API: register_extension_dtype class decorator

doc/source/api.rst

@@ -2559,6 +2559,7 @@ objects.
 .. autosummary::
    :toctree: generated/
 
+   api.extensions.register_extension_dtype
    api.extensions.register_dataframe_accessor
    api.extensions.register_series_accessor
    api.extensions.register_index_accessor

doc/source/whatsnew/v0.24.0.txt

@@ -491,6 +491,7 @@ ExtensionType Changes
 - :meth:`Series.combine()` works correctly with :class:`~pandas.api.extensions.ExtensionArray` inside of :class:`Series` (:issue:`20825`)
 - :meth:`Series.combine()` with scalar argument now works for any function type (:issue:`21248`)
 - :meth:`Series.astype` and :meth:`DataFrame.astype` now dispatch to :meth:`ExtensionArray.astype` (:issue:`21185:`).
+- Added :meth:`pandas.api.types.register_extension_dtype` to register an extension type with pandas (:issue:`22664`)
 
 .. _whatsnew_0240.api.incompatibilities:
 

pandas/api/extensions/__init__.py

@@ -5,4 +5,6 @@
 from pandas.core.algorithms import take  # noqa
 from pandas.core.arrays.base import (ExtensionArray,    # noqa
                                      ExtensionScalarOpsMixin)
-from pandas.core.dtypes.dtypes import ExtensionDtype  # noqa
+from pandas.core.dtypes.dtypes import (  # noqa
+    ExtensionDtype, register_extension_dtype
+)

pandas/core/arrays/integer.py

@@ -19,7 +19,7 @@
     is_list_like)
 from pandas.core.arrays import ExtensionArray, ExtensionOpsMixin
 from pandas.core.dtypes.base import ExtensionDtype
-from pandas.core.dtypes.dtypes import registry
+from pandas.core.dtypes.dtypes import register_extension_dtype
 from pandas.core.dtypes.missing import isna, notna
 
 from pandas.io.formats.printing import (
@@ -614,9 +614,9 @@ def integer_arithmetic_method(self, other):
     classname = "{}Dtype".format(name)
     attributes_dict = {'type': getattr(np, dtype),
                        'name': name}
-    dtype_type = type(classname, (_IntegerDtype, ), attributes_dict)
+    dtype_type = register_extension_dtype(
+        type(classname, (_IntegerDtype, ), attributes_dict)
+    )
     setattr(module, classname, dtype_type)
 
-    # register
-    registry.register(dtype_type)
     _dtypes[dtype] = dtype_type()

pandas/core/dtypes/base.py

@@ -127,7 +127,8 @@ class ExtensionDtype(_DtypeOpsMixin):
     * _is_numeric
 
     Optionally one can override construct_array_type for construction
-    with the name of this dtype via the Registry
+    with the name of this dtype via the Registry. See
+    :meth:`pandas.api.extensions.register_extension_dtype`.
 
     * construct_array_type
 
@@ -138,6 +139,11 @@ class ExtensionDtype(_DtypeOpsMixin):
     Methods and properties required by the interface raise
     ``pandas.errors.AbstractMethodError`` and no ``register`` method is
     provided for registering virtual subclasses.
+
+    See Also
+    --------
+    pandas.api.extensions.register_extension_dtype
+    pandas.api.extensions.ExtensionArray
     """
 
     def __str__(self):

pandas/core/dtypes/dtypes.py

@@ -8,6 +8,26 @@
 from .base import ExtensionDtype, _DtypeOpsMixin
 
 
+def register_extension_dtype(cls):
+    """Class decorator to register an ExtensionType with pandas.
+
+    .. versionadded:: 0.24.0
+
+    This enables operations like ``.astype(name)`` for the name
+    of the ExtensionDtype.
+
+    Examples
+    --------
+    >>> from pandas.api.extensions import register_extension_dtype
+    >>> from pandas.api.extensions import ExtensionDtype
+    >>> @register_extension_dtype
+    ... class MyExtensionDtype(ExtensionDtype):
+    ...     pass
+    """
+    registry.register(cls)
+    return cls
+
+
 class Registry(object):
     """
     Registry for dtype inference
@@ -17,10 +37,6 @@ class Registry(object):
 
     Multiple extension types can be registered.
     These are tried in order.
-
-    Examples
-    --------
-    registry.register(MyExtensionDtype)
     """
     def __init__(self):
         self.dtypes = []
@@ -65,9 +81,6 @@ def find(self, dtype):
 
 
 registry = Registry()
-# TODO(Extension): remove the second registry once all internal extension
-# dtypes are real extension dtypes.
-_pandas_registry = Registry()
 
 
 class PandasExtensionDtype(_DtypeOpsMixin):
@@ -145,6 +158,7 @@ class CategoricalDtypeType(type):
     pass
 
 
+@register_extension_dtype
 class CategoricalDtype(PandasExtensionDtype, ExtensionDtype):
     """
     Type for categorical data with the categories and orderedness
@@ -692,6 +706,7 @@ class IntervalDtypeType(type):
     pass
 
 
+@register_extension_dtype
 class IntervalDtype(PandasExtensionDtype, ExtensionDtype):
     """
     A Interval duck-typed class, suitable for holding an interval
@@ -824,8 +839,9 @@ def is_dtype(cls, dtype):
         return super(IntervalDtype, cls).is_dtype(dtype)
 
 
-# register the dtypes in search order
-registry.register(IntervalDtype)
-registry.register(CategoricalDtype)
+# TODO(Extension): remove the second registry once all internal extension
+# dtypes are real extension dtypes.
+_pandas_registry = Registry()
+
 _pandas_registry.register(DatetimeTZDtype)
 _pandas_registry.register(PeriodDtype)

pandas/tests/extension/base/dtype.py

@@ -1,4 +1,3 @@
-import pytest
 import numpy as np
 import pandas as pd
 
@@ -51,10 +50,6 @@ def test_eq_with_numpy_object(self, dtype):
     def test_array_type(self, data, dtype):
         assert dtype.construct_array_type() is type(data)
 
-    def test_array_type_with_arg(self, data, dtype):
-        with pytest.raises(NotImplementedError):
-            dtype.construct_array_type('foo')
-
     def test_check_dtype(self, data):
         dtype = data.dtype
 

pandas/tests/extension/decimal/test_decimal.py

@@ -105,9 +105,7 @@ def assert_frame_equal(self, left, right, *args, **kwargs):
 
 
 class TestDtype(BaseDecimal, base.BaseDtypeTests):
-
-    def test_array_type_with_arg(self, data, dtype):
-        assert dtype.construct_array_type() is DecimalArray
+    pass
 
 
 class TestInterface(BaseDecimal, base.BaseInterfaceTests):

pandas/tests/extension/json/test_json.py

@@ -116,9 +116,7 @@ def assert_frame_equal(self, left, right, *args, **kwargs):
 
 
 class TestDtype(BaseJSON, base.BaseDtypeTests):
-
-    def test_array_type_with_arg(self, data, dtype):
-        assert dtype.construct_array_type() is JSONArray
+    pass
 
 
 class TestInterface(BaseJSON, base.BaseInterfaceTests):

pandas/tests/extension/test_categorical.py

@@ -77,9 +77,7 @@ def data_for_grouping():
 
 
 class TestDtype(base.BaseDtypeTests):
-
-    def test_array_type_with_arg(self, data, dtype):
-        assert dtype.construct_array_type() is Categorical
+    pass
 
 
 class TestInterface(base.BaseInterfaceTests):

pandas/tests/extension/test_integer.py

@@ -20,7 +20,7 @@
 from pandas.tests.extension import base
 from pandas.core.dtypes.common import is_extension_array_dtype
 
-from pandas.core.arrays import IntegerArray, integer_array
+from pandas.core.arrays import integer_array
 from pandas.core.arrays.integer import (
     Int8Dtype, Int16Dtype, Int32Dtype, Int64Dtype,
     UInt8Dtype, UInt16Dtype, UInt32Dtype, UInt64Dtype)
@@ -92,9 +92,6 @@ def test_is_dtype_unboxes_dtype(self):
         # we have multiple dtypes, so skip
         pass
 
-    def test_array_type_with_arg(self, data, dtype):
-        assert dtype.construct_array_type() is IntegerArray
-
 
 class TestArithmeticOps(base.BaseArithmeticOpsTests):
 

pandas/tests/extension/test_interval.py

@@ -84,9 +84,7 @@ class BaseInterval(object):
 
 
 class TestDtype(BaseInterval, base.BaseDtypeTests):
-
-    def test_array_type_with_arg(self, data, dtype):
-        assert dtype.construct_array_type() is IntervalArray
+    pass
 
 
 class TestCasting(BaseInterval, base.BaseCastingTests):