Thanks to visit codestin.com
Credit goes to github.com

Skip to content

Bug in pandas assert_*_equal wrapper test causes other tests to pass when they should fail #1447

Closed
@kandersolar

Description

@kandersolar

Describe the bug
Using the 36-min environment, I get different test results in the iotools tests depending on whether I include test_conftest.py in the test run.

To Reproduce

$ conda env create --file ci/requirements-py36-min.yml -n pvlib36min
$ conda activate pvlib36min
$ pytest pvlib/tests/test_conftest.py pvlib/tests/iotools  # no failures
$ pytest pvlib/tests/iotools/  # some failures

Expected behavior
I expect to be able to run test modules individually and get the same results I'd get by running the entire test suite.

Versions:

  • pvlib.__version__: master (0.9.1+2.g8d0f863)
  • pandas.__version__: 0.22.0
  • python: 3.6.13

Additional context
The failures are mostly about whether dataframe columns and indexes are equal, so it might have something to do with the assert_X_equal definitions in conftest.py? I also mentioned the psm3 failure here: #1374 (review)

Metadata

Metadata

Assignees

No one assigned

    Labels

    Type

    No type

    Projects

    No projects

    Milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions