update.  ;-(

        - Loading non-binary string pickles checks for insecure
          strings. This is needed because cPickle (still)
          uses a restricted eval to parse non-binary string pickles.
          This change is needed to prevent untrusted
          pickles like::

            "S'hello world'*2000000\012p0\012."

          from hosing an application.

        - User-defined types can now support unpickling without
          executing a constructor.

          The second value returned from __reduce__ can now be None,
          rather than an argument tuple. On unpickling, if the second
          value returned from __reduce__ during pickling was None, then
          rather than calling the first value returned from __reduce__,
          directly, the __basicnew__ method of the first value returned
          from __reduce__ is called without arguments.

package, and -lcrypt for the crypt module.

by the Python runtime: 'site' and 'exceptions'.

as in pickle: the new instance is created without calling __init__().

functions and constants together).

Make explicit datadesc sections for each of the constants which might appear,
and have a description of each.  (Descriptions are based on the Linux
documentation and sources and the Solaris man pages.)

Hopefully Jeremy won't mind, because I didn't ask.  ;-)

mapping chapter and section headings to more appropriate nodes so that
the TOC actually works.  Now generates one HTML page per chapter.

Updated the to do list.

I hope this time I've taken out all // comments and indended #ifdefs.

- New option -x, to skip first line of script

- Use the correct platform-specific delimiter and library location in
the usage message

(Also removed two blank lines and moved one line around so that each
part of the usage message is again under 512 bytes and the whole usage
message still fits in 23 lines.)

Also added one more os2 specific piece of code, by Jeff Rush.

not define __getinitargs__, bypass the __init__ constructor
completely.  This uses the trick of instantiating an empty dummy class
and then changing inst.__class__ to the real class.  This is done in
two places: once for the INST and once for the OBJ format code.

Also replaced the much outdated long doc string with a short summary
of the module; the information of that doc string is already
incorporated in the library reference manual.

(Like ntpath.py, this was contributed by "Charles G. Waldman" <cgw@pgt.com>)

exclude "configure".

Use \file{} for file names.

Prefer \code{blat} and \emph{blat} to {\tt blat} and {\em blat}; this matches
current style in the Library Reference a bit better.

Made the example startup banner current.  The version number should be
bumped before the next release.

Remove spurious underscore following book title.  Added specific reference to
a Win32 networking book.

Changed \indexsubitem from (in module SocketServer) to (SocketServer protocol),
since it's talking about a protocol supported by a collection of classes.

Removed the large comment remaining from the template documentation
section; the template tells us to remove these once they're not needed.

Remove some trailing whitespace from VM instruction pseudo-code.

shift.

Adjust whitespace around "=" in VM machine descriptions to be consistent
throughout the document.

Fixed \indexsubitem to gdbm instead of dbm.

Changed dbm to \code{dbm} everywhere.

but otherwise unchanged.

distribution.

enabled.  This is done through a substitution in Modules/Setup.thread(.in).
Bill Janssen will be happy.  The original idea was by Lele Gaifax (though
I changed the implementation to use a separate file).

which is edited by the config.status script depending on whether
--with-thread is specified or not.  Also needs changes to configure(.in).