Add some dtypes (#179)

* add some dtypes

* remove DType import (just use the DType class from the same file)

* bool8 and bool1

* bool1, bool8 -> bool

Author: MarcoGorelli

spec/API_specification/dataframe_api/__init__.py

@@ -9,8 +9,6 @@
 from .dataframe_object import *
 from .groupby_object import *
 
-from ._types import DType
-
 
 __all__ = [
     "__dataframe_api_version",
@@ -19,6 +17,18 @@
     "dataframe_from_dict",
     "is_null",
     "null",
+    "DType",
+    "Int64",
+    "Int32",
+    "Int16",
+    "Int8",
+    "UInt64",
+    "UInt32",
+    "UInt16",
+    "UInt8",
+    "Float64",
+    "Float32",
+    "Bool",
 ]
 
 
@@ -129,3 +139,43 @@ def is_null(value: object, /) -> bool:
         implements the dataframe API standard, False otherwise.
 
     """
+
+##########
+# Dtypes #
+##########
+
+class DType:
+    """Base class for all dtypes."""
+
+class Int64(DType):
+    """Integer type with 64 bits of precision."""
+
+class Int32(DType):
+    """Integer type with 32 bits of precision."""
+
+class Int16(DType):
+    """Integer type with 16 bits of precision."""
+
+class Int8(DType):
+    """Integer type with 8 bits of precision."""
+
+class UInt64(DType):
+    """Unsigned integer type with 64 bits of precision."""
+
+class UInt32(DType):
+    """Unsigned integer type with 32 bits of precision."""
+
+class UInt16(DType):
+    """Unsigned integer type with 16 bits of precision."""
+
+class UInt8(DType):
+    """Unsigned integer type with 8 bits of precision."""
+
+class Float64(DType):
+    """Floating point type with 64 bits of precision."""
+
+class Float32(DType):
+    """Floating point type with 32 bits of precision."""
+
+class Bool(DType):
+    """Boolean type with 8 bits of precision."""

spec/API_specification/dataframe_api/column_object.py

@@ -3,7 +3,8 @@
 from typing import NoReturn, Sequence, TYPE_CHECKING
 
 if TYPE_CHECKING:
-    from ._types import Scalar, DType
+    from ._types import Scalar
+    from . import DType
 
 
 __all__ = ['Column']
@@ -52,7 +53,7 @@ def get_rows(self, indices: Column[int]) -> Column:
         """
         ...
 
-    def get_value(self, row_number: int) -> DType:
+    def get_value(self, row_number: int) -> Scalar:
         """
         Select the value at a row number, similar to `ndarray.__getitem__(<int>)`.
 
@@ -63,7 +64,7 @@ def get_value(self, row_number: int) -> DType:
         
         Returns
         -------
-        dtype
+        Scalar
             Depends on the dtype of the Column, and may vary
             across implementations.
         """
@@ -369,56 +370,56 @@ def all(self, *, skip_nulls: bool = True) -> bool:
             If column is not boolean.
         """
 
-    def min(self, *, skip_nulls: bool = True) -> DType:
+    def min(self, *, skip_nulls: bool = True) -> Scalar:
         """
         Reduction returns a scalar. Any data type that supports comparisons
         must be supported. The returned value has the same dtype as the column.
         """
 
-    def max(self, *, skip_nulls: bool = True) -> DType:
+    def max(self, *, skip_nulls: bool = True) -> Scalar:
         """
         Reduction returns a scalar. Any data type that supports comparisons
         must be supported. The returned value has the same dtype as the column.
         """
 
-    def sum(self, *, skip_nulls: bool = True) -> DType:
+    def sum(self, *, skip_nulls: bool = True) -> Scalar:
         """
         Reduction returns a scalar. Must be supported for numerical and
         datetime data types. The returned value has the same dtype as the
         column.
         """
 
-    def prod(self, *, skip_nulls: bool = True) -> DType:
+    def prod(self, *, skip_nulls: bool = True) -> Scalar:
         """
         Reduction returns a scalar. Must be supported for numerical data types.
         The returned value has the same dtype as the column.
         """
 
-    def median(self, *, skip_nulls: bool = True) -> DType:
+    def median(self, *, skip_nulls: bool = True) -> Scalar:
         """
         Reduction returns a scalar. Must be supported for numerical and
         datetime data types. Returns a float for numerical data types, and
         datetime (with the appropriate timedelta format string) for datetime
         dtypes.
         """
 
-    def mean(self, *, skip_nulls: bool = True) -> DType:
+    def mean(self, *, skip_nulls: bool = True) -> Scalar:
         """
         Reduction returns a scalar. Must be supported for numerical and
         datetime data types. Returns a float for numerical data types, and
         datetime (with the appropriate timedelta format string) for datetime
         dtypes.
         """
 
-    def std(self, *, skip_nulls: bool = True) -> DType:
+    def std(self, *, skip_nulls: bool = True) -> Scalar:
         """
         Reduction returns a scalar. Must be supported for numerical and
         datetime data types. Returns a float for numerical data types, and
         datetime (with the appropriate timedelta format string) for datetime
         dtypes.
         """
 
-    def var(self, *, skip_nulls: bool = True) -> DType:
+    def var(self, *, skip_nulls: bool = True) -> Scalar:
         """
         Reduction returns a scalar. Must be supported for numerical and
         datetime data types. Returns a float for numerical data types, and

spec/API_specification/index.rst

@@ -16,6 +16,18 @@ of objects and functions in the top-level namespace. The latter are:
    __dataframe_api_version__
    is_null
    null
+   DType
+   Int64
+   Int32
+   Int16
+   Int8
+   UInt64
+   UInt32
+   UInt16
+   UInt8
+   Float64
+   Float32
+   Bool
 
 The ``DataFrame``, ``Column`` and ``GroupBy`` objects have the following
 methods and attributes: