Let ctypes._CData.from_buffer[_copy] accept any buffer object (#2610)

Fixes the issue discussed here:
https://github.com/python/typeshed/pull/1906#discussion_r234798828

stdlib/2and3/ctypes/__init__.pyi

@@ -1,5 +1,6 @@
 # Stubs for ctypes
 
+from array import array
 from typing import (
     Any, Callable, ClassVar, Iterator, Iterable, List, Mapping, Optional, Sequence, Sized, Text,
     Tuple, Type, Generic, TypeVar, overload,
@@ -45,6 +46,13 @@ if sys.platform == 'win32':
 pydll: LibraryLoader[PyDLL] = ...
 pythonapi: PyDLL = ...
 
+# Anything that implements the read-write buffer interface.
+# The buffer interface is defined purely on the C level, so we cannot define a normal Protocol
+# for it. Instead we have to list the most common stdlib buffer classes in a Union.
+_WritableBuffer = _UnionT[bytearray, memoryview, array, _CData]
+# Same as _WritableBuffer, but also includes read-only buffer types (like bytes).
+_ReadOnlyBuffer = _UnionT[_WritableBuffer, bytes]
+
 class _CDataMeta(type):
     # By default mypy complains about the following two methods, because strictly speaking cls
     # might not be a Type[_CT]. However this can never actually happen, because the only class that
@@ -56,9 +64,9 @@ class _CData(metaclass=_CDataMeta):
     _b_needsfree_: bool = ...
     _objects: Optional[Mapping[Any, int]] = ...
     @classmethod
-    def from_buffer(cls: Type[_CT], source: bytearray, offset: int = ...) -> _CT: ...
+    def from_buffer(cls: Type[_CT], source: _WritableBuffer, offset: int = ...) -> _CT: ...
     @classmethod
-    def from_buffer_copy(cls: Type[_CT], source: bytearray, offset: int = ...) -> _CT: ...
+    def from_buffer_copy(cls: Type[_CT], source: _ReadOnlyBuffer, offset: int = ...) -> _CT: ...
     @classmethod
     def from_address(cls: Type[_CT], address: int) -> _CT: ...
     @classmethod