Remove mention of narrow/wide builds and update array doc, add a test.

Doc/library/array.rst

@@ -21,7 +21,7 @@ defined:
 +-----------+--------------------+-------------------+-----------------------+-------+
 | ``'B'``   | unsigned char      | int               | 1                     |       |
 +-----------+--------------------+-------------------+-----------------------+-------+
-| ``'u'``   | Py_UNICODE         | Unicode character | 2                     | \(1)  |
+| ``'u'``   | Py_UCS4            | Unicode character | 4                     |       |
 +-----------+--------------------+-------------------+-----------------------+-------+
 | ``'h'``   | signed short       | int               | 2                     |       |
 +-----------+--------------------+-------------------+-----------------------+-------+
@@ -35,9 +35,9 @@ defined:
 +-----------+--------------------+-------------------+-----------------------+-------+
 | ``'L'``   | unsigned long      | int               | 4                     |       |
 +-----------+--------------------+-------------------+-----------------------+-------+
-| ``'q'``   | signed long long   | int               | 8                     | \(2)  |
+| ``'q'``   | signed long long   | int               | 8                     | \(1)  |
 +-----------+--------------------+-------------------+-----------------------+-------+
-| ``'Q'``   | unsigned long long | int               | 8                     | \(2)  |
+| ``'Q'``   | unsigned long long | int               | 8                     | \(1)  |
 +-----------+--------------------+-------------------+-----------------------+-------+
 | ``'f'``   | float              | float             | 4                     |       |
 +-----------+--------------------+-------------------+-----------------------+-------+
@@ -47,10 +47,6 @@ defined:
 Notes:
 
 (1)
-   The ``'u'`` type code corresponds to Python's unicode character.  On narrow
-   Unicode builds this is 2-bytes, on wide builds this is 4-bytes.
-
-(2)
    The ``'q'`` and ``'Q'`` type codes are available only if
    the platform C compiler used to build Python supports C :c:type:`long long`,
    or, on Windows, :c:type:`__int64`.

Lib/test/test_array.py

@@ -1015,7 +1015,7 @@ class UnicodeTest(StringTest):
     smallerexample = '\x01\u263a\x00\ufefe'
     biggerexample = '\x01\u263a\x01\ufeff'
     outside = str('\x33')
-    minitemsize = 2
+    minitemsize = 4
 
     def test_unicode(self):
         self.assertRaises(TypeError, array.array, 'b', 'foo')
@@ -1027,6 +1027,7 @@ class UnicodeTest(StringTest):
         a.fromunicode('\x11abc\xff\u1234')
         s = a.tounicode()
         self.assertEqual(s, '\xa0\xc2\u1234 \x11abc\xff\u1234')
+        self.assertEqual(a.itemsize, 4)
 
         s = '\x00="\'a\\b\x80\xff\u0000\u0001\u1234'
         a = array.array('u', s)

Modules/arraymodule.c

@@ -2593,7 +2593,7 @@ is a single character.  The following type codes are defined:\n\
     Type code   C Type             Minimum size in bytes \n\
     'b'         signed integer     1 \n\
     'B'         unsigned integer   1 \n\
-    'u'         Unicode character  2 (see note) \n\
+    'u'         Unicode character  4 \n\
     'h'         signed integer     2 \n\
     'H'         unsigned integer   2 \n\
     'i'         signed integer     2 \n\
@@ -2605,9 +2605,6 @@ is a single character.  The following type codes are defined:\n\
     'f'         floating point     4 \n\
     'd'         floating point     8 \n\
 \n\
-NOTE: The 'u' type code corresponds to Python's unicode character. On \n\
-narrow builds this is 2-bytes on wide builds this is 4-bytes.\n\
-\n\
 NOTE: The 'q' and 'Q' type codes are only available if the platform \n\
 C compiler used to build Python supports 'long long', or, on Windows, \n\
 '__int64'.\n\