked out the main master branch and still have the same issue. Am I following the right procedure to run tests?

[kevingerman]

I find that I have to specify the test directory, a glob pattern or the specific module that I want to use when running pytest from the command line with asv.

e.g.

pytest test
pytest test/test_benchmark*

[pv]

FWIW, for me it works out of the box:

$ pytest 
============================= test session starts ==============================
platform linux2 -- Python 2.7.16, pytest-4.6.5, py-1.8.0, pluggy-0.12.0
rootdir: /home/pauli/prj/scipy/asv, inifile: pytest.ini
collected 210 items                                                            
... tests run ok ...
$ pytest-3
Test session starts (platform: linux, Python 3.7.4, pytest 5.1.2, pytest-sugar 0.9.2)
rootdir: /home/pauli/prj/scipy/asv, inifile: pytest.ini
plugins: cov-2.7.1, leaks-0.3.0, sugar-0.9.2, timeout-1.3.3, xdist-1.27.0, rerunfailures-6.0, forked-1.0.1
collecting ... 
... tests run ok ...

[pv]

I guess you could also try from clean checkout, git clean -fdx etc., in case there are some old py/pyc files lying around.

[Alphare]

It was indeed some kind of local issue. Note @pv I'm not certain recommending to run git clean -fdx without a warning is such a good idea considering its destructive nature.

Thanks a lot!