Add more numeric tests for NumIndex

pandas/core/indexes/base.py

@@ -442,7 +442,7 @@ def __new__(
             return Index._simple_new(data, name=name)
 
         # index-like
-        elif type(data) is NumIndex and dtype is None:
+        elif isinstance(data, NumIndex) and data._is_num_index() and dtype is None:
             return NumIndex(data, name=name, copy=copy)
         elif isinstance(data, (np.ndarray, Index, ABCSeries)):
 
@@ -2407,6 +2407,26 @@ def is_all_dates(self) -> bool:
         )
         return self._is_all_dates
 
+    def _is_num_index(self) -> bool:
+        """
+        Whether self is a NumIndex, but not *not* Int64Index, UInt64Index, FloatIndex.
+
+        Typically used to check if an operation should return NumIndex or plain Index.
+        """
+        from pandas.core.indexes.numeric import (
+            Float64Index,
+            Int64Index,
+            NumIndex,
+            UInt64Index,
+        )
+
+        if not isinstance(self, NumIndex):
+            return False
+        elif isinstance(self, (Int64Index, UInt64Index, Float64Index)):
+            return False
+        else:
+            return True
+
     # --------------------------------------------------------------------
     # Pickle Methods
 
@@ -5488,8 +5508,8 @@ def map(self, mapper, na_action=None):
             # empty
             attributes["dtype"] = self.dtype
 
-        if type(self) is NumIndex:
-            return type(self)(new_values, **attributes)
+        if self._is_num_index() and issubclass(new_values.dtype.type, np.number):
+            return NumIndex(new_values, **attributes)
 
         return Index(new_values, **attributes)
 

pandas/core/indexes/numeric.py

@@ -288,11 +288,11 @@ def astype(self, dtype, copy=True):
                 # TODO(jreback); this can change once we have an EA Index type
                 # GH 13149
                 arr = astype_nansafe(self._values, dtype=dtype)
-                if isinstance(self, Float64Index):
+                if not self._is_num_index():
                     return Int64Index(arr, name=self.name)
                 else:
                     return NumIndex(arr, name=self.name, dtype=dtype)
-        elif is_categorical_dtype(dtype):
+        if is_categorical_dtype(dtype):
             from pandas import CategoricalIndex
 
             return CategoricalIndex(self, name=self.name, dtype=dtype, copy=copy)

pandas/tests/indexes/common.py

@@ -12,7 +12,7 @@
 from pandas.core.dtypes.dtypes import CategoricalDtype
 
 import pandas as pd
-from pandas import (  # noqa
+from pandas import (
     CategoricalIndex,
     DatetimeIndex,
     Float64Index,
@@ -29,6 +29,7 @@
 )
 import pandas._testing as tm
 from pandas.core.indexes.datetimelike import DatetimeIndexOpsMixin
+from pandas.core.indexes.numeric import NumIndex
 
 
 class Base:
@@ -343,12 +344,13 @@ def test_numpy_argsort(self, index):
     def test_repeat(self, simple_index):
         rep = 2
         idx = simple_index.copy()
-        expected = Index(idx.values.repeat(rep), name=idx.name)
+        new_index_cls = type(idx) if not isinstance(idx, RangeIndex) else Int64Index
+        expected = new_index_cls(idx.values.repeat(rep), name=idx.name)
         tm.assert_index_equal(idx.repeat(rep), expected)
 
         idx = simple_index
         rep = np.arange(len(idx))
-        expected = Index(idx.values.repeat(rep), name=idx.name)
+        expected = new_index_cls(idx.values.repeat(rep), name=idx.name)
         tm.assert_index_equal(idx.repeat(rep), expected)
 
     def test_numpy_repeat(self, simple_index):
@@ -649,7 +651,12 @@ def test_map_dictlike(self, mapper, simple_index):
         tm.assert_index_equal(result, expected)
 
         # empty mappable
-        expected = Index([np.nan] * len(idx))
+        if idx._is_num_index():
+            new_index_cls = NumIndex
+        else:
+            new_index_cls = Float64Index
+
+        expected = new_index_cls([np.nan] * len(idx))
         result = idx.map(mapper(expected, idx))
         tm.assert_index_equal(result, expected)
 

pandas/tests/indexes/numeric/test_numeric.py

@@ -13,17 +13,25 @@
     UInt64Index,
 )
 import pandas._testing as tm
+from pandas.core.indexes.numeric import NumIndex
 from pandas.tests.indexes.common import NumericBase
 
 
 class TestFloat64Index(NumericBase):
     _index_cls = Float64Index
     _dtype = np.float64
 
-    @pytest.fixture
-    def simple_index(self) -> Index:
-        values = np.arange(5, dtype=self._dtype)
-        return self._index_cls(values)
+    @pytest.fixture(
+        params=[
+            (Float64Index, None),
+            (NumIndex, np.float64),
+            (NumIndex, np.float32),
+        ],
+    )
+    def simple_index(self, request) -> Index:
+        index_cls, dtype = request.param
+        values = np.arange(5, dtype=dtype)
+        return index_cls(values)
 
     @pytest.fixture(
         params=[
@@ -392,9 +400,19 @@ class TestInt64Index(NumericInt):
     _index_cls = Int64Index
     _dtype = np.int64
 
-    @pytest.fixture
-    def simple_index(self) -> Index:
-        return self._index_cls(range(0, 20, 2), dtype=self._dtype)
+    @pytest.fixture(
+        params=[
+            (Int64Index, None),
+            (NumIndex, np.int64),
+            (NumIndex, np.int32),
+            (NumIndex, np.int16),
+            (NumIndex, np.int8),
+        ],
+    )
+    def simple_index(self, request) -> Index:
+        index_cls, dtype = request.param
+        values = np.arange(5, dtype=dtype)
+        return index_cls(values)
 
     @pytest.fixture(
         params=[range(0, 20, 2), range(19, -1, -1)], ids=["index_inc", "index_dec"]
@@ -490,10 +508,19 @@ class TestUInt64Index(NumericInt):
     _index_cls = UInt64Index
     _dtype = np.uint64
 
-    @pytest.fixture
-    def simple_index(self) -> Index:
-        # compat with shared Int64/Float64 tests
-        return self._index_cls(np.arange(5, dtype=self._dtype))
+    @pytest.fixture(
+        params=[
+            (UInt64Index, None),
+            (NumIndex, np.uint64),
+            (NumIndex, np.uint32),
+            (NumIndex, np.uint16),
+            (NumIndex, np.uint8),
+        ],
+    )
+    def simple_index(self, request) -> Index:
+        index_cls, dtype = request.param
+        values = np.arange(5, dtype=dtype)
+        return index_cls(values)
 
     @pytest.fixture(
         params=[