Similar to an old bug in ntpath.py.

splittype():  Always lower-case the URL scheme; these are supposed to be
               normalized according to RFC 1738 anyway.

getatime() returned the mtime instead of the atime.
Similar to an old bug in posixpath.py.

Correct typo that caused a well-formedness failure.

They should all be binary (cvs admin -kv has now been applied).

the bind(), connect(), and connect_ex() methods must be a single
parameter.

Originally part of 1.98, reverted in 1.100.

changes to the bsddb source.

Update all projects so Debug builds use "/Zi" instead of "/ZI", which means everything should still build under MSVC5

NOTE:  All these projects had messed-up line endings (but MSVC didnt mind).  Executing a diff with "--ignore-all-space" will show the specific changes, but I made a complete checkin with correct line endiings as it is the correct thing to do!

Include limits.h if we have it.

If we have stat.h include it if we don't have sys/stat.h

Include stat.h if needed; different Mac filename compare

CHAR_MAX, use hardcoded -128 and 127.  This may seem strange, unless
you realize that we're talking about signed bytes here!  Bytes are
always 8 bits and 2's complement.  CHAR_MIN and CHAR_MAX are
properties of the char data type, which is guaranteed to hold at least
8 bits anyway.

Otherwise you'd get failing tests on platforms where unsigned char is
the default (e.g. AIX).

Thanks, Vladimir Marangozov, for finding this nit!

what the dozen+ subprojects are for, which are and aren't
expected to build out of the box, and what else is needed to get
them all to build.  Also explained that Alpha configurations
don't refer to pre-beta, but to the Alpha processor!  That's
baffled me for years <0.7 wink>.

Test case for the pty module.

- fixed parser flag propagation (of all stupid bugs...)