Backport PR #45212: CLN: suppress warnings (#45269)

Co-authored-by: jbrockmendel <[email protected]>

Author: meeseeksmachine

doc/source/whatsnew/v1.4.0.rst

@@ -816,7 +816,7 @@ Indexing
 
 Missing
 ^^^^^^^
-- Bug in :meth:`DataFrame.fillna` with limit and no method ignores axis='columns' or ``axis = 1`` (:issue:`40989`)
+- Bug in :meth:`DataFrame.fillna` with limit and no method ignores axis='columns' or ``axis = 1`` (:issue:`40989`, :issue:`17399`)
 - Bug in :meth:`DataFrame.fillna` not replacing missing values when using a dict-like ``value`` and duplicate column names (:issue:`43476`)
 - Bug in constructing a :class:`DataFrame` with a dictionary ``np.datetime64`` as a value and ``dtype='timedelta64[ns]'``, or vice-versa, incorrectly casting instead of raising (:issue:`44428`)
 - Bug in :meth:`Series.interpolate` and :meth:`DataFrame.interpolate` with ``inplace=True`` not writing to the underlying array(s) in-place (:issue:`44749`)

pandas/_libs/lib.pyx

@@ -714,7 +714,7 @@ cpdef ndarray[object] ensure_string_array(
             return out
 
         arr = arr.to_numpy()
-    elif not isinstance(arr, np.ndarray):
+    elif not util.is_array(arr):
         arr = np.array(arr, dtype="object")
 
     result = np.asarray(arr, dtype="object")
@@ -729,7 +729,7 @@ cpdef ndarray[object] ensure_string_array(
             continue
 
         if not checknull(val):
-            if not isinstance(val, np.floating):
+            if not util.is_float_object(val):
                 # f"{val}" is faster than str(val)
                 result[i] = f"{val}"
             else:

pandas/core/arrays/masked.py

@@ -767,7 +767,7 @@ def _quantile(
         We assume that all impacted cases are 1D-only.
         """
         mask = np.atleast_2d(np.asarray(self.isna()))
-        npvalues = np.atleast_2d(np.asarray(self))
+        npvalues: np.ndarray = np.atleast_2d(np.asarray(self))
 
         res = quantile_with_mask(
             npvalues,

pandas/core/dtypes/cast.py

@@ -459,7 +459,7 @@ def ensure_dtype_can_hold_na(dtype: DtypeObj) -> DtypeObj:
         # TODO: ExtensionDtype.can_hold_na?
         return dtype
     elif dtype.kind == "b":
-        return np.dtype(object)
+        return _dtype_obj
     elif dtype.kind in ["i", "u"]:
         return np.dtype(np.float64)
     return dtype
@@ -522,7 +522,7 @@ def _maybe_promote(dtype: np.dtype, fill_value=np.nan):
             # with object dtype there is nothing to promote, and the user can
             #  pass pretty much any weird fill_value they like
             raise ValueError("fill_value must be a scalar")
-        dtype = np.dtype(object)
+        dtype = _dtype_obj
         return dtype, fill_value
 
     kinds = ["i", "u", "f", "c", "m", "M"]
@@ -532,7 +532,7 @@ def _maybe_promote(dtype: np.dtype, fill_value=np.nan):
         return dtype, fv
 
     elif isna(fill_value):
-        dtype = np.dtype(object)
+        dtype = _dtype_obj
         if fill_value is None:
             # but we retain e.g. pd.NA
             fill_value = np.nan
@@ -551,7 +551,7 @@ def _maybe_promote(dtype: np.dtype, fill_value=np.nan):
         #      fv = dta._validate_setitem_value(fill_value)
         #      return dta.dtype, fv
         #  except (ValueError, TypeError):
-        #      return np.dtype(object), fill_value
+        #      return _dtype_obj, fill_value
         if isinstance(fill_value, date) and not isinstance(fill_value, datetime):
             # deprecate casting of date object to match infer_dtype_from_scalar
             #  and DatetimeArray._validate_setitem_value
@@ -699,7 +699,7 @@ def infer_dtype_from_scalar(val, pandas_dtype: bool = False) -> tuple[DtypeObj,
         If False, scalar belongs to pandas extension types is inferred as
         object
     """
-    dtype: DtypeObj = np.dtype(object)
+    dtype: DtypeObj = _dtype_obj
 
     # a 1-element ndarray
     if isinstance(val, np.ndarray):
@@ -718,13 +718,13 @@ def infer_dtype_from_scalar(val, pandas_dtype: bool = False) -> tuple[DtypeObj,
         # instead of np.empty (but then you still don't want things
         # coming out as np.str_!
 
-        dtype = np.dtype(object)
+        dtype = _dtype_obj
 
     elif isinstance(val, (np.datetime64, datetime)):
         try:
             val = Timestamp(val)
         except OutOfBoundsDatetime:
-            return np.dtype(object), val
+            return _dtype_obj, val
 
         # error: Non-overlapping identity check (left operand type: "Timestamp",
         # right operand type: "NaTType")
@@ -736,13 +736,13 @@ def infer_dtype_from_scalar(val, pandas_dtype: bool = False) -> tuple[DtypeObj,
                 dtype = DatetimeTZDtype(unit="ns", tz=val.tz)
             else:
                 # return datetimetz as object
-                return np.dtype(object), val
+                return _dtype_obj, val
 
     elif isinstance(val, (np.timedelta64, timedelta)):
         try:
             val = Timedelta(val)
         except (OutOfBoundsTimedelta, OverflowError):
-            dtype = np.dtype(object)
+            dtype = _dtype_obj
         else:
             dtype = np.dtype("m8[ns]")
             val = np.timedelta64(val.value, "ns")
@@ -1911,7 +1911,7 @@ def construct_1d_arraylike_from_scalar(
         try:
             dtype, value = infer_dtype_from_scalar(value, pandas_dtype=True)
         except OutOfBoundsDatetime:
-            dtype = np.dtype(object)
+            dtype = _dtype_obj
 
     if isinstance(dtype, ExtensionDtype):
         cls = dtype.construct_array_type()

pandas/core/indexing.py

@@ -4,6 +4,7 @@
 from typing import (
     TYPE_CHECKING,
     Hashable,
+    Sequence,
 )
 import warnings
 
@@ -2016,6 +2017,7 @@ def _ensure_iterable_column_indexer(self, column_indexer):
         """
         Ensure that our column indexer is something that can be iterated over.
         """
+        ilocs: Sequence[int]
         if is_integer(column_indexer):
             ilocs = [column_indexer]
         elif isinstance(column_indexer, slice):

pandas/core/internals/array_manager.py

@@ -1203,7 +1203,7 @@ def make_empty(self, axes=None) -> SingleArrayManager:
         """Return an empty ArrayManager with index/array of length 0"""
         if axes is None:
             axes = [Index([], dtype=object)]
-        array = np.array([], dtype=self.dtype)
+        array: np.ndarray = np.array([], dtype=self.dtype)
         return type(self)([array], axes)
 
     @classmethod

pandas/core/internals/managers.py

@@ -1090,14 +1090,12 @@ def iset(
             #  containing (self._blknos[loc], BlockPlacement(slice(0, 1, 1)))
 
             # Check if we can use _iset_single fastpath
+            loc = cast(int, loc)
             blkno = self.blknos[loc]
             blk = self.blocks[blkno]
             if len(blk._mgr_locs) == 1:  # TODO: fastest way to check this?
                 return self._iset_single(
-                    # error: Argument 1 to "_iset_single" of "BlockManager" has
-                    # incompatible type "Union[int, slice, ndarray[Any, Any]]";
-                    # expected "int"
-                    loc,  # type:ignore[arg-type]
+                    loc,
                     value,
                     inplace=inplace,
                     blkno=blkno,

pandas/core/reshape/reshape.py

@@ -1021,7 +1021,7 @@ def get_empty_frame(data) -> DataFrame:
         if isinstance(data, Series):
             index = data.index
         else:
-            index = np.arange(len(data))
+            index = Index(range(len(data)))
         return DataFrame(index=index)
 
     # if all NaN
@@ -1031,7 +1031,7 @@ def get_empty_frame(data) -> DataFrame:
     codes = codes.copy()
     if dummy_na:
         codes[codes == -1] = len(levels)
-        levels = np.append(levels, np.nan)
+        levels = levels.insert(len(levels), np.nan)
 
     # if dummy_na, we just fake a nan level. drop_first will drop it again
     if drop_first and len(levels) == 1:

pandas/tests/frame/test_ufunc.py

@@ -269,6 +269,11 @@ def test_alignment_deprecation_many_inputs(request):
         )
         request.node.add_marker(mark)
 
+        mark = pytest.mark.filterwarnings(
+            "ignore:`np.MachAr` is deprecated.*:DeprecationWarning"
+        )
+        request.node.add_marker(mark)
+
     @vectorize([float64(float64, float64, float64)])
     def my_ufunc(x, y, z):
         return x + y + z

pandas/tests/io/test_parquet.py

@@ -43,6 +43,9 @@
     with catch_warnings():
         # `np.bool` is a deprecated alias...
         filterwarnings("ignore", "`np.bool`", category=DeprecationWarning)
+        # accessing pd.Int64Index in pd namespace
+        filterwarnings("ignore", ".*Int64Index.*", category=FutureWarning)
+
         import fastparquet
 
     _HAVE_FASTPARQUET = True

pandas/tests/plotting/test_datetimelike.py

@@ -21,6 +21,7 @@
     Index,
     NaT,
     Series,
+    concat,
     isna,
     to_datetime,
 )
@@ -244,7 +245,7 @@ def test_fake_inferred_business(self):
         _, ax = self.plt.subplots()
         rng = date_range("2001-1-1", "2001-1-10")
         ts = Series(range(len(rng)), index=rng)
-        ts = ts[:3].append(ts[5:])
+        ts = concat([ts[:3], ts[5:]])
         ts.plot(ax=ax)
         assert not hasattr(ax, "freq")
 

pandas/tests/test_downstream.py

@@ -103,6 +103,10 @@ def test_oo_optimized_datetime_index_unpickle():
     # patsy needs to update their imports
     "ignore:Using or importing the ABCs from 'collections:DeprecationWarning"
 )
+@pytest.mark.filterwarnings(
+    # numpy 1.22
+    "ignore:`np.MachAr` is deprecated.*:DeprecationWarning"
+)
 def test_statsmodels():
 
     statsmodels = import_module("statsmodels")  # noqa:F841

pandas/util/_test_decorators.py

@@ -72,6 +72,14 @@ def safe_import(mod_name: str, min_version: str | None = None):
             message=".*decorator is deprecated since Python 3.8.*",
         )
 
+        # fastparquet import accesses pd.Int64Index
+        warnings.filterwarnings(
+            "ignore",
+            category=FutureWarning,
+            module="fastparquet",
+            message=".*Int64Index.*",
+        )
+
         try:
             mod = __import__(mod_name)
         except ImportError: