names in the source code (they already had those for the linker,
through some smart macros; but the source still had the old, un-Py names).

(Also added an XXX comment requesting clarification of module
initialization -- TBD later.)

strings are not actually modified.  Problem reported by Dr. Peter
Stoehr <peter.stoehr@weihenstephan.org>.

the process table.

"""
The message ID is returned lowercased and there is no way to access
the original ID the server sent. Now at least some news servers
are very picky about the case of the ID and return errors when
fetching articles with mixed case given a lowercased version
of the ID.

The solution is simple: remove the string.lower() call.
"""

(I might add that the lowercasing was probably introduced as a result
of sloppy copy-and-paste coding; there's a string.lower in a similar
piece of code a bit higher in the source, that makes more sense --
it's lowercasing the group name.)

(AMK: you should probably copy this into your patch set for pcre.)

After a version by Fredrik Lundh that I found on the web somewhere.
(I like mine better, but of course that may have to do with my alcohol
consumption. :-)

_PyThreadState_Current, defined in pystate.c.

support a macro in pystate.h.

widget -- no time to do this right now.

explore this now.

yours, please let me know for propoer acknowledgement.)

This avoids recompiling files that haven't changed; it adds a -f
option to force recompilation.

- Fixed a bug where a syntax error was reported when a document
  started with white space.  (White space at the start of a document
  is valid if there is no XML declaration.)
- Improved the speed quite a bit for documents that don't make use of
  namespaces.

successfully)

checkin)

assignments.

NOTE: There is still a bug of some sort in the behavior of zlib.  In
at least one case, inflate returns Z_OK (which is typically
interpreted to mean that more output space is needed) when it has
finished inflating a buffer.  This has been reported as a bug to the
zlib maintainers; we may need to change the Python interface.

If we're going to put the PyOS_strtol() proto in intobject.h we should
include it here or the symbol isn't exported on PowerPC.

Here is my current version of xmllib.py and the documentation.  This
version has some API changes with respect to the version currently in
Python (also the one in 1.5.2a).
This version supports XML namespaces.

string we wanted to resize is set to NULL.  Don't Py_DECREF() those
variables!  (5 places)

File names with "funny" characters get translated wrong by
pathname2url (any variety).  E.g. the (Unix) file "/ufs/sjoerd/#tmp"
gets translated into "/ufs/sjoerd/#tmp" which, when interpreted as a
URL is file "/ufs/sjoerd/" with fragment ID "tmp".

Here's an easy fix.  (An alternative fix would be to change the
various implementations of pathname2url and url2pathname to include
calls to quote and unquote.

[The main problem is with the normal use of URLs:
	url = url2pathname(file)
	transmit url
	url, tag = splittag(url)
	urlopen(url)
]

In addition, this patch fixes some uses of unquote:
- the host part of URLs should be unquoted
- the file path in the FTP URL should be unquoted before it is split
  into components.
- because of the latter, I removed all unquoting from ftpwrapper,
  and moved it to the caller, but that is not essential

    offending files to be printed.  Good for emacs `tabnanny.py *.py`

- remove PowerPC specific compiler switch for BeOS (Chris Herborth)
- FreeBSD/3 support
- check for pthread_detach in libc instead of _create

have to use a const-correct prototype on BeOS or the compiler gets uppity.

BeOS headers live in various non-standard places;
luckily, there's an environment variable that lists them all.

This fixes a bug that can cause core dumps when doing seeks in input
StringIO objects. This has a number of other clean-ups.

(This only applies when COUNT_ALLOCS is defined.)