svn+ssh://pythondev@svn.python.org/python/trunk

........
  r57127 | neal.norwitz | 2007-08-16 21:10:55 -0700 (Thu, 16 Aug 2007) | 1 line

  Update the tools before building the docs
........

svn+ssh://pythondev@svn.python.org/python/trunk

........
  r57123 | guido.van.rossum | 2007-08-16 20:30:08 -0700 (Thu, 16 Aug 2007) | 2 lines

  Ignore *.py[co] files.
........
  r57124 | guido.van.rossum | 2007-08-16 20:36:03 -0700 (Thu, 16 Aug 2007) | 2 lines

  Ignore a few more files.
........

using the message generated by svnmerge, because it contains a lot of
stuff about the Doc tree, which I'm not merging this time due to the
way the Doc tree was initially added.  I am however adding roman.py
which was added later to Doc/tools.  I'll try to diff the two Doc
trees separately to see if there's stuff I missed.

svn+ssh://pythondev@svn.python.org/python/trunk

ut-32-be). On narrow builds the codecs combine surrogate pairs in the unicode
object into one codepoint on encoding and create surrogate pairs for
codepoints outside the BMP on decoding. Lone surrogates are passed through
unchanged in all cases.

Backport to the trunk will follow.

Refactor the urllib open_http() and open_https() to use the same
internal logic.  The only difference between the two is the call to
constructor an HTTPConnection object.

Change various socket APIs to use bytes or str instead str8.

The two clients in the std library have been updated to use the newer
interface.  A couple of minor changes to the httplib tests were
needed.

Also, reformat some long lines in the httplib tests.

Note that it's hard to test xmlrpclib, because the server it attempts
to connect to doesn't seem to support the expected interfaces.  Many
of the links via xmlrpc.com are dead, so I couldn't find another
server to use for tests.

Even if it isn't, it shouldn't be changed based on the locale setting IMO.
(Incidentally, this makes test_csv.py pass again on my box.)

 - fix "h[sbseek] == b'\1'" and "ratecode = ord(h[sbseek+4])" in test_voc()
 - avoid division by zero
 - use startswith method: replace h[:2] == b'BM' by h.startswith(b'BM')
 - use aifc.open() instead of old aifc.openfp()
 - use ord(b'P') instead of ord('P')