DEPR: datetimelike.astype(int) (#38544)

Author: jbrockmendel

doc/source/whatsnew/v1.3.0.rst

@@ -146,6 +146,7 @@ Deprecations
 ~~~~~~~~~~~~
 - Deprecating allowing scalars passed to the :class:`Categorical` constructor (:issue:`38433`)
 - Deprecated allowing subclass-specific keyword arguments in the :class:`Index` constructor, use the specific subclass directly instead (:issue:`14093`,:issue:`21311`,:issue:`22315`,:issue:`26974`)
+- Deprecated ``astype`` of datetimelike (``timedelta64[ns]``, ``datetime64[ns]``, ``Datetime64TZDtype``, ``PeriodDtype``) to integer dtypes, use ``values.view(...)`` instead (:issue:`38544`)
 -
 -
 

pandas/core/arrays/datetimelike.py

@@ -352,6 +352,14 @@ def astype(self, dtype, copy=True):
         elif is_integer_dtype(dtype):
             # we deliberately ignore int32 vs. int64 here.
             # See https://github.com/pandas-dev/pandas/issues/24381 for more.
+            warnings.warn(
+                f"casting {self.dtype} values to int64 with .astype(...) is "
+                "deprecated and will raise in a future version. "
+                "Use .view(...) instead.",
+                FutureWarning,
+                stacklevel=3,
+            )
+
             values = self.asi8
 
             if is_unsigned_integer_dtype(dtype):

pandas/core/dtypes/cast.py

@@ -1,7 +1,6 @@
 """
 Routines for casting.
 """
-
 from contextlib import suppress
 from datetime import datetime, timedelta
 from typing import (
@@ -17,6 +16,7 @@
     Type,
     Union,
 )
+import warnings
 
 import numpy as np
 
@@ -997,6 +997,14 @@ def astype_nansafe(
 
     elif is_datetime64_dtype(arr):
         if dtype == np.int64:
+            warnings.warn(
+                f"casting {arr.dtype} values to int64 with .astype(...) "
+                "is deprecated and will raise in a future version. "
+                "Use .view(...) instead.",
+                FutureWarning,
+                # stacklevel chosen to be correct when reached via Series.astype
+                stacklevel=7,
+            )
             if isna(arr).any():
                 raise ValueError("Cannot convert NaT values to integer")
             return arr.view(dtype)
@@ -1009,6 +1017,14 @@ def astype_nansafe(
 
     elif is_timedelta64_dtype(arr):
         if dtype == np.int64:
+            warnings.warn(
+                f"casting {arr.dtype} values to int64 with .astype(...) "
+                "is deprecated and will raise in a future version. "
+                "Use .view(...) instead.",
+                FutureWarning,
+                # stacklevel chosen to be correct when reached via Series.astype
+                stacklevel=7,
+            )
             if isna(arr).any():
                 raise ValueError("Cannot convert NaT values to integer")
             return arr.view(dtype)

pandas/tests/arrays/test_datetimes.py

@@ -184,13 +184,18 @@ def test_astype_copies(self, dtype, other):
     @pytest.mark.parametrize("dtype", [int, np.int32, np.int64, "uint32", "uint64"])
     def test_astype_int(self, dtype):
         arr = DatetimeArray._from_sequence([pd.Timestamp("2000"), pd.Timestamp("2001")])
-        result = arr.astype(dtype)
+        with tm.assert_produces_warning(FutureWarning):
+            # astype(int..) deprecated
+            result = arr.astype(dtype)
 
         if np.dtype(dtype).kind == "u":
             expected_dtype = np.dtype("uint64")
         else:
             expected_dtype = np.dtype("int64")
-        expected = arr.astype(expected_dtype)
+
+        with tm.assert_produces_warning(FutureWarning):
+            # astype(int..) deprecated
+            expected = arr.astype(expected_dtype)
 
         assert result.dtype == expected_dtype
         tm.assert_numpy_array_equal(result, expected)

pandas/tests/arrays/test_period.py

@@ -123,26 +123,36 @@ def test_astype(dtype):
     # We choose to ignore the sign and size of integers for
     # Period/Datetime/Timedelta astype
     arr = period_array(["2000", "2001", None], freq="D")
-    result = arr.astype(dtype)
+    with tm.assert_produces_warning(FutureWarning):
+        # astype(int..) deprecated
+        result = arr.astype(dtype)
 
     if np.dtype(dtype).kind == "u":
         expected_dtype = np.dtype("uint64")
     else:
         expected_dtype = np.dtype("int64")
-    expected = arr.astype(expected_dtype)
+
+    with tm.assert_produces_warning(FutureWarning):
+        # astype(int..) deprecated
+        expected = arr.astype(expected_dtype)
 
     assert result.dtype == expected_dtype
     tm.assert_numpy_array_equal(result, expected)
 
 
 def test_astype_copies():
     arr = period_array(["2000", "2001", None], freq="D")
-    result = arr.astype(np.int64, copy=False)
+    with tm.assert_produces_warning(FutureWarning):
+        # astype(int..) deprecated
+        result = arr.astype(np.int64, copy=False)
+
     # Add the `.base`, since we now use `.asi8` which returns a view.
     # We could maybe override it in PeriodArray to return ._data directly.
     assert result.base is arr._data
 
-    result = arr.astype(np.int64, copy=True)
+    with tm.assert_produces_warning(FutureWarning):
+        # astype(int..) deprecated
+        result = arr.astype(np.int64, copy=True)
     assert result is not arr._data
     tm.assert_numpy_array_equal(result, arr._data.view("i8"))
 

pandas/tests/arrays/test_timedeltas.py

@@ -82,13 +82,18 @@ def test_from_sequence_dtype(self):
     @pytest.mark.parametrize("dtype", [int, np.int32, np.int64, "uint32", "uint64"])
     def test_astype_int(self, dtype):
         arr = TimedeltaArray._from_sequence([Timedelta("1H"), Timedelta("2H")])
-        result = arr.astype(dtype)
+        with tm.assert_produces_warning(FutureWarning):
+            # astype(int..) deprecated
+            result = arr.astype(dtype)
 
         if np.dtype(dtype).kind == "u":
             expected_dtype = np.dtype("uint64")
         else:
             expected_dtype = np.dtype("int64")
-        expected = arr.astype(expected_dtype)
+
+        with tm.assert_produces_warning(FutureWarning):
+            # astype(int..) deprecated
+            expected = arr.astype(expected_dtype)
 
         assert result.dtype == expected_dtype
         tm.assert_numpy_array_equal(result, expected)

pandas/tests/dtypes/test_common.py

@@ -719,7 +719,9 @@ def test_astype_nansafe(val, typ):
 
     msg = "Cannot convert NaT values to integer"
     with pytest.raises(ValueError, match=msg):
-        astype_nansafe(arr, dtype=typ)
+        with tm.assert_produces_warning(FutureWarning, check_stacklevel=False):
+            # datetimelike astype(int64) deprecated
+            astype_nansafe(arr, dtype=typ)
 
 
 @pytest.mark.parametrize("from_type", [np.datetime64, np.timedelta64])

pandas/tests/indexes/datetimes/methods/test_astype.py

@@ -29,7 +29,8 @@ def test_astype(self):
         )
         tm.assert_index_equal(result, expected)
 
-        result = idx.astype(int)
+        with tm.assert_produces_warning(FutureWarning, check_stacklevel=False):
+            result = idx.astype(int)
         expected = Int64Index(
             [1463356800000000000] + [-9223372036854775808] * 3,
             dtype=np.int64,
@@ -38,7 +39,8 @@ def test_astype(self):
         tm.assert_index_equal(result, expected)
 
         rng = date_range("1/1/2000", periods=10, name="idx")
-        result = rng.astype("i8")
+        with tm.assert_produces_warning(FutureWarning, check_stacklevel=False):
+            result = rng.astype("i8")
         tm.assert_index_equal(result, Index(rng.asi8, name="idx"))
         tm.assert_numpy_array_equal(result.values, rng.asi8)
 
@@ -48,9 +50,9 @@ def test_astype_uint(self):
             np.array([946684800000000000, 946771200000000000], dtype="uint64"),
             name="idx",
         )
-
-        tm.assert_index_equal(arr.astype("uint64"), expected)
-        tm.assert_index_equal(arr.astype("uint32"), expected)
+        with tm.assert_produces_warning(FutureWarning, check_stacklevel=False):
+            tm.assert_index_equal(arr.astype("uint64"), expected)
+            tm.assert_index_equal(arr.astype("uint32"), expected)
 
     def test_astype_with_tz(self):
 

pandas/tests/indexes/interval/test_astype.py

@@ -197,10 +197,13 @@ def index(self, request):
     @pytest.mark.parametrize("subtype", ["int64", "uint64"])
     def test_subtype_integer(self, index, subtype):
         dtype = IntervalDtype(subtype)
-        result = index.astype(dtype)
-        expected = IntervalIndex.from_arrays(
-            index.left.astype(subtype), index.right.astype(subtype), closed=index.closed
-        )
+        with tm.assert_produces_warning(FutureWarning, check_stacklevel=False):
+            result = index.astype(dtype)
+            expected = IntervalIndex.from_arrays(
+                index.left.astype(subtype),
+                index.right.astype(subtype),
+                closed=index.closed,
+            )
         tm.assert_index_equal(result, expected)
 
     def test_subtype_float(self, index):

pandas/tests/indexes/interval/test_constructors.py

@@ -72,13 +72,21 @@ def test_constructor(self, constructor, breaks, closed, name):
     )
     def test_constructor_dtype(self, constructor, breaks, subtype):
         # GH 19262: conversion via dtype parameter
-        expected_kwargs = self.get_kwargs_from_breaks(breaks.astype(subtype))
+        warn = None
+        if subtype == "int64" and breaks.dtype.kind in ["M", "m"]:
+            # astype(int64) deprecated
+            warn = FutureWarning
+
+        with tm.assert_produces_warning(warn, check_stacklevel=False):
+            expected_kwargs = self.get_kwargs_from_breaks(breaks.astype(subtype))
         expected = constructor(**expected_kwargs)
 
         result_kwargs = self.get_kwargs_from_breaks(breaks)
         iv_dtype = IntervalDtype(subtype)
         for dtype in (iv_dtype, str(iv_dtype)):
-            result = constructor(dtype=dtype, **result_kwargs)
+            with tm.assert_produces_warning(warn, check_stacklevel=False):
+
+                result = constructor(dtype=dtype, **result_kwargs)
             tm.assert_index_equal(result, expected)
 
     @pytest.mark.filterwarnings("ignore:Passing keywords other:FutureWarning")