[issue34688] Segfault in pandas that works fine on 3.7

Karthikeyan Singaravelan added the comment: This has an issue in GitHub and pandas is also tested with 3.8 from https://github.com/pandas-dev/pandas/pull/28730. I am closing this as third party. -- resolution: -> third party stage: -> resolved status: open -> closed

[issue34688] Segfault in pandas that works fine on 3.7

Karthikeyan Singaravelan added the comment: Pandas issue that contains other segfaults : https://github.com/pandas-dev/pandas/issues/22714 Thanks -- ___ Python tracker ___

[issue34688] Segfault in pandas that works fine on 3.7

Vladimir Matveev added the comment: somewhat shortened repro that fails with the same error on master: ``` import pandas import numpy now = pandas.Timestamp.now() arr = numpy.array([ ['a', now] for i in range(0, 3)]) arr.sum(0) ``` -- nosy: +v2m

[issue34688] Segfault in pandas that works fine on 3.7

Karthikeyan Singaravelan added the comment: Copying my GitHub comment here : If I have only D or no D then there is no segfault. It occurs when I have either A, D or B, D, or C, D. **Leaving out column D** ```python from pandas import DataFrame import pandas as pd df = DataFrame({'A':

[issue34688] Segfault in pandas that works fine on 3.7

New submission from Karthikeyan Singaravelan : I have tried running pandas test suite under Python 3.8 and the below code segfault as part of the test suite. Pandas uses Python 3.7 in Travis and it runs fine. There is also NumPy related code in the trace but it also seems to run fine under