pull_requests: 525378289

This data as json

id node_id number state locked title user body created_at updated_at closed_at merged_at merge_commit_sha assignee milestone draft head base author_association auto_merge repo url merged_by
525378289 MDExOlB1bGxSZXF1ZXN0NTI1Mzc4Mjg5 4600 closed 0 plot tests: ensure all figures are closed 24938400 Test `testxyincrease_false_changes_axes` is flaky in nature. TestCase passes consistently in the actual order with command: ``` pytest tests/test_plot.py ``` TestCase fails when one randomises the order in *test_plot.py* module by using the plugin [pytest-random-order](https://pypi.org/project/pytest-random-order/) with command: ``` pytest tests/test_plot.py --random-order-seed=802425 -v ``` The victim `test_xyincrease_false_changes_axes` fails in a failing test order because there is at least one test that runs before the victim and “pollute” the state (e.g., global state: here, it is `matplotlib's plt`) on which the victim depends. [(Paper)](https://taoxie.cs.illinois.edu/publications/esecfse19-ifixflakies.pdf) After investigation, we found the polluter test to be *test_ylim_kwarg* Potential Fix: Adding a `plt.clf` or `plt.cla` at the end of the polluter test. Let us know your preference. Please let us know if you require the pass and fail logs. Also, if one runs test_plot.py in any random order, Teardown phase throws an error which is raised by test_all_figures_closed(). Should I raise a separate PR for the same? Contributors: [Yash Saboo](https://github.com/yashsaboo) [Nirupam K N](https://github.com/Nirupamkn) 2020-11-22T23:45:04Z 2020-11-27T12:03:15Z 2020-11-27T10:18:38Z 2020-11-27T10:18:38Z a04aa25fea8a283493b87190c75aaec6118b7623     0 37f570d3a876711ea0d4eacf8a3e0cb2a2151856 5883a460b8fa9a72a2f12ee6495665bb7cc304f6 CONTRIBUTOR   13221727 https://github.com/pydata/xarray/pull/4600  

Links from other tables