Merged revisions 74000-74001 via svnmerge from
svn+ssh://pythondev@svn.python.org/python/trunk

........
  r74000 | amaury.forgeotdarc | 2009-07-13 22:01:11 +0200 (lun., 13 juil. 2009) | 4 lines

  #1616979: Add the cp720 (Arabic DOS) encoding.
  Since there is no official mapping file from unicode.org,
  the codec file is generated on Windows with the new genwincodec.py script.
........
  r74001 | amaury.forgeotdarc | 2009-07-13 22:03:21 +0200 (lun., 13 juil. 2009) | 2 lines

  NEWS entry for r74000.
........
diff --git a/Tools/unicode/genwincodec.py b/Tools/unicode/genwincodec.py
new file mode 100644
index 0000000..17ba809
--- /dev/null
+++ b/Tools/unicode/genwincodec.py
@@ -0,0 +1,61 @@
+"""This script generates a Python codec module from a Windows Code Page.

+

+It uses the function MultiByteToWideChar to generate a decoding table.

+"""

+

+import ctypes

+from ctypes import wintypes

+from gencodec import codegen

+import unicodedata

+

+def genwinmap(codepage):

+    MultiByteToWideChar = ctypes.windll.kernel32.MultiByteToWideChar

+    MultiByteToWideChar.argtypes = [wintypes.UINT, wintypes.DWORD,

+                                    wintypes.LPCSTR, ctypes.c_int,

+                                    wintypes.LPWSTR, ctypes.c_int]

+    MultiByteToWideChar.restype = ctypes.c_int

+

+    enc2uni = {}

+

+    for i in list(range(32)) + [127]:

+        enc2uni[i] = (i, 'CONTROL CHARACTER')

+

+    for i in range(256):

+        buf = ctypes.create_unicode_buffer(2)

+        ret = MultiByteToWideChar(

+            codepage, 0,

+            chr(i), 1,

+            buf, 2)

+        assert ret == 1, "invalid code page"

+        assert buf[1] == '\x00'

+        try:

+            name = unicodedata.name(buf[0])

+        except ValueError:

+            try:

+                name = enc2uni[i][1]

+            except KeyError:

+                name = ''

+

+        enc2uni[i] = (ord(buf[0]), name)

+

+    return enc2uni

+

+def genwincodec(codepage):

+    import platform

+    map = genwinmap(codepage)

+    encodingname = 'cp%d' % codepage

+    code = codegen("", map, encodingname)

+    # Replace first lines with our own docstring

+    code = '''\

+"""Python Character Mapping Codec %s generated on Windows:

+%s with the command:

+  python Tools/unicode/genwincodec.py %s

+"""#"

+''' % (encodingname, ' '.join(platform.win32_ver()), codepage

+      ) + code.split('"""#"', 1)[1]

+

+    print(code)

+

+if __name__ == '__main__':

+    import sys

+    genwincodec(int(sys.argv[1]))