]> granicus.if.org Git - python/log
python
18 years agoForward-port of rev. 51857:
Georg Brandl [Mon, 11 Sep 2006 09:38:35 +0000 (09:38 +0000)]
Forward-port of rev. 51857:

Building with HP's cc on HP-UX turned up a couple of problems.
_PyGILState_NoteThreadState was declared as static inconsistently.
Make it static as it's not necessary outside of this module.

Some tests failed because errno was reset to 0. (I think the tests
that failed were at least: test_fcntl and test_mailbox).
Ensure that errno doesn't change after a call to Py_END_ALLOW_THREADS.
This only affected debug builds.

18 years agoAdd a "crasher" taken from the sgml bug report referenced in the comment
Neal Norwitz [Mon, 11 Sep 2006 04:32:57 +0000 (04:32 +0000)]
Add a "crasher" taken from the sgml bug report referenced in the comment

18 years agoProperly handle a NULL returned from PyArena_New().
Neal Norwitz [Mon, 11 Sep 2006 04:28:16 +0000 (04:28 +0000)]
Properly handle a NULL returned from PyArena_New().
(Also fix some whitespace)

Klocwork #364.

18 years agoForward port of 51850 from release25-maint branch.
Neal Norwitz [Mon, 11 Sep 2006 04:24:09 +0000 (04:24 +0000)]
Forward port of 51850 from release25-maint branch.

As mentioned on python-dev, reverting patch #1504333 because it introduced
an infinite loop in rev 47154.

This patch also adds a test to prevent the regression.

18 years agoRemove the __unicode__ method from exceptions. Allows unicode() to be called
Brett Cannon [Sat, 9 Sep 2006 07:11:46 +0000 (07:11 +0000)]
Remove the __unicode__ method from exceptions.  Allows unicode() to be called
on exception classes.  Would require introducing a tp_unicode slot to make it
work otherwise.

Fixes bug #1551432 and will be backported.

18 years agoFix typo in example
Andrew M. Kuchling [Fri, 8 Sep 2006 14:05:10 +0000 (14:05 +0000)]
Fix typo in example

18 years agoUse native SQLite types
Andrew M. Kuchling [Fri, 8 Sep 2006 14:03:01 +0000 (14:03 +0000)]
Use native SQLite types

18 years agoUse native SQLite types
Andrew M. Kuchling [Fri, 8 Sep 2006 14:02:45 +0000 (14:02 +0000)]
Use native SQLite types

18 years agoExplain SQLite a bit more clearly
Andrew M. Kuchling [Fri, 8 Sep 2006 13:36:36 +0000 (13:36 +0000)]
Explain SQLite a bit more clearly

18 years agoExplain SQLite a bit more clearly
Andrew M. Kuchling [Fri, 8 Sep 2006 13:35:49 +0000 (13:35 +0000)]
Explain SQLite a bit more clearly

18 years agoAdd missing word
Andrew M. Kuchling [Fri, 8 Sep 2006 13:25:23 +0000 (13:25 +0000)]
Add missing word

18 years agoAdd missing NEWS entry for rev 51803
Nick Coghlan [Fri, 8 Sep 2006 10:04:38 +0000 (10:04 +0000)]
Add missing NEWS entry for rev 51803

18 years agoThe cast function did not accept c_char_p or c_wchar_p instances
Thomas Heller [Thu, 7 Sep 2006 19:09:54 +0000 (19:09 +0000)]
The cast function did not accept c_char_p or c_wchar_p instances
as first argument, and failed with a 'bad argument to internal function'
error message.

18 years agoAnonymous structure fields that have a bit-width specified did not work,
Thomas Heller [Thu, 7 Sep 2006 18:56:28 +0000 (18:56 +0000)]
Anonymous structure fields that have a bit-width specified did not work,
and they gave a strange error message from PyArg_ParseTuple:
    function takes exactly 2 arguments (3 given).

With tests.

18 years agoAdd a warning notice on top of the generated grammar.txt.
Armin Rigo [Thu, 7 Sep 2006 15:06:00 +0000 (15:06 +0000)]
Add a warning notice on top of the generated grammar.txt.

18 years ago[Bug #1552726] Avoid repeatedly polling in interactive mode -- only put a timeout...
Andrew M. Kuchling [Thu, 7 Sep 2006 13:59:38 +0000 (13:59 +0000)]
[Bug #1552726] Avoid repeatedly polling in interactive mode -- only put a timeout on the select()
if an input hook has been defined.  Patch by Richard Boulton.

This select() code is only executed with readline 2.1, or if
READLINE_CALLBACKS is defined.

Backport candidate for 2.5, 2.4, probably earlier versions too.

18 years agoTypo fix
Andrew M. Kuchling [Thu, 7 Sep 2006 13:56:23 +0000 (13:56 +0000)]
Typo fix

18 years agoFix a glaring error and update some version numbers.
Ronald Oussoren [Thu, 7 Sep 2006 12:03:10 +0000 (12:03 +0000)]
Fix a glaring error and update some version numbers.

18 years agoFix the speed regression in inspect.py by adding another cache to speed up getmodule...
Nick Coghlan [Thu, 7 Sep 2006 10:50:34 +0000 (10:50 +0000)]
Fix the speed regression in inspect.py by adding another cache to speed up getmodule(). Patch #1553314

18 years agoFix refcounts and add error checks.
Raymond Hettinger [Thu, 7 Sep 2006 02:42:48 +0000 (02:42 +0000)]
Fix refcounts and add error checks.

18 years agoFixed subprocess bug #1531862 again, after removing tests
Gustavo Niemeyer [Thu, 7 Sep 2006 00:48:33 +0000 (00:48 +0000)]
Fixed subprocess bug #1531862 again, after removing tests
offending buildbot

18 years agoAdd news item for bug fix of SF bug report #1546372.
Marc-André Lemburg [Wed, 6 Sep 2006 20:40:22 +0000 (20:40 +0000)]
Add news item for bug fix of SF bug report #1546372.

18 years agoFix missing import of the types module in logging.config.
Georg Brandl [Wed, 6 Sep 2006 20:05:58 +0000 (20:05 +0000)]
Fix missing import of the types module in logging.config.

18 years agoBug #1542051: Exceptions now correctly call PyObject_GC_UnTrack.
Georg Brandl [Wed, 6 Sep 2006 06:50:05 +0000 (06:50 +0000)]
Bug #1542051: Exceptions now correctly call PyObject_GC_UnTrack.
Also make sure that every exception class has __module__ set to
'exceptions'.

18 years agowith and as are now keywords. There are some generated files I can't recreate.
Neal Norwitz [Wed, 6 Sep 2006 06:28:06 +0000 (06:28 +0000)]
with and as are now keywords.  There are some generated files I can't recreate.

18 years agoBug #1550983: emit better error messages for erroneous relative
Georg Brandl [Wed, 6 Sep 2006 06:09:31 +0000 (06:09 +0000)]
Bug #1550983: emit better error messages for erroneous relative
imports (if not in package and if beyond toplevel package).

18 years agoBug #1551427: fix a wrong NULL pointer check in the win32 version
Georg Brandl [Wed, 6 Sep 2006 06:03:59 +0000 (06:03 +0000)]
Bug #1551427: fix a wrong NULL pointer check in the win32 version
of os.urandom().

18 years agoRevert 51758 because it broke all the buildbots
Neal Norwitz [Wed, 6 Sep 2006 03:58:34 +0000 (03:58 +0000)]
Revert 51758 because it broke all the buildbots

18 years agoFixing #1531862: Do not close standard file descriptors in the
Gustavo Niemeyer [Wed, 6 Sep 2006 01:58:52 +0000 (01:58 +0000)]
Fixing #1531862: Do not close standard file descriptors in the
subprocess module.

18 years agoUpdate the PCBuild8 solution.
Kristján Valur Jónsson [Tue, 5 Sep 2006 17:58:12 +0000 (17:58 +0000)]
Update the PCBuild8 solution.
Facilitate cross-compilation by having binaries in separate Win32 and x64 directories.
Rationalized configs by making proper use of platforms/configurations.
Remove pythoncore_pgo project.
Add new PGIRelease and PGORelease configurations to perform Profile Guided Optimisation.
Removed I64 support, but this can be easily added by copying the x64 platform settings.

18 years ago[Bug #1526834] Fix crash in pdb when you do 'b f(';
Andrew M. Kuchling [Tue, 5 Sep 2006 13:19:18 +0000 (13:19 +0000)]
[Bug #1526834] Fix crash in pdb when you do 'b f(';
the function name was placed into a regex pattern and the unbalanced paren
caused re.compile() to report an error

18 years ago[Bug #1525469] SimpleXMLRPCServer still uses the sys.exc_{value,type} module-level...
Andrew M. Kuchling [Tue, 5 Sep 2006 13:15:41 +0000 (13:15 +0000)]
[Bug #1525469] SimpleXMLRPCServer still uses the sys.exc_{value,type} module-level globals instead of calling sys.exc_info().  Reported by Russell Warren

18 years agoRearrange example a bit, and show rpartition() when separator is not found
Andrew M. Kuchling [Tue, 5 Sep 2006 13:02:40 +0000 (13:02 +0000)]
Rearrange example a bit, and show rpartition() when separator is not found

18 years agoBug #1552618: change docs of dict.has_key() to reflect recommendation
Georg Brandl [Tue, 5 Sep 2006 12:44:58 +0000 (12:44 +0000)]
Bug #1552618: change docs of dict.has_key() to reflect recommendation
to use "in".

18 years agoFix a typo: 2013 -> 0213
Hye-Shik Chang [Tue, 5 Sep 2006 12:14:57 +0000 (12:14 +0000)]
Fix a typo: 2013 -> 0213

18 years agoFix a few bugs on cjkcodecs found by Oren Tirosh:
Hye-Shik Chang [Tue, 5 Sep 2006 12:07:09 +0000 (12:07 +0000)]
Fix a few bugs on cjkcodecs found by Oren Tirosh:
- 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 2013:2
  codepoints now.

18 years agoM-x untabify
Neal Norwitz [Tue, 5 Sep 2006 04:00:12 +0000 (04:00 +0000)]
M-x untabify

18 years agoAdd a comment about some refactoring. (There's probably more that should be done...
Neal Norwitz [Tue, 5 Sep 2006 03:58:26 +0000 (03:58 +0000)]
Add a comment about some refactoring.  (There's probably more that should be done.)  I will reformat this file in the next checkin due to the inconsistent tabs/spaces.

18 years agoBug #1520864 (again): unpacking singleton tuples in list comprehensions and
Neal Norwitz [Tue, 5 Sep 2006 03:53:08 +0000 (03:53 +0000)]
Bug #1520864 (again): unpacking singleton tuples in list comprehensions and
generator expressions (x for x, in ... ) works again.

Sigh, I only fixed for loops the first time, not list comps and genexprs too.
I couldn't find any more unpacking cases where there is a similar bug lurking.

This code should be refactored to eliminate the duplication.  I'm sure
the listcomp/genexpr code can be refactored.  I'm not sure if the for loop
can re-use any of the same code though.

Will backport to 2.5 (the only place it matters).

18 years agoPatch #1540470, for OpenBSD 4.0. Backport candidate for 2.[34].
Neal Norwitz [Tue, 5 Sep 2006 02:57:01 +0000 (02:57 +0000)]
Patch #1540470, for OpenBSD 4.0.  Backport candidate for 2.[34].

18 years agoAdd a NEWS entry for str.rpartition() change
Neal Norwitz [Tue, 5 Sep 2006 02:36:20 +0000 (02:36 +0000)]
Add a NEWS entry for str.rpartition() change

18 years agoThis was found by Guido AFAIK on p3yk (sic) branch.
Neal Norwitz [Tue, 5 Sep 2006 02:35:08 +0000 (02:35 +0000)]
This was found by Guido AFAIK on p3yk (sic) branch.

18 years agoFix SF #1552093, eval docstring typo (3 ps in mapping)
Neal Norwitz [Tue, 5 Sep 2006 02:25:41 +0000 (02:25 +0000)]
Fix SF #1552093, eval docstring typo (3 ps in mapping)

18 years agoFix SF bug #1546288, crash in dict_equal.
Neal Norwitz [Tue, 5 Sep 2006 02:24:03 +0000 (02:24 +0000)]
Fix SF bug #1546288, crash in dict_equal.

18 years agoAdd missing svn:eol-style property to text files.
Tim Peters [Tue, 5 Sep 2006 02:22:17 +0000 (02:22 +0000)]
Add missing svn:eol-style property to text files.

18 years agoWhitespace normalization.
Tim Peters [Tue, 5 Sep 2006 02:21:19 +0000 (02:21 +0000)]
Whitespace normalization.

18 years ago"Conceptual" merge of rev 51711 from the 2.5 branch.
Tim Peters [Tue, 5 Sep 2006 02:18:09 +0000 (02:18 +0000)]
"Conceptual" merge of 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.

This differs from the corresponding change on the 2.5 and 2.4
branches, using a less obscure approach, but one that /may/
tickle platform idiocies in their definitions of LONG_MIN.
The 2.4 + 2.5 change avoided introducing a dependence on
LONG_MIN, at the cost of substantially goofier code.

18 years agoFix endcase for str.rpartition()
Raymond Hettinger [Mon, 4 Sep 2006 15:32:48 +0000 (15:32 +0000)]
Fix endcase for str.rpartition()

18 years agoNEWS entry on trunk for decimal module changes
Nick Coghlan [Sun, 3 Sep 2006 01:20:46 +0000 (01:20 +0000)]
NEWS entry on trunk for decimal module changes

18 years agoTypo fix for decimal docs
Nick Coghlan [Sun, 3 Sep 2006 01:06:07 +0000 (01:06 +0000)]
Typo fix for decimal docs

18 years agoFix final documentation nits before backporting decimal module fixes to 2.5
Nick Coghlan [Sun, 3 Sep 2006 01:02:00 +0000 (01:02 +0000)]
Fix final documentation nits before backporting decimal module fixes to 2.5

18 years agoHmm, this test has failed at least twice recently on the OpenBSD and
Neal Norwitz [Sat, 2 Sep 2006 19:40:19 +0000 (19:40 +0000)]
Hmm, this test has failed at least twice recently on the OpenBSD and
Debian sparc buildbots.  Since this goes through a lot of tests
and hits the disk a lot it could be slow (especially if NFS is involved).
I'm not sure if that's the problem, but printing periodic msgs shouldn't hurt.
The code was stolen from test_compiler.

18 years agoAdd missing word in comment
Neal Norwitz [Sat, 2 Sep 2006 18:51:34 +0000 (18:51 +0000)]
Add missing word in comment

18 years agoFix documentation nits for decimal context managers.
Raymond Hettinger [Sat, 2 Sep 2006 17:07:23 +0000 (17:07 +0000)]
Fix documentation nits for decimal context managers.

18 years agoFurther corrections to the decimal module context management documentation
Nick Coghlan [Sat, 2 Sep 2006 04:04:18 +0000 (04:04 +0000)]
Further corrections to the decimal module context management documentation

18 years agoMake decimal.ContextManager a private implementation detail of decimal.localcontext()
Nick Coghlan [Sat, 2 Sep 2006 03:54:17 +0000 (03:54 +0000)]
Make decimal.ContextManager a private implementation detail of decimal.localcontext()

18 years agoBug #1550714: fix SystemError from itertools.tee on negative value for n.
Neal Norwitz [Sat, 2 Sep 2006 02:58:13 +0000 (02:58 +0000)]
Bug #1550714: fix SystemError from itertools.tee on negative value for n.

Needs backport to 2.5.1 and earlier.

18 years agoBug #1548092: fix curses.tparm seg fault on invalid input. Needs backport to 2.5...
Neal Norwitz [Sat, 2 Sep 2006 02:50:35 +0000 (02:50 +0000)]
Bug #1548092: fix curses.tparm seg fault on invalid input.  Needs backport to 2.5.1 and earlier.

18 years agoSF #1547931, fix typo (missing and). Will backport to 2.5
Neal Norwitz [Sat, 2 Sep 2006 02:43:17 +0000 (02:43 +0000)]
SF #1547931, fix typo (missing and).  Will backport to 2.5

18 years agoevalfile() should be execfile().
Georg Brandl [Fri, 1 Sep 2006 22:30:52 +0000 (22:30 +0000)]
evalfile() should be execfile().

18 years ago- SF patch #1550263: Enhance and correct unittest docs
Fred Drake [Fri, 1 Sep 2006 03:57:19 +0000 (03:57 +0000)]
- SF patch #1550263: Enhance and correct unittest docs
- various minor cleanups for improved consistency

18 years agoHave pre-existing C files use 8 spaces indents (to match old PEP 7 style), but
Brett Cannon [Thu, 31 Aug 2006 22:42:37 +0000 (22:42 +0000)]
Have pre-existing C files use 8 spaces indents (to match old PEP 7 style), but
have all new files use 4 spaces (to match current PEP 7 style).

18 years agoFix comment about indentation level in C files.
Brett Cannon [Thu, 31 Aug 2006 21:47:52 +0000 (21:47 +0000)]
Fix comment about indentation level in C files.

18 years agoMake sure memory is properly cleaned up in file_init.
Brett Cannon [Thu, 31 Aug 2006 18:54:26 +0000 (18:54 +0000)]
Make sure memory is properly cleaned up in file_init.

Backport candidate.

18 years agoRemove the old decimal context management tests from test_contextlib (guess who didn...
Nick Coghlan [Thu, 31 Aug 2006 12:51:25 +0000 (12:51 +0000)]
Remove the old decimal context management tests from test_contextlib (guess who didn't run the test suite before committing...)

18 years agoFix the wrongheaded implementation of context management in the decimal module and...
Nick Coghlan [Thu, 31 Aug 2006 12:00:43 +0000 (12:00 +0000)]
Fix the wrongheaded implementation of context management in the decimal module and add unit tests. (python-dev discussion is ongoing regarding what we do about Python 2.5)

18 years agoDoc fix: hashlib objects don't always return a digest of 16 bytes.
Armin Rigo [Thu, 31 Aug 2006 08:51:06 +0000 (08:51 +0000)]
Doc fix: hashlib objects don't always return a digest of 16 bytes.
Backport candidate for 2.5.

18 years agoFix a buglet in the error reporting (SF bug report #1546372).
Marc-André Lemburg [Tue, 29 Aug 2006 10:34:12 +0000 (10:34 +0000)]
Fix a buglet in the error reporting (SF bug report #1546372).

This should probably go into Python 2.5 or 2.5.1 as well.

18 years agoFix a couple of typos.
Neal Norwitz [Tue, 29 Aug 2006 05:40:58 +0000 (05:40 +0000)]
Fix a couple of typos.

18 years ago- Move functions common to all path modules into genericpath.py and have the
Jack Diederich [Sat, 26 Aug 2006 18:42:06 +0000 (18:42 +0000)]
- Move functions common to all path modules into genericpath.py and have the
  OS speicifc path modules import them.
- Have os2emxpath import common functions fron ntpath instead of using copies

18 years agoA new test here relied on preserving invisible trailing
Tim Peters [Fri, 25 Aug 2006 22:26:21 +0000 (22:26 +0000)]
A new test here relied on preserving invisible trailing
whitespace in expected output.  Stop that.

18 years agoAdd missing svn:eol-style property to text files.
Tim Peters [Fri, 25 Aug 2006 22:06:44 +0000 (22:06 +0000)]
Add missing svn:eol-style property to text files.

18 years agoWhitespace normalization.
Tim Peters [Fri, 25 Aug 2006 22:05:39 +0000 (22:05 +0000)]
Whitespace normalization.

18 years agoThe regular expression engine in '_sre' can segfault when interpreting
Armin Rigo [Fri, 25 Aug 2006 12:44:28 +0000 (12:44 +0000)]
The regular expression engine in '_sre' can segfault when interpreting
bogus bytecode.  It is unclear whether this is a real bug or a "won't
fix" case like bogus_code_obj.py.

18 years agoBuild _ctypes.pyd for win AMD64 into the MSVC project file.
Thomas Heller [Fri, 25 Aug 2006 09:26:33 +0000 (09:26 +0000)]
Build _ctypes.pyd for win AMD64 into the MSVC project file.
Since MSVC doesn't know about .asm files, a helper batch file is needed
to find ml64.exe in predefined locations.  The helper script hardcodes
the path to the MS Platform SDK.

18 years agoAdd missing file for _ctypes.pyd port to win64 on AMD64.
Thomas Heller [Fri, 25 Aug 2006 07:34:51 +0000 (07:34 +0000)]
Add missing file for _ctypes.pyd port to win64 on AMD64.

18 years agoPort _ctypes.pyd to win64 on AMD64.
Thomas Heller [Fri, 25 Aug 2006 07:27:33 +0000 (07:27 +0000)]
Port _ctypes.pyd to win64 on AMD64.

18 years agoimporting types is not necessary if we use isinstance
Neal Norwitz [Fri, 25 Aug 2006 01:52:49 +0000 (01:52 +0000)]
importing types is not necessary if we use isinstance

18 years agoPatch #1545507: Exclude ctypes package in Win64 MSI file.
Martin v. Löwis [Fri, 25 Aug 2006 00:03:34 +0000 (00:03 +0000)]
Patch #1545507: Exclude ctypes package in Win64 MSI file.
Will backport to 2.5.

18 years agoFix SF bug #1545837: array.array borks on deepcopy.
Thomas Wouters [Thu, 24 Aug 2006 18:40:20 +0000 (18:40 +0000)]
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.)

18 years agoReverting the patch that tried to fix the issue whereby x**2 raises
Alex Martelli [Wed, 23 Aug 2006 22:17:59 +0000 (22:17 +0000)]
Reverting the patch that tried to fix the issue whereby x**2 raises
OverflowError while x*x succeeds and produces infinity; apparently
these inconsistencies cannot be fixed across ``all'' platforms and
there's a widespread feeling that therefore ``every'' platform
should keep suffering forevermore.  Ah well.

18 years agoReplace dead code with an assert.
Jeremy Hylton [Wed, 23 Aug 2006 21:26:46 +0000 (21:26 +0000)]
Replace dead code with an assert.

Now that COMMENT tokens are reliably followed by NL or NEWLINE,
there is never a need to add extra newlines in untokenize.

18 years agoBug fixes large and small for tokenize.
Jeremy Hylton [Wed, 23 Aug 2006 21:14:03 +0000 (21:14 +0000)]
Bug fixes large and small for tokenize.

Small: Always generate a NL or NEWLINE token following
       a COMMENT token.  The old code did not generate an NL token if
       the comment was on a line by itself.

Large: The output of untokenize() will now match the
       input exactly if it is passed the full token sequence.  The
       old, crufty output is still generated if a limited input
       sequence is provided, where limited means that it does not
       include position information for tokens.

Remaining bug: There is no CONTINUATION token (\) so there is no way
for untokenize() to handle such code.

Also, expanded the number of doctests in hopes of eventually removing
the old-style tests that compare against a golden file.

Bug fix candidate for Python 2.5.1. (Sigh.)

18 years agox**2 should about equal x*x (including for a float x such that the result is
Alex Martelli [Wed, 23 Aug 2006 20:42:02 +0000 (20:42 +0000)]
x**2 should about equal x*x (including for a float x such that the result is
inf) but didn't; added a test to test_float to verify that, and ignored the
ERANGE value for errno in the pow operation to make the new test pass (with
help from Marilyn Davis at the Google Python Sprint -- thanks!).

18 years agoBaby steps towards better tests for tokenize
Jeremy Hylton [Wed, 23 Aug 2006 18:37:43 +0000 (18:37 +0000)]
Baby steps towards better tests for tokenize

18 years agoMove functional howto into trunk
Andrew M. Kuchling [Tue, 22 Aug 2006 23:13:43 +0000 (23:13 +0000)]
Move functional howto into trunk

18 years agoExpose column offset information in parse trees.
Jeremy Hylton [Tue, 22 Aug 2006 20:46:00 +0000 (20:46 +0000)]
Expose column offset information in parse trees.

18 years agoAlexander Belopolsky pointed out that pos is a size_t
Neal Norwitz [Tue, 22 Aug 2006 13:56:56 +0000 (13:56 +0000)]
Alexander Belopolsky pointed out that pos is a size_t

18 years agopatch for documentation for recent uuid changes (from ping)
Anthony Baxter [Tue, 22 Aug 2006 07:36:59 +0000 (07:36 +0000)]
patch for documentation for recent uuid changes (from ping)

18 years agoTry to get the windows bots working again with the new peephole.c
Neal Norwitz [Mon, 21 Aug 2006 23:44:48 +0000 (23:44 +0000)]
Try to get the windows bots working again with the new peephole.c

18 years agoPatch #1541585: fix buffer overrun when performing repr() on
Neal Norwitz [Mon, 21 Aug 2006 22:21:19 +0000 (22:21 +0000)]
Patch #1541585: fix buffer overrun when performing repr() on
a unicode string in a build with wide unicode (UCS-4) support.

This code could be improved, so add an XXX comment.

18 years agoHandle a few more error conditions.
Neal Norwitz [Mon, 21 Aug 2006 20:16:24 +0000 (20:16 +0000)]
Handle a few more error conditions.

Klocwork 301 and 302.  Will backport.

18 years agoPatch #1542451: disallow continue anywhere under a finally
Neal Norwitz [Mon, 21 Aug 2006 19:47:08 +0000 (19:47 +0000)]
Patch #1542451: disallow continue anywhere under a finally

I'm undecided if this should be backported to 2.5 or 2.5.1.
Armin suggested to wait (I'm of the same opinion).  Thomas W thinks
it's fine to go in 2.5.

18 years agoFix a couple of ssize-t issues reported by Alexander Belopolsky on python-dev
Neal Norwitz [Mon, 21 Aug 2006 18:20:10 +0000 (18:20 +0000)]
Fix a couple of ssize-t issues reported by Alexander Belopolsky on python-dev

18 years agoAdd assert to make Klocwork happy (#276)
Neal Norwitz [Mon, 21 Aug 2006 18:01:30 +0000 (18:01 +0000)]
Add assert to make Klocwork happy (#276)

18 years agoFix bug #1543303, tarfile adds padding that breaks gunzip.
Neal Norwitz [Mon, 21 Aug 2006 17:59:46 +0000 (17:59 +0000)]
Fix bug #1543303, tarfile adds padding that breaks gunzip.
Patch # 1543897.

Will backport to 2.5

18 years agoMove peephole optimizer to separate file. (Forgot .h in previous checkin.)
Jeremy Hylton [Mon, 21 Aug 2006 16:20:29 +0000 (16:20 +0000)]
Move peephole optimizer to separate file.  (Forgot .h in previous checkin.)

18 years agoMove peephole optimizer to separate file.
Jeremy Hylton [Mon, 21 Aug 2006 16:19:37 +0000 (16:19 +0000)]
Move peephole optimizer to separate file.

18 years agoPatch #1542948: fix urllib2 header casing issue. With new test.
Georg Brandl [Sun, 20 Aug 2006 13:15:39 +0000 (13:15 +0000)]
Patch #1542948: fix urllib2 header casing issue. With new test.

18 years agoHandle PyString_FromInternedString() failing (unlikely, but possible).
Neal Norwitz [Sat, 19 Aug 2006 04:52:03 +0000 (04:52 +0000)]
Handle PyString_FromInternedString() failing (unlikely, but possible).

Klocwork #325

(I'm not backporting this, but if someone wants to, feel free.)