summaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
...
* Backport:Neal Norwitz2003-06-291-0/+6
| | | | | | Fix SF #762455, segfault when sys.stdout is changed in getattr Note: in 2.2, the problem was an infinite loop (at least for me).
* backport of fix from 1.15 to make sure installed scripts have the read bitSkip Montanaro2003-06-271-1/+1
| | | | set.
* Don't assume the name of the table of contents will be contents.html;Fred Drake2003-06-271-2/+4
| | | | | | | | | that won't be right when mkhowto is called with --numeric (as it is for the Python Tutorial). Save the actual name we use when we build the table of contents, then use that in the page head metadata. The node2label.pl script will munge this properly when --numeric isn't given, so this works in both cases. Closes SF bug #761830.
* Fix for SF bug 620190: getargspec() doesn't understand methods.Jeremy Hylton2003-06-271-0/+2
|
* Add definitions of the test and testlist symbols in the grammar soFred Drake2003-06-271-0/+5
| | | | | there are no gaps in the definitions. Closes SF bug #726150.
* backport -uall,-resource stuffSkip Montanaro2003-06-211-25/+92
| | | | add sunos5 and hpux11 expected skips
* check getservbyname with several protocols, not just telnetSkip Montanaro2003-06-212-4/+12
|
* Bump version info.Fred Drake2003-06-202-3/+3
|
* Remove heading from orphan section.Fred Drake2003-06-201-1/+1
|
* Update link to Python book information.Fred Drake2003-06-201-1/+1
|
* Avoid 'dir' as a variable name. Use os.path.join() to join path elementsSkip Montanaro2003-06-191-4/+4
| | | | together.
* Include "instances of most classes" in a warning about mutable objectsFred Drake2003-06-181-2/+3
| | | | as default values of function/method parameters.
* Patch #755987 (bug #755031): backport from trunk.Greg Ward2003-06-181-16/+14
|
* Backport 1.194:Neal Norwitz2003-06-161-0/+16
| | | | Add test for bug #751998.
* Backport 2.237 by Guido:Neal Norwitz2003-06-161-5/+14
| | | | | - SF patch 751998 fixes an unwanted side effect of the previous fix for SF bug 742860 (the next item).
* Backport FreeBSD recursion limit. Fixes #553736.Martin v. Löwis2003-06-141-0/+14
|
* Treat empty dat/dir files as dumbdbm. Fixes #744687.Martin v. Löwis2003-06-141-3/+7
|
* copy and deepcopy builtin functions atomically. Fixes #746304.Martin v. Löwis2003-06-142-0/+8
| | | | Also backport 2.36 (deepcopy classes atomically).
* Patch #746366: Update to current automake install-sh.Martin v. Löwis2003-06-141-62/+105
|
* Patch #754340: Fix typo in redirect_request.Martin v. Löwis2003-06-141-1/+1
|
* Install Sean's updated specBarry Warsaw2003-06-131-0/+2
|
* Backport of my checkin to rev 1.46:Thomas Heller2003-06-121-1/+1
| | | | | | | | Fix for sf # 749210, wininst isn't build correctly after building zip. The problem was that subcommands were not reinitialized. Bugfix candidate, will backport myself.
* Clarify docstring for symlink.Brett Cannon2003-06-111-1/+1
|
* Patch #750595: Refer to type complex using builtin. Fixes #595837.Martin v. Löwis2003-06-071-3/+9
|
* Patch #746801: FreeBSD 4 expected failures, by Charles Swiger.Martin v. Löwis2003-06-071-0/+23
|
* Patch #749191: Delete commands in after_cancel.Martin v. Löwis2003-06-071-0/+6
|
* Bump the version numberBarry Warsaw2003-05-311-1/+1
|
* This commit was manufactured by cvs2svn to create tag 'r223'.v2.2.3cvs2svn2003-05-301-0/+1
|
* Last minute NEWS file updatesBarry Warsaw2003-05-301-1/+4
|
* Update Windows installer for 2.2.3.Tim Peters2003-05-301-4/+4
|
* Update the Windows DLL resource info for 2.2.3.Tim Peters2003-05-301-5/+5
|
* Update the buildno diplayed by Windows Python 2.2.3.Tim Peters2003-05-301-2/+2
|
* Updated Windows buildno for 2.2.3 final.Tim Peters2003-05-301-0/+2
|
* Bump the version to 2.2.3 (final).Guido van Rossum2003-05-301-3/+3
|
* Backport from 2.3 trunk:Barry Warsaw2003-05-292-3/+7
| | | | | | | | _make_boundary(): Fix for SF bug #745478, broken boundary calculation in some locales. This code simplifies the boundary algorithm to use randint() which is what we wanted anyway. Bump package version to 2.5.3.
* Typo repair.Tim Peters2003-05-291-2/+2
|
* Fleshed out the WeakKeyDictionary.__delitem__ news, since this can changeTim Peters2003-05-291-2/+14
| | | | visible behavior.
* All the news that's fit to print (I reviewed the CVS log since theGuido van Rossum2003-05-291-0/+15
| | | | bump of patchlevel.h).
* Add section for 2.2.3 final and memory leak plug for tkinterNeal Norwitz2003-05-291-0/+7
|
* Backport 1.160:Neal Norwitz2003-05-291-4/+11
| | | | | | Fix SF #745055, Memory leak in _tkinter.c/Tkapp_SplitList() Also fix a memory leak in Tkapp_Split.
* Give the Windows socket wrapper function default values for its 'family'Tim Peters2003-05-291-1/+1
| | | | | | | | and 'type' arguments. This improves x-platform compatibility (e.g., on Unix systems socket() already supported these default arguments). No need to do anything here for 2.3 -- it already fixed this glitch in a different way.
* Set release information for Python 2.2.3.Fred Drake2003-05-292-3/+3
|
* crash_in_get_sf736892(): A test of SF 736892, forcing function to actBarry Warsaw2003-05-291-0/+14
| | | | like an unbound method dumps core.
* Backport patch 2.206:Barry Warsaw2003-05-291-0/+9
| | | | | | | | | | | | | | | ---------------------------- revision 2.206 date: 2003/02/11 16:25:43; author: gvanrossum; state: Exp; lines: +9 -0 Add basic arg sanity checking to wrap_descr_get(). This is called when Python code calls a descriptor's __get__ method. It should translate None to NULL in both argument positions, and insist that at least one of the argument positions is not NULL after this transformation. ---------------------------- which fixes SF bug # 736892, forcing function to act like an unbound method dumps core.
* Remove debug code from weakref_segfault().Guido van Rossum2003-05-291-3/+1
| | | | | Also move the call of the test to the top; the crash is caused by a double free.
* Fix for SF 742911. We now clear the weakrefs *before* calling __del__Guido van Rossum2003-05-291-8/+8
| | | | or emptying __dict__, just as we do for classic classes.
* Add testcase for SF 742911.Guido van Rossum2003-05-291-0/+24
|
* The backport gets Fred's seal of approval:Barry Warsaw2003-05-282-5/+59
| | | | | | | | | | | | | | | | | | | SF 742860: WeakKeyDictionary __delitem__ uses iterkeys Someone review this, please! Final releases are getting close, Fred (the weakref guy) won't be around until Tuesday, and the pre-patch code can indeed raise spurious RuntimeErrors in the presence of threads or mutating comparison functions. See the bug report for my confusions: I can't see any reason for why __delitem__ iterated over the keys. The new one-liner implementation is much faster, can't raise RuntimeError, and should be better-behaved in all respects wrt threads. New tests test_weak_keyed_bad_delitem and test_weak_keyed_cascading_deletes fail before this patch. Backported the tests and the patch.
* Spec file update from Sean Reifschneider.Barry Warsaw2003-05-271-3/+2
|
* Backport of 1.48: on Mac OS X don't use -R for runtime_library_dirs, useJack Jansen2003-05-232-0/+6
| | | | -L in stead. Fixes #723495.
generated by cgit v0.12 at 2025-08-09 19:38:20 (GMT)