skip test_float_precision_options, test_rolling_var_numerical_issues and test_rolling_skew_kurt_large_value_range #113
Conversation
…heels to be published for 1.2rc)
|
@jreback i'll close this for a bit, since there are other 32bit tests failing, test_rolling_var_numerical_issues and test_rolling_skew_kurt_large_value_range that are failing on all linux 32bit, linux py_3.7_32, linux py_3.8_32 and linux py_3.9_32 (and also aarch) Unlike test_float_precision_options which was failing on windows py_3.8_32 and windows py_3.9_32 (not testing for 3.7[1]), i'm surprised these other 32bit failures aren't appearing on our ci. (It would be easier to debug/fix if they were) so I think we should first investigate why these failures are not appearing on our ci before attempting any fixes. [1] It appears that the py37 can be re-enabled will open a PR shortly It also appears that we no longer need to skip test_file_descriptor_leak (and can unpin pytest-dist #95) - will open a PR shortly for this given the planned tidy up, I think the best course of action is to skip the 3 32bit tests here for now so that nightly wheels build. Once the other PRs are complete and merged will reopen this. |
|
|
ok sure |
…_kurt_large_value_range
simonjayhawkins
changed the title
|
@jreback I will be merging this before the end of the day so that the nightly builds of the 32bit wheels succeeds. |
xref pandas-dev/pandas#36429 (comment)