Backport PR #55460 on branch 2.1.x (COMPAT: Update old numpy aliases) (#55463)

Backport PR #55460: COMPAT: Update old numpy aliases

Co-authored-by: Matthew Roeschke <[email protected]>

Author: meeseeksmachine

pandas/io/stata.py

@@ -977,7 +977,7 @@ def __init__(self) -> None:
         # with a label, but the underlying variable is -127 to 100
         # we're going to drop the label and cast to int
         self.DTYPE_MAP = dict(
-            list(zip(range(1, 245), [np.dtype("a" + str(i)) for i in range(1, 245)]))
+            [(i, np.dtype(f"S{i}")) for i in range(1, 245)]
             + [
                 (251, np.dtype(np.int8)),
                 (252, np.dtype(np.int16)),

pandas/tests/frame/constructors/test_from_records.py

@@ -281,7 +281,7 @@ def test_frame_from_records_utc(self):
 
     def test_from_records_to_records(self):
         # from numpy documentation
-        arr = np.zeros((2,), dtype=("i4,f4,a10"))
+        arr = np.zeros((2,), dtype=("i4,f4,S10"))
         arr[:] = [(1, 2.0, "Hello"), (2, 3.0, "World")]
 
         DataFrame.from_records(arr)

pandas/tests/frame/methods/test_select_dtypes.py

@@ -339,9 +339,7 @@ def test_select_dtypes_datetime_with_tz(self):
         expected = df3.reindex(columns=[])
         tm.assert_frame_equal(result, expected)
 
-    @pytest.mark.parametrize(
-        "dtype", [str, "str", np.bytes_, "S1", "unicode", np.str_, "U1"]
-    )
+    @pytest.mark.parametrize("dtype", [str, "str", np.bytes_, "S1", np.str_, "U1"])
     @pytest.mark.parametrize("arg", ["include", "exclude"])
     def test_select_dtypes_str_raises(self, dtype, arg):
         df = DataFrame(

pandas/tests/frame/methods/test_to_records.py

@@ -253,7 +253,7 @@ def test_to_records_with_categorical(self):
             ),
             # Pass in a dtype instance.
             (
-                {"column_dtypes": np.dtype("unicode")},
+                {"column_dtypes": np.dtype(np.str_)},
                 np.rec.array(
                     [("0", "1", "0.2", "a"), ("1", "2", "1.5", "bc")],
                     dtype=[

pandas/tests/indexes/base_class/test_setops.py

@@ -182,23 +182,23 @@ def test_symmetric_difference(self):
                 "intersection",
                 np.array(
                     [(1, "A"), (2, "A"), (1, "B"), (2, "B")],
-                    dtype=[("num", int), ("let", "a1")],
+                    dtype=[("num", int), ("let", "S1")],
                 ),
                 False,
             ),
             (
                 "intersection",
                 np.array(
                     [(1, "A"), (1, "B"), (2, "A"), (2, "B")],
-                    dtype=[("num", int), ("let", "a1")],
+                    dtype=[("num", int), ("let", "S1")],
                 ),
                 None,
             ),
             (
                 "union",
                 np.array(
                     [(1, "A"), (1, "B"), (1, "C"), (2, "A"), (2, "B"), (2, "C")],
-                    dtype=[("num", int), ("let", "a1")],
+                    dtype=[("num", int), ("let", "S1")],
                 ),
                 None,
             ),
@@ -208,13 +208,13 @@ def test_tuple_union_bug(self, method, expected, sort):
         index1 = Index(
             np.array(
                 [(1, "A"), (2, "A"), (1, "B"), (2, "B")],
-                dtype=[("num", int), ("let", "a1")],
+                dtype=[("num", int), ("let", "S1")],
             )
         )
         index2 = Index(
             np.array(
                 [(1, "A"), (2, "A"), (1, "B"), (2, "B"), (1, "C"), (2, "C")],
-                dtype=[("num", int), ("let", "a1")],
+                dtype=[("num", int), ("let", "S1")],
             )
         )
 

pandas/tests/io/json/test_pandas.py

@@ -599,7 +599,7 @@ def test_blocks_compat_GH9037(self):
         )
 
         # JSON deserialisation always creates unicode strings
-        df_mixed.columns = df_mixed.columns.astype("unicode")
+        df_mixed.columns = df_mixed.columns.astype(np.str_)
         data = StringIO(df_mixed.to_json(orient="split"))
         df_roundtrip = read_json(data, orient="split")
         tm.assert_frame_equal(

pandas/tests/reshape/concat/test_append.py

@@ -91,10 +91,10 @@ def test_append_length0_frame(self, sort):
         tm.assert_frame_equal(df5, expected)
 
     def test_append_records(self):
-        arr1 = np.zeros((2,), dtype=("i4,f4,a10"))
+        arr1 = np.zeros((2,), dtype=("i4,f4,S10"))
         arr1[:] = [(1, 2.0, "Hello"), (2, 3.0, "World")]
 
-        arr2 = np.zeros((3,), dtype=("i4,f4,a10"))
+        arr2 = np.zeros((3,), dtype=("i4,f4,S10"))
         arr2[:] = [(3, 4.0, "foo"), (5, 6.0, "bar"), (7.0, 8.0, "baz")]
 
         df1 = DataFrame(arr1)

pandas/tests/series/methods/test_astype.py

@@ -403,12 +403,12 @@ def test_astype_unicode(self):
             #  bytes with obj.decode() instead of str(obj)
             item = "野菜食べないとやばい"
             ser = Series([item.encode()])
-            result = ser.astype("unicode")
+            result = ser.astype(np.str_)
             expected = Series([item])
             tm.assert_series_equal(result, expected)
 
         for ser in test_series:
-            res = ser.astype("unicode")
+            res = ser.astype(np.str_)
             expec = ser.map(str)
             tm.assert_series_equal(res, expec)