Since I can't test this, I'm just adding a comment. If we get access
to AIX boxes, we can test this and really resolve. Anyone from IBM
want to offer help?
[ 1232517 ] OverflowError in time.utime() causes strange traceback
A needed error check was missing.
(Actually, this error check may only have become necessary in fairly
recent Python, not sure).
Backport candidate.
[A few lines below the code in 2.4 touched by the patch, there's already
a similar check of (intval == -1 && PyErr_Occurred()), so I think
this function can already report such errors, and therefore the fix
still applies. Perhaps Michael can clarify what he was referring to. --amk]
[Possibly controversial because it adds a parameter to a method.
This parameter isn't documented, however, so arguably it's a private
detail, and the fix is necessary to make GzipFile.flush() behave more
similarly to regular file object.]
Patch #1110248: SYNC_FLUSH the zlib buffer for GZipFile.flush.
Partially fixes #1110242.
Fixed a few bugs on cjkcodecs:
- gbk and gb18030 codec now handle U+30FB KATAKANA MIDDLE DOT correctly.
- iso2022_jp_2 codec now encodes into G0 for KS X 1001, GB2312
codepoints to conform the standard.
- iso2022_jp_3 and iso2022_jp_2004 codec can encode JIS X 0213:2
codepoints now.
Tim Peters [Tue, 5 Sep 2006 02:00:47 +0000 (02:00 +0000)]
Merge rev 51711 from the 2.5 branch.
i_divmod(): As discussed on Python-Dev, changed the overflow
checking to live happily with recent gcc optimizations that
assume signed integer arithmetic never overflows.
Thomas Wouters [Thu, 24 Aug 2006 18:55:01 +0000 (18:55 +0000)]
Backport trunk's revision 51565:
Fix SF bug #1545837: array.array borks on deepcopy.
array.__deepcopy__() needs to take an argument, even if it doesn't actually
use it. Will backport to 2.5 and 2.4 (if applicable.)
Georg Brandl [Fri, 18 Aug 2006 07:25:22 +0000 (07:25 +0000)]
Bug #1541682: Fix example in the "Refcount details" API docs.
Additionally, remove a faulty example showing PySequence_SetItem applied
to a newly created list object and add notes that this isn't a good idea.
(backport)
Martin v. Löwis [Sun, 30 Jul 2006 13:34:47 +0000 (13:34 +0000)]
Backport r50856:
Don't kill a normal instance of python running on windows when checking
to kill a cygwin instance. build\\python.exe was matching a normal
windows instance. Prefix that with a \\ to ensure build is a directory
and not PCbuild. As discussed on python-dev.
Barry Warsaw [Fri, 28 Jul 2006 03:18:56 +0000 (03:18 +0000)]
Backport r50840 to Python 2.4:
Forward port some fixes that were in email 2.5 but for some reason didn't
make it into email 4.0. Specifically, in Message.get_content_charset(),
handle RFC 2231 headers that contain an encoding not known to Python, or a
character in the data that isn't in the charset encoding. Also forward
port the appropriate unit tests.
Tim Peters [Thu, 27 Jul 2006 23:56:12 +0000 (23:56 +0000)]
Merge rev 50882 from the trunk.
Bug #1529297: The rewrite of doctest for Python 2.4 unintentionally
lost that tests are sorted by name before being run. ``DocTestFinder``
has been changed to sort the list of tests it returns.
[Bug #1471938] Fix build problem on Solaris 8 by conditionalizing the use of mvwgetnstr(); it was conditionalized a few lines below. Fix from Paul Eggert. I also tried out the STRICT_SYSV_CURSES case and am therefore removing the 'untested' comment.
Fred Drake [Thu, 29 Jun 2006 00:52:32 +0000 (00:52 +0000)]
SF bug #1504333: sgmlib should allow angle brackets in quoted values
(modified patch by Sam Ruby; changed to use separate REs for start and end
tags to reduce matching cost for end tags; extended tests; updated to avoid
breaking previous changes to support IPv6 addresses in unquoted attribute
values)
Neal Norwitz [Tue, 27 Jun 2006 04:32:20 +0000 (04:32 +0000)]
Use a different install dir than 2.5. Otherwise running the 2.4 version at the same time as the 2.5 versions would break the 2.5 tests since the install dir was removed.
Brett Cannon [Tue, 13 Jun 2006 21:50:24 +0000 (21:50 +0000)]
Classic class that defined ``def __coerce__(self, other): return other, self``
would infinitely recourse and segfault the interpreter. Now a recursion check
occurs after a coercion.
Tim Peters [Sat, 10 Jun 2006 22:54:19 +0000 (22:54 +0000)]
Merge rev 46840 from trunk.
shuffle() doscstring: Removed warning about sequence length
versus generator period. While this was a real weakness of the
older WH generator for lists with just a few dozen elements,
and so could potentially bite the naive ;-), the Twister should
show excellent behavior up to at least 600 elements.
Fred Drake [Sat, 10 Jun 2006 20:04:06 +0000 (20:04 +0000)]
SF patch #1303595: improve description of __builtins__, explaining how it
varies between __main__ and other modules, and strongly suggest not touching
it but using __builtin__ if absolutely necessary