Mirror of git.openembedded.org/meta-openembedded https://git.enea.com/cgit/linux/meta-openembedded.git/atom?h=kirkstone-next 2020-09-15T16:28:35+00:00 2020-09-15T16:28:35+00:00 Leon Anavi leon.anavi@konsulko.com 2020-09-14T13:15:34+00:00 urn:sha1:34309528040995273be36b6eee634f8b68867a96 Upgrade to release 1.1.2: - Regression in DatetimeIndex.intersection() incorrectly raising AssertionError when intersecting against a list - Fix regression in updating a column inplace - Fix regression in DataFrame.append() mixing tz-aware and tz-naive datetime columns - Performance regression for RangeIndex.format() - Regression where MultiIndex.get_loc() would return a slice spanning the full index when passed an empty list - Bug in DataFrame.eval() with object dtype column binary operations - Bug in Series constructor raising a TypeError when constructing sparse datetime64 dtypes - Bug in DataFrame.apply() with result_type="reduce" returning with incorrect index - Bug in Series.astype() and DataFrame.astype() not respecting the errors argument when set to "ignore" for extension dtypes - Bug in DateTimeIndex.format() and PeriodIndex.format() with name=True setting the first item to "None" where it should be "" - Other bug fixes Signed-off-by: Leon Anavi <leon.anavi@konsulko.com> Acked-by: Trevor Gamblin <trevor.gamblin@windriver.com> Signed-off-by: Khem Raj <raj.khem@gmail.com> 2020-08-25T17:34:27+00:00 Leon Anavi leon.anavi@konsulko.com 2020-08-24T14:50:03+00:00 urn:sha1:28619bb56c9aeb3d71d15bd2924205a7936702a6 Upgrade to release 1.1.1: - Fixed regression in CategoricalIndex.format() where, when stringified scalars had different lengths, the shorter string would be right-filled with spaces, so it had the same length as the longest string - Fixed regression in Series.truncate() when trying to truncate a single-element series - Fixed regression where DataFrame.to_numpy() would raise a RuntimeError for mixed dtypes when converting to str - Fixed regression where read_csv() would raise a ValueError when pandas.options.mode.use_inf_as_na was set to True - Fixed regression where pandas.testing.assert_series_equal() would raise an error when non-numeric dtypes were passed with check_exact=True - Fixed regression in .groupby(..).rolling(..) where column selection was ignored - Fixed regression where DataFrame.interpolate() would raise a TypeError when the DataFrame was empty - Fixed regression in DataFrame.shift() with axis=1 and heterogeneous dtypes - Fixed regression in DataFrame.diff() with read-only data - Fixed regression in .groupby(..).rolling(..) where a segfault would occur with center=True and an odd number of values - Fixed regression in DataFrame.apply() where functions that altered the input in-place only operated on a single row - Fixed regression in DataFrame.reset_index() would raise a ValueError on empty DataFrame with a MultiIndex with a datetime64 dtype level - Fixed regression where pandas.merge_asof() would raise a UnboundLocalError when left_index, right_index and tolerance were set - Fixed regression in .groupby(..).rolling(..) where a custom BaseIndexer would be ignored - Fixed regression in DataFrame.replace() and Series.replace() where compiled regular expressions would be ignored during replacement - Bug in Styler whereby cell_ids argument had no effect due to other recent changes - Bug in pandas.testing.assert_series_equal() and pandas.testing.assert_frame_equal() where extension dtypes were not ignored when check_dtypes was set to False - Bug in to_timedelta() fails when arg is a Series with Int64 dtype containing null values - Bug in .groupby(..).rolling(..) where passing closed with column selection would raise a ValueError - Bug in DataFrame constructor failing to raise ValueError in some cases when data and index have mismatched lengths Signed-off-by: Leon Anavi <leon.anavi@konsulko.com> Acked-by: Trevor Gamblin <trevor.gamblin@windriver.com> Signed-off-by: Khem Raj <raj.khem@gmail.com>