Bug#1004870: python-xarray: autopkgtest regression on s390x

[Alastair McKinstry]

reopen 1004870

thanks


I've just tested on zelenka with python-xarray & pandas 1.4.3;

FAILED 
xarray/tests/test_accessor_dt.py::test_calendar_cftime_2D[365_day] - 
pandas._libs.tslibs.np_datetime.OutOfBoundsDatetime: Out of bounds 
nanosecond timestamp: -809793270280017-0...
FAILED 
xarray/tests/test_accessor_dt.py::test_calendar_cftime_2D[360_day] - 
pandas._libs.tslibs.np_datetime.OutOfBoundsDatetime: Out of bounds 
nanosecond timestamp: -809793270280017-0...
FAILED xarray/tests/test_accessor_dt.py::test_calendar_cftime_2D[julian] 
- pandas._libs.tslibs.np_datetime.OutOfBoundsDatetime: Out of bounds 
nanosecond timestamp: -809793270280017-04...
FAILED 
xarray/tests/test_accessor_dt.py::test_calendar_cftime_2D[all_leap] - 
pandas._libs.tslibs.np_datetime.OutOfBoundsDatetime: Out of bounds 
nanosecond timestamp: -809793270280017-...
FAILED 
xarray/tests/test_accessor_dt.py::test_calendar_cftime_2D[366_day] - 
pandas._libs.tslibs.np_datetime.OutOfBoundsDatetime: Out of bounds 
nanosecond timestamp: -809793270280017-0...
FAILED 
xarray/tests/test_accessor_dt.py::test_calendar_cftime_2D[gregorian] - 
pandas._libs.tslibs.np_datetime.OutOfBoundsDatetime: Out of bounds 
nanosecond timestamp: -809793270280017...
FAILED 
xarray/tests/test_accessor_dt.py::test_calendar_cftime_2D[proleptic_gregorian] 
- pandas._libs.tslibs.np_datetime.OutOfBoundsDatetime: Out of bounds 
nanosecond timestamp: -80979...
FAILED xarray/tests/test_indexes.py::TestPandasIndex::test_join - 
AssertionError: assert dtype('>U4') == '<U4'
FAILED xarray/tests/test_indexes.py::TestPandasMultiIndex::test_concat - 
AssertionError: assert {'bar': dtype...type('int32')} == {'bar': 
'<U1'...numpy.int32'>}
FAILED xarray/tests/test_indexes.py::TestPandasMultiIndex::test_join - 
AssertionError: assert {'one': dtype...type('int32')} == {'one': '<U2', 
'two': 'i'}


On 27/08/2022 08:21, Rebecca N. Palmer wrote:
> pandas 1.4.3 in experimental has now been built for s390x - can 
> someone try the python-xarray autopkgtest with that (i.e. test whether 
> that actually fixes this bug)?
>
-- 
Alastair McKinstry,
GPG: 82383CE9165B347C787081A2CBE6BB4E5D9AD3A5
ph: +353 87 6847928 e: alastair at sceal.ie, im: @sceal.ie:mckinstry