Patch to make _codecs a builtin module. This is necessary since

Python 2.3 will support source code encodings which rely on the
builtin codecs being available to the parser.

Remove struct dependency from codecs.py

Lib/codecs.py

@@ -7,7 +7,7 @@ Written by Marc-Andre Lemburg (mal@lemburg.com).
 
 """#"
 
-import struct, __builtin__
+import __builtin__, sys
 
 ### Registry and builtin stateless codec functions
 
@@ -48,11 +48,21 @@ BOM_UTF32_LE = '\xff\xfe\x00\x00'
 # UTF-32, big endian
 BOM_UTF32_BE = '\x00\x00\xfe\xff'
 
-# UTF-16, native endianness
-BOM = BOM_UTF16 = struct.pack('=H', 0xFEFF)
+if sys.byteorder == 'little':
 
-# UTF-32, native endianness
-BOM_UTF32 = struct.pack('=L', 0x0000FEFF)
+    # UTF-16, native endianness
+    BOM = BOM_UTF16 = BOM_UTF16_LE
+
+    # UTF-32, native endianness
+    BOM_UTF32 = BOM_UTF32_LE
+
+else:
+
+    # UTF-16, native endianness
+    BOM = BOM_UTF16 = BOM_UTF16_BE
+
+    # UTF-32, native endianness
+    BOM_UTF32 = BOM_UTF32_BE
 
 # Old broken names (don't use in new code)
 BOM32_LE = BOM_UTF16_LE

Modules/Setup.config.in

@@ -10,4 +10,4 @@
 @USE_SIGNAL_MODULE@signal signalmodule.c
 
 # The rest of the modules previously listed in this file are built
-# by the setup.py script in Python 2.1.
+# by the setup.py script in Python 2.1 and later.

Modules/Setup.dist

@@ -111,6 +111,7 @@ PYTHONPATH=$(COREPYTHONPATH)
 posix posixmodule.c		# posix (UNIX) system calls
 errno errnomodule.c		# posix (UNIX) errno values
 _sre _sre.c			# Fredrik Lundh's new regular expressions
+_codecs _codecsmodule.c		# access to the builtin codecs and codec registry
 
 # The rest of the modules listed in this file are all commented out by
 # default.  Usually they can be detected and built as dynamically
@@ -163,7 +164,6 @@ GLHACK=-Dclear=__GLclear
 #time timemodule.c # -lm # time operations and variables
 #operator operator.c	# operator.add() and similar goodies
 #_weakref _weakref.c	# basic weak reference support
-#_codecs _codecsmodule.c	# access to the builtin codecs and codec registry
 #_testcapi _testcapimodule.c    # Python C API test module
 
 #unicodedata unicodedata.c    # static Unicode character database

setup.py

@@ -313,8 +313,6 @@ class PyBuildExt(build_ext):
                                libraries=math_libs) )
         # operator.add() and similar goodies
         exts.append( Extension('operator', ['operator.c']) )
-        # access to the builtin codecs and codec registry
-        exts.append( Extension('_codecs', ['_codecsmodule.c']) )
         # Python C API test module
         exts.append( Extension('_testcapi', ['_testcapimodule.c']) )
         # static Unicode character database