DEPR: log of deprecations in 1.x (to be removed in 2.0)

[jorisvandenbossche]

Similar issue as #6581, but starting a fresh one for keeping a log of the deprecations we do in the 1.x releases.

1.0.0

• API/DEPR: Change default skipna behaviour + deprecate numeric_only in Categorical.min and max #27929 numeric_only to Categorical.min/max
• DEPR: is_extension_type #29457 is_extension_type
• DEPR: change DTI.to_series keep_tz default to True #29731 remove keep_tz kwarg from DatetimeIndex.to_series
• DEPR: Deprecate Index.set_value #28621 Index.set_value (presumably also IndexEngine.set_value)
• DEPR: deprecate truediv param in pd.eval #29812 Remove truediv kwarg from pd.eval
• DEPR: remove alias of Categorical.take_nd to Categorical.take #27745 Categorical.take_nd (closed by DEPR: Categorical, Index #49220)
• API: Use object dtype for empty Series #29405 Use object dtype for empty Series
• Added FutureWarning to Series.str.__iter__ #29909 str.__iter__
• REF: change parameter name fname -> path #30338 "fname" argument in to_parquet, to_stata, to_feather changed to "path"
• DEPR. Change RangeIndex._start/_stop/_step to FutureWarning #30482 RangeIndex._start, _stop, _step (closed by DEPR: to_native_types, set_value, iteritems, union_many, to_perioddelta #49244)
• DEPR: camelCase in offsets, get_offset #30340 onOffset, isAnchored, get_offset (closed by DEPR: enforce deprecations on DateOffset methods #48685)
• DEPR: Deprecate pandas.np module #30386 pd.np
• DEPR: Deprecate pandas.datetime #30489 pd.datetime
• DEPR: DataFrame GroupBy indexing with single items DeprecationWarning #30546 groupby single-list indexing
• DEPR: is_copy arg of take #30615 is_copy arg in .take
• DEPR: Deprecate numpy argument in read_json #30636 numpy arg in read_json
• CLN: Deprecate pandas.SparseArray for pandas.arrays.SparseArray #30656 deprecated pd.SparseArray (leave in designated pd.array.SparseArray)
• DEPR: CategoricalIndex.take_nd #30702 CategoricalIndex.take_nd
• DEPR/REGR: Fix pandas.util.testing deprecation #30745 pandas.util.testing module.

1.0.1

• BUG accept and deprecate negative integer for max_colwidth #31569 accepting a negative int in display.max_colwidth

1.1.0

• DEPR: indexing Series with single-entry list #31333 allowing single-item lists containing slice in Series.__getitem__ (closed by DEPR: Enforce diallowing indexing with single item slice #49343)
• API: Make describe changes backwards compatible #34798 DataFrame/Series.describe treats datelike as numeric.
• ENH: support datetime64, datetime64tz in nanops.mean, nanops.median #29941 DataFrame.mean and DataFrame.median with numeric_only=None will include datetime64 and datetime64tz columns in a future version (closed by DEPR: DataFrame.median/mean with numeric_only=None and dt64 columns #49250)
• BUG/DEPR: loc.__setitem__ incorrectly accepting positional slices #31840 loc.__setitem__ incorrectly accepting positional slices
• Deprecate Aliases as orient Argument in DataFrame.to_dict #32516 .to_dict(orient=), for abbreviations
• DEPR: Categorical.to_dense #32639 Categorical.to_dense (closed by DEPR: Categorical, Index #49220)
• Deprecate Aliases as orient Argument in DataFrame.to_dict #32516 .to_dict() aliases
• DEPR: Index.is_mixed #33291 Index.is_mixed (closed by DEPR: Categorical, Index #49220)
• API: Make most arguments for read_html and read_json keyword-ony #27573 read_html and read_json non-keyword args (closed by DEPR: non-keyword arguments #49302, DEPR: non-keyword args, errors arg #49415)
• DEPR: deprecate non keyword arguments in read_excel #34418 read_excel non-keyword args
• DEPR: is_categorical #33385 is_categorical (closed by DEPR: is_categorical #49253)
• DEPR: Deprecate week, weekofyear in Series.dt,DatetimeIndex #33595 week, weekofyear in Series.dt,DatetimeIndex (closed by DEPR: Remove week & weekofyear for datetimes #49380)
• REF: _AXIS_TO_AXIS_NUMBER to simplify axis access #33637 _AXIS_NUMBERS and _AXIS_NAMES properties on NDFrame and child classes (closed by DEPR: to_native_types, set_value, iteritems, union_many, to_perioddelta #49244)
• DEPR: get_value #33907 Index.get_value (closed by DEPR: remove Index.get_value #49208)
• ENH: add 'origin' and 'offset' arguments to 'resample' and 'pd.Grouper' #31809 `.resample(base=, loffset=)
• DEPR: DateOffset.__call__ #34171 DateOffset.__call__ (closed by DEPR: enforce deprecations on DateOffset methods #48685)
• 32380 deprecate squeeze in groupby #33218 groupby(squeeze=)
• DEPR: deprecate Index.__getitem__ with float key #34193 Index.__getitem__ with float key
• DEPR: tz kwarg in Period.to_timestamp #34522 Period.to_timestamp "tz" kwarg (closed by DEPR: tz arg in Period.to_timestamp #49280)
• DEPR: Deprecate tshift and integrate it to shift #34545 DataFrame/Series.tshift
• API: Allow non-tuples in pandas.merge #34810 pd.merge handling suffixes as a set.
• DEPR: to_perioddelta #34853 DatetimeIndex.to_perioddelta (closed by DEPR: to_native_types, set_value, iteritems, union_many, to_perioddelta #49244)
• Deprecate center on df.expanding #34887 center keyword in .expanding()
• ENH: Allow absolute precision in assert_almost_equal (#13357) #30562 check_less_precise in assert_frame_equal/assert_series_equal
• BUG: DataFrame.melt gives unexpected result when column "value" already exists #35003 deprecate a same column as the value parameter in DataFrame.melt
• DEPR: Deprecate n-dim indexing for Series #35141 multi-dimensional key to Series.__getitem__
• API: User-control of result keys in GroupBy.apply #34998 group_keys ignored for transform-like apply.
• Fixed apply_index #35165 Offset.apply_index (closed by DEPR: enforce deprecations on DateOffset methods #48685)

1.2.0

• DEPR: Deprecate inplace param in MultiIndex.set_codes and MultiIndex.set_levels #35626 inplace parameter in MultiIndex.set_codes and MultiIndex.set_levels
• DEPR: deprecate dtype param in Index.copy #35853 dtype parameters in Index.copy
• DEPR: Deprecate pandas/io/date_converters.py #35741 pandas/io/date_converters.py
• DEPR: Index.to_native_types #36418 Index.to_native_type (closed by DEPR: to_native_types, set_value, iteritems, union_many, to_perioddelta #49244)
• DEPR: string indexing along index for datetimes #36179 string indexing along index for datetimes (closed by DEPR: enforce indexing deprecations #49511)
• DEPR: Deprecate params levels & codes in MultiIndex.copy #36685 MultiIndex.copy(level, codes)
• DEPR: min_periods=None behavior for Rolling.count #36649 .rolling(min_periods=None ...).count()
• DEPR: automatic alignment on frame.__cmp__(series) #36795 automatic alignment on DataFrame cmp Series (closed by DEPR: enforce not-automatically aligning in DataFrame comparisons #49946)
• DEPR: disallow tznaive datetimes when indexing tzaware datetimeindex #36148 Indexing tz-aware with tz-naive (closed by DEPR: DatetimeIndex indexing with mismatched tzawareness #49492)
• API: Deprecate regex=True default in Series.str.replace #36695 regex=True in str.replace()
• DEPR: Index.ravel returning an ndarray #36900 Index.ravel
• API: Deprecate regex=True default in Series.str.replace #36695 default value of regex for Series.str.replace
• DEPR: Deprecate use of strings denoting units with 'M', 'Y' or 'y' in pd.to_timedelta (36666) #36838 'Y', 'M', 'y' units in to_timedelta (closed by BUG: Timestamp with unit=Y or unit=M #47266)
• DEPR: Index.__and__, __or__, __xor__ behaving as set ops #37374 Index.__and__, __or__, __xor__ behaving as setops (closed by DEPR: Index.__and__, __or__, __xor__ #49503)
• REF: Categorical.is_dtype_equal -> categories_match_up_to_permutation #37545 Categorical.is_dtype_equal and CategoricalIndex.is_dtype_equal (closed by DEPR: Categorical, Index #49220)
• DEPR: DataFrame/Series.slice_shift #37601 DataFrame.slice_shift
• Deprecate partial slicing of unordered DatetimeIndex when both keys are not present #37819 partial indexing of a non-monotonic DTI
• DEPR: Index.asi8 #37877 Index.asi8 (closed by DEPR: Categorical, Index #49220)
• Deprecate inplace in Categorical.remove_unused_categories #37918 Categorical.remove_unused_categories
• DEPR: how keyword in PeriodIndex.astype #37982 how keyword in Period.astype (closed by DEPR: PeriodIndex.astype how keyword #49234)
• DOC: Deprecate null_counts parameter of DataFrame.info #37999 null_counts in DataFrame.info
• DEPR: Deprecate using xlrd engine for read_excel #35029 default for read_excel with engine=None changing
• REF: de-duplicate Block.__init__ #38134 ndim arg in Block.__init__
• Deprecate groupby/pivot observed=False default #35967 observed=False keyword in groupby (this is pending)
• DEPR: xlwt for writing excel files #38317 xlwt in excel writing

1.2.1

• DEPR: raise deprecation warning in numpy ufuncs on DataFrames if not aligned + fallback to <1.2.0 behaviour #39239

1.2.5

• Bug in xs raising KeyError for MultiIndex columns with droplevel False and list indexe #41789 lists in .xs (closed by DEPR: enforce indexing deprecations #49511)

1.3.0

• DEPR: datetimelike.astype(int) #38544 .astype(int) on datetimelikes
• DEPR: allowing subclass-specific keywords in pd.Index.__new__ #38597 Index constuction sub-class keywords, additional tests needed (closed by DEPR: disallow subclass-specific keywords in Index.__new__ #49311)
• BUG: disallow scalar in Categorical constructor #38472 passing a scalar to Categorical constructor
• DEPR: is_all_dates #36697 Index.is_all_dates (tentative as reverted)
• BUG: Timestamp == date match stdlib #36131 Timestamp comparsion to datetime.date (closed by DEPR: Timestamp comparison with pydate #49394)
• DEPR: try_cast kwarg in mask, where #38836 try_cast in .where/.mask
• DEPR MultiIndex.is_lexsorted and MultiIndex.lexsort_depth #38701 MultiIndex.is_lexsorted
• DEPR: Rolling.win_type returning freq & is_datetimelike #38963 Rolling.win_type returning "freq" & Rolling.is_datetimelike
• DEPR: Rolling.win_type returning freq & is_datetimelike #38963 Rolling.is_datetimelike
• Deprecate DataFrame indexer for iloc setitem and getitem #39022 DataFrame.iloc[DataFrame.indexer] = ....
• DEPR: EWM.vol #39220 EWM.vol
• DEPR: astype dt64<->dt64tz #39258 .astype() with a tz-naive timestamp/index. (closed by DEPR: enforce DatetimeArray.astype deprecations #49235)
• Deprecate DataFrame indexer for iloc setitem and getitem #39022 DataFrame.iloc alignment with boolean DataFrame input
• DEPR: casting date to dt64 in maybe_promote #39767 .reindex with an incorrect fill value
• DEPR: Styler.set_na_rep and .set_precision in favour of .format(na_rep='x', precision=3) #40134 Styler.set_precision and Styler.set_na_rep
• DEPR: Partial failure in Series.transform and DataFrame.transform #40238 partial failure in Series.transform or DataFrame.transform
• DEPR: **kwargs in ExcelWriter #40430 kwargs in read_excel
• DEPR: DataFrame(MaskedRecords) #40363 MaskedRecords (closed by DEPR: enforce a bunch #49552)
• DEPR: CategoricalBlock; combine Block.replace methods #40527 CategoricalBlock
• CLN/DEPR: remove Block._holder, deprecated Block.is_categorical #40571 Block.is_categorical
• Deprecate level keyword for dataframe and series aggregations #40869 level= kwarg in DataFrame reduction methods
• Deprecate suffixes in merge producing duplicate columns #40991 merging with duplicated suffixes
• Deprecate joining over a different number of levels #40993 merging with non-equal number of levels
• Deprecate inplace in Categorical.remove_categories #37981 Categorial.remove_categories
• Deprecate inplace in Categorical.add_categories. #41118 Categorical.add_categories
• Deprecate inplace in Categorical.reorder_categories. #41133 Categorical.reorder_categories
• DEPR: Styler.where possible deprecation #40821 Styler.where
• DEPR: Deprecate convert_float #41176 ExcelFile are convert_float
• DEPR: kind kwarg in _maybe_cast_slice_bound #41378 Index._maybe_slice_bound loc keyword
• DEPR: setting Categorical._codes #41429 setting Categorical._codes
• Deprecated usecols with out of bounds indices in read_csv #41130 usecols in read_csv that are out of bounds
• DEPR: DatetimeIndex.union with mixed timezones #41458 DatatimeIndex.union with mixed-timezones
• DEPR: dropping nuisance columns in DataFrame reductions #41480 DataFrame reduction with nuiscance columns
• DEPR: dropping nuisance columns in DataFrameGroupby apply, agg, transform #41475 DataFrame groupby reductions with nuiscance columns
• Deprecate passing args as positional in DataFrame/Series.interpolate #41510 Positional args in .interpolate (closed by DEPR: non-keyword arguments #49302)
• ENH: Deprecate positional arguments for DataFrame.fillna and Series.fillna (GH41485) #41559 Positional args in .fillna
• ENH: Deprecate non-keyword arguments for Index.set_names. #41551 Positional args in .set_names (closed by DEPR: non-keyword arguments #49302)
• added deprecate_nonkeyword_arguments to function where #41523 Positional args in .where (closed by DEPR: disallow non-keyword arguments #49259)
• Deprecate non-keyword arguments in mask #41580 Positional args in .mask (closed by DEPR: disallow non-keyword arguments #49259)
• Deprecate passing args as positional in sort_index #41506 Positional args in .sort_index (closed by DEPR: non-keyword arguments #49302)
• Deprecate non-keyword arguments in drop #41486 Positional args in .drop
• Deprecate passing args as positional in dropna #41504 Positional args in .dropna (closed by DEPR: non-keyword arguments #49359)
• Deprecate passing args as positional in sort_values #41505 Positional args in .sort_values (closed by DEPR: disallow non-keyword arguments #49259)
• Deprecate passing args as positional DataFrame/Series.ffill #41508 Positional args in ffill/bfill (closed by DEPR: disallow non-keyword arguments #49259)
• Deprecate passing args as positional in DataFrame/Series.clip #41511 Positional args in .clip (closed by DEPR: disallow non-keyword arguments #49259)
• Deprecate nonkeyword args set axis #41491 Postiional args in .set_axis (closed by DEPR: disallow non-keyword arguments #49259)
• Deprecate passing default args as positional in DataFrame.set_index #41495 Positional args in .set_index (closed by DEPR: non-keyword arguments #49359)
• Deprecate passing default arguments as positional in reset_index #41496 Positional args in .reset_index (closed by DEPR: non-keyword arguments #49359)
• ENH: Deprecate non-keyword arguments for MultiIndex.set_levels #41589 Positional args in .set_levels
• Deprecated nonkeyword arguments for set_codes function #41650 Positional args in set_codes
• DEPR: DataFrame([categorical, ...]) special casing #41557 Dataframe construction with categorical in a list of arrays
• DEPR: Series/DataFrame with tzaware data and tznaive dtype #41555 Series/DataFrame with tzaware data and tznaive dtype (closed by DEPR: enforce Series/DataFrame awareness-mismatch deprecations #48739)
• DEPR: error_bad_lines and warn_bad_lines for read_csv #40413 read_csv, error_bad_lines, warn_bad_lines
• ENH: Deprecate arguments #41485 #41657 Positional args in read_csv (closed by DEPR: non-keyword arguments #49359)
• CLN: Deprecate non-keyword arguments in read_table #41485 #41717 Positional args in read_table (closed by DEPR: non-keyword arguments #49359)
• DEPR: Series(dt64naive, dtype=dt64tz) -> will match DatetimeIndex #41662 Series construction with i8 and DatetimeTZDtype as UTC (closed by DEPR: Series(dt64_naive, dtype=dt64tz) #49242)
• DEPR: silent overflow on Series construction #41734 Series construction with smaller dtypes than possible
• DEPR: ignoring dtype in DataFrame constructor failures #41733 Dataframe construction ignoring invalid dtypes (closed by DEPR: DataFrame dtype keyword match Series behavior #49313)
• DEPR: datetimelike inference with strings #41731 Series inference of M8, m8 w/o dtype (closed by DEPR: stop inferring dt64/td64 from strings in Series construtor #49319)
• DEPR: DataFrame(floaty, dtype=inty) match Series #41770 Dataframe construction with int dtypes and floats. (closed by DEPR: DataFrame dtype keyword match Series behavior #49313)
• ENH: New boundary inputs #40628 Series.between, inclusive options

1.4.0

• DEPR: Index.is_type_compatible #42113 Index.is_type_compatible (closed by DEPR: Categorical, Index #49220)
• DEPR: Series.__setitem__ with Float64Index falling back to positional #42215 Setitem with integer and Float64Index (closed by DEPR: enforce Series.__setitem__ behavior with Float64Index and non-present int key #49530)
• DEPR: Index.get_loc with method #42269 Index.get_loc with method=
• DEPR: ignoring missing labels when indexing on MultiIndex level #42351 Indexing with a MultiIndex and missing keys
• DEPR: Int64Index, UInt64Index & Float64Index #43028 Int64Index, Float64Index, UInt64Index
• DEPR: Passing in a string column label for DataFrame.ewm(times=...) #43265 .ewm(times=) as a string
• DEPR: squeeze argument in read_csv/read_table/read_excel #43427 squeeze arg in pd.read_csv, pd.read_excel
• DEPR: index argument to SparseArray #43523 index arg to SparseArray (closed by DEPR: enforce a bunch #49552)
• ENH: consistency of input args for boundaries (pd.date_range) #43504 closed arg in pd.date_range
• ENH: consistency of input args for boundaries in DataFrame.between_time() #40245 #43248 includes_* args in .between_time
• DEPR: rolling/expanding/ewm.validate #43665 Rolling/Expanding.validate
• DEPR: Silent dropping of nuisance columns in transform_dict_like #43743,DEPR: Silent dropping of nuisance columns in agg_list_like #43741 silent dropping of nuscance columns in .transform
• DEPR REF: hide(axis=..) replaces hide_index and hide_columns #43771 .hide_index / .hide_columns in Styler
• Deprecate skipna=None for mad #44580 skipna for .mad
• DEP: Deprecate non boolean sort for concat #44629 passing non boolean value to sort in concat
• DEPR: Categorical.replace #44929 Categorical.replace
• DEPR: numeric_only=None in DataFrame.rank #45059 numeric_only=None in DataFrame.rank

1.5.0

• DEPR: treating object-dtype blocks as bool_only #46188 treaing object-dtype blocks as bool_only (closed by DEPR: object-dtype bool_only #49371)
• BUG: Change numeric_only default to True #46096 numeric_only default in .quantile
• DEPR: Timedelta.freq, Timedelta.is_populated #46430 Timestamp.freq
• DEPR: Timedelta.delta #46476 Timedelta.delta
• DEPR: mad #46707 .mad
• DEPR: non-keyword arguments in any #44896 .any non-kwargs (closed by DEPR: non-keyword arguments #49302)
• DEPR: numeric_only default in DataFrame methods with None/True #46906, DEPR: numeric_only default in resampler ops #47177 numeric_only on DataFrame ops / resampler / groupby
• BUG: DatetimeIndex.resolution with nanosecond reso #46903 indexing a tz-naive DTI/series with a tz-aware convertible string
• TYP: Index.join #46518 Index.join for positional args (closed by DEPR: non-keyword arguments #49302)
• ENH: consistency of input args for boundaries - Interval #46522 remove closed keyword in Interval in favor of inclusive
• DEPR: DataFrameGroupBy numeric_only defaulting to True #46072
• DEPR: numeric_only default on DataFrame ops #46852
• DEPR: Timestamp(150.5, unit=Y) #47267 Timestamp(non_round_float, unit="Y/M") (closed by DEPR: enforce a bunch #49552)
• DEPR: display.column_space #47280 `display.column_space'
• DEPR: groupby.transform not aligning with input index #47244 groupby.transform aligning input
• DEPR: na_sentinel in factorize #47157 na_sentinel in pd.factorize
• Deprecate non-keyword arguments for rsplit #47446 non-keyword args in split (closed by DEPR: non-keyword args, errors arg #49415)
• DEPR: numeic_only in Series and SeriesGroupBy consistency #47561 numeic_only in Series.rank
• DEPR: Deprecate set_closed and add set_incluive #47636 IntervalIndex/IntervalArray.set_closed
• DEPR: deprecate unused errors in NDFrame.where/mask #47728 Series/DataFrame.mask/where(errors=...) (closed by DEPR: non-keyword args, errors arg #49415)
• DEPR: returning tuple when grouping by a list containing single element #47761 iter(df.groupby([length-1-key])) returning a scalar
• DEPR: inplace keyword for Categorical.set_ordered, setting .categories directly #47834 Categorical.set_ordered, setting .categories inplace
• TYP: pandas.core.generic from pandas-stubs #47912 to_excel(encoding=..., verbose...) (unused)
• DEPR: sort_columns in plot (#47563) #48073 deprecate sort_columns in plot
• DEPR: Deprecate mangle_dupe_cols #48137 mangle_dupe_cols
• DEPR: set_axis inplace keyword #48130
• DEPR: positional arguments of read/to_state #48128
• DEPR: inplace kwarg in set_index #48115Reverted
• DEPR: Deprecate positional arguments in pivot #48301
• DEPR: args and kwargs in rolling, expanding, and ewm ops #47851
• DEPR: datetimelike.astype(int_other_than_i8) return requested dtype #45574
• DEPR: df.iloc[:, foo] = bar attempt to set inplace #45333

1.5.1

• BUG: ExcelWriter.book - attribute not accessible for setting #48780

1.5.2

• REGR: Better warning in pivot_table when dropping nuisance columns #49615

[jbrockmendel]

For this one, let's use the checkboxes instead of cut/pasting to a separate issue when they are removed.

[jorisvandenbossche]

For this one, let's use the checkboxes instead of cut/pasting to a separate issue when they are removed.

Yes, +1, that's also what I wanted to propose

[mroeschke]

Noting some things to look out for to ensure there's a "thorough" removal when enforcing these deprecations (referencing the original PR may not encompass the extent of the feature)

• Doc changes
  • Remove/modify docstrings
  • Remove/modify API reference (if applicable)
  • Remove/modify references in User Guide
• Code changes
  • Remove warning.warn call & modify code as needed
  • Remove related warnings.filterwarnings in other parts of the codebase (if any)
  • Adjust type annotations (if related)
• Test changes
  • Remove tests checking the FutureWarning
  • Modify/add tests (if any) to confirm the future behavior
  • Remove any related filterwarnings (pytest.mark.filterwarnings, pyproject.toml)

[rhshadrach]

I searched through the pandas code, found a few more deprecations that weren't previously in the list, and added them.

PR authors: @jbrockmendel, @jorisvandenbossche, and myself

[mroeschke]

All prior deprecations have been enforced so closing

[rhshadrach]