]> granicus.if.org Git - python/log
python
18 years agoUse GetModuleFileNameEx instead of GetProcessImageFileName,
Martin v. Löwis [Mon, 17 Apr 2006 17:26:42 +0000 (17:26 +0000)]
Use GetModuleFileNameEx instead of GetProcessImageFileName,
as the latter is not available on Windows 2000.

18 years agoPatch #790710: Add breakpoint command lists in pdb.
Martin v. Löwis [Mon, 17 Apr 2006 17:08:37 +0000 (17:08 +0000)]
Patch  #790710: Add breakpoint command lists in pdb.

18 years agofix long option markup
George Yoshida [Mon, 17 Apr 2006 15:44:59 +0000 (15:44 +0000)]
fix long option markup

18 years agodisutils checks if MACOSX_DEPLOYMENT_TARGET is consistent with the value at
Ronald Oussoren [Mon, 17 Apr 2006 14:43:30 +0000 (14:43 +0000)]
disutils checks if MACOSX_DEPLOYMENT_TARGET is consistent with the value at
configure time. The current check is too strict and doesn't allow building
extensions that can only run on newer versions of the OS than the version
python was build for, that is python build for 10.3 or later and an extension
for 10.4. This patch relaxes this check.

This turned out to be a reimplementation of patch 1193190.

18 years agoUpdate status of document
Andrew M. Kuchling [Mon, 17 Apr 2006 14:01:36 +0000 (14:01 +0000)]
Update status of document

18 years agoRemove translated code hidden inside a comment environment; latex2html gets confused...
Andrew M. Kuchling [Mon, 17 Apr 2006 14:00:31 +0000 (14:00 +0000)]
Remove translated code hidden inside a comment environment; latex2html gets confused and includes half of it anyway

18 years agoThis patches fixes a number of byteorder problems in MacOSX specific code.
Ronald Oussoren [Mon, 17 Apr 2006 13:40:08 +0000 (13:40 +0000)]
This patches fixes a number of byteorder problems in MacOSX specific code.

18 years agoTeach platform about darwin/x86
Ronald Oussoren [Mon, 17 Apr 2006 13:37:15 +0000 (13:37 +0000)]
Teach platform about darwin/x86

18 years agoRevert to 45478, disable kill_python command for now.
Martin v. Löwis [Mon, 17 Apr 2006 10:39:39 +0000 (10:39 +0000)]
Revert to 45478, disable kill_python command for now.

18 years agoCheck whether disk space is full.
Martin v. Löwis [Mon, 17 Apr 2006 10:36:18 +0000 (10:36 +0000)]
Check whether disk space is full.

18 years agoRename binary again; increase noise; stop trying to actually kill the process.
Martin v. Löwis [Mon, 17 Apr 2006 10:31:35 +0000 (10:31 +0000)]
Rename binary again; increase noise; stop trying to actually kill the process.

18 years agoRename binaries again; increase noise.
Martin v. Löwis [Mon, 17 Apr 2006 10:27:28 +0000 (10:27 +0000)]
Rename binaries again; increase noise.

18 years agoRename binary to avoid conflicts with hanging processes on x86 w2k.
Martin v. Löwis [Mon, 17 Apr 2006 10:23:23 +0000 (10:23 +0000)]
Rename binary to avoid conflicts with hanging processes on x86 w2k.

18 years agoTry some tracing
Martin v. Löwis [Mon, 17 Apr 2006 10:19:25 +0000 (10:19 +0000)]
Try some tracing

18 years agoAdd kill_python command.
Martin v. Löwis [Mon, 17 Apr 2006 09:46:47 +0000 (09:46 +0000)]
Add kill_python command.

18 years agoFix for a bug exposed by r45232:
Armin Rigo [Mon, 17 Apr 2006 09:22:35 +0000 (09:22 +0000)]
Fix for a bug exposed by r45232:

    /path/to/uninstalled/python setup.py build_ext

now failed with pyconfig.h not found.  Prior to r45232
the above command did not look for pyconfig.h, but the
bug is really in the look-up code: expecting to find it
in os.curdir is a rather fragile idea.

18 years agoNo need to cast a Py_ssize_t, use %z in PyErr_Format
Neal Norwitz [Mon, 17 Apr 2006 05:56:32 +0000 (05:56 +0000)]
No need to cast a Py_ssize_t, use %z in PyErr_Format

18 years agoAdd a comment to explain why we are calling _cleanup()
Neal Norwitz [Mon, 17 Apr 2006 02:41:25 +0000 (02:41 +0000)]
Add a comment to explain why we are calling _cleanup()

18 years agoTry to stop the test from leaking and yet still work on windows
Neal Norwitz [Mon, 17 Apr 2006 02:39:37 +0000 (02:39 +0000)]
Try to stop the test from leaking and yet still work on windows

18 years agoRemove unused field
Neal Norwitz [Mon, 17 Apr 2006 01:49:28 +0000 (01:49 +0000)]
Remove unused field

18 years agoAdd some notes about a couple of poorly behaved tests
Neal Norwitz [Mon, 17 Apr 2006 01:49:14 +0000 (01:49 +0000)]
Add some notes about a couple of poorly behaved tests

18 years agotest_cmd_line should not leak any more, ensure an empty reflog file exists if no...
Neal Norwitz [Mon, 17 Apr 2006 01:48:41 +0000 (01:48 +0000)]
test_cmd_line should not leak any more, ensure an empty reflog file exists if no leaks are found

18 years agoGet test to consistently show no leaks
Neal Norwitz [Mon, 17 Apr 2006 01:48:06 +0000 (01:48 +0000)]
Get test to consistently show no leaks

18 years agoNo reason to export get_decomp_record, make static
Neal Norwitz [Mon, 17 Apr 2006 00:36:29 +0000 (00:36 +0000)]
No reason to export get_decomp_record, make static

18 years agomoduleName can be NULL
Neal Norwitz [Mon, 17 Apr 2006 00:33:23 +0000 (00:33 +0000)]
moduleName can be NULL

18 years agoWhitespace normalization.
Tim Peters [Sun, 16 Apr 2006 22:22:36 +0000 (22:22 +0000)]
Whitespace normalization.

18 years agoAdd missing SVN eol-style property to text files.
Tim Peters [Sun, 16 Apr 2006 22:11:28 +0000 (22:11 +0000)]
Add missing SVN eol-style property to text files.

18 years agoUse %zd instead of %i as format character (in call to PyErr_Format) for
Thomas Wouters [Sun, 16 Apr 2006 22:04:49 +0000 (22:04 +0000)]
Use %zd instead of %i as format character (in call to PyErr_Format) for
Py_ssize_t argument.

18 years agoPatch #1063914: Add clipboard_get.
Martin v. Löwis [Sun, 16 Apr 2006 20:55:38 +0000 (20:55 +0000)]
Patch #1063914: Add clipboard_get.

18 years agoAdd item
Andrew M. Kuchling [Sun, 16 Apr 2006 19:53:27 +0000 (19:53 +0000)]
Add item

18 years agoInitialize structseq types only once.
Martin v. Löwis [Sun, 16 Apr 2006 18:55:50 +0000 (18:55 +0000)]
Initialize structseq types only once.

18 years agoUpdate instructions for EXTRA_CFLAGS: configure ignores
Martin v. Löwis [Sun, 16 Apr 2006 18:55:07 +0000 (18:55 +0000)]
Update instructions for EXTRA_CFLAGS: configure ignores
them; they have to be passed to make.

18 years agoDescribe contextlib module. (Done for today...)
Andrew M. Kuchling [Sun, 16 Apr 2006 18:45:11 +0000 (18:45 +0000)]
Describe contextlib module.  (Done for today...)

18 years agoWrite most of the 'writing context managers' section. I'd like comments on it,
Andrew M. Kuchling [Sun, 16 Apr 2006 18:20:05 +0000 (18:20 +0000)]
Write most of the 'writing context managers' section.  I'd like comments on it,
    but wait for a few hours before you read it; I'm still revising it
    and will be tackling contextlib next.
Untabify

18 years agoMake test_timeout not fail on systems with no dots in their fqdn.
Thomas Wouters [Sun, 16 Apr 2006 16:26:28 +0000 (16:26 +0000)]
Make test_timeout not fail on systems with no dots in their fqdn.

18 years agoMake test_warnings play nice with regrtest -R:: now that regrtest doesn't
Thomas Wouters [Sun, 16 Apr 2006 15:43:39 +0000 (15:43 +0000)]
Make test_warnings play nice with regrtest -R:: now that regrtest doesn't
always reload the module (specifically, it doesn't reload if the module has
a 'test_main'.)

18 years agoSpecialcase 'xs4all' (.nl/.net/.com/whatever else we have) as well as
Thomas Wouters [Sun, 16 Apr 2006 15:22:41 +0000 (15:22 +0000)]
Specialcase 'xs4all' (.nl/.net/.com/whatever else we have) as well as
'python.org' when deciding what server to use for the timeout tests; getting
tired of seeing the test fail on all my boxes ;P This'll still allow the
test to fail for hosts in the XS4ALL network that don't have an 'xs4all'
hostname, so maybe it should use a fallback scheme instead.

18 years agoThis test no longer leaks, and test_generators sufficiently tests it to
Thomas Wouters [Sun, 16 Apr 2006 15:11:33 +0000 (15:11 +0000)]
This test no longer leaks, and test_generators sufficiently tests it to
prevent unreported regression.

18 years agoFix valgrind problem with invalid memory read
Neal Norwitz [Sun, 16 Apr 2006 03:37:19 +0000 (03:37 +0000)]
Fix valgrind problem with invalid memory read

18 years agoFix memory leak
Neal Norwitz [Sun, 16 Apr 2006 03:28:17 +0000 (03:28 +0000)]
Fix memory leak

18 years agoerr is no longer used
Neal Norwitz [Sun, 16 Apr 2006 00:02:59 +0000 (00:02 +0000)]
err is no longer used

18 years agoAdd missing DECREF to PyErr_WriteUnraisable(). That function reports
Thomas Wouters [Sat, 15 Apr 2006 23:27:28 +0000 (23:27 +0000)]
Add missing DECREF to PyErr_WriteUnraisable(). That function reports
exceptions that can't be raised any further, because (for instance) they
occur in __del__ methods. The coroutine tests in test_generators was
triggering this leak. Remove the leakers' testcase, and add a simpler
testcase that explicitly tests this leak to test_generators.

test_generators now no longer leaks at all, on my machine. This fix may also
solve other leaks, but my full refleakhunting run is still busy, so who
knows?

18 years agogen_del(): Looks like much this was copy/pasted from
Tim Peters [Sat, 15 Apr 2006 22:59:10 +0000 (22:59 +0000)]
gen_del():  Looks like much this was copy/pasted from
slot_tp_del(), but while the latter had to cater to types
that don't participate in GC, we know that generators do.
That allows strengthing an assert().

18 years agoRemove now-unused variables from tp_traverse and tp_clear methods.
Tim Peters [Sat, 15 Apr 2006 22:51:26 +0000 (22:51 +0000)]
Remove now-unused variables from tp_traverse and tp_clear methods.

18 years agoConsolidate 'leak_test1' and 'refleaks_tests', since they both test for the
Thomas Wouters [Sat, 15 Apr 2006 22:44:07 +0000 (22:44 +0000)]
Consolidate 'leak_test1' and 'refleaks_tests', since they both test for the
same kind of thing.

18 years agoAdd missing PyObject_GC_Track call, causing *some* itertools.tee objects to
Thomas Wouters [Sat, 15 Apr 2006 22:33:13 +0000 (22:33 +0000)]
Add missing PyObject_GC_Track call, causing *some* itertools.tee objects to
not be tracked by GC. This fixes 254 of test_generators' refleaks on my
machine, but I'm sure something else will make them come back :>

Not adding a separate test for this kind of cycle, since the existing
fib/m235 already test them in more extensive ways than any 'minimal' test
has been able to manage.

18 years agoUse Py_VISIT in all tp_traverse methods, instead of traversing manually or
Thomas Wouters [Sat, 15 Apr 2006 21:47:09 +0000 (21:47 +0000)]
Use Py_VISIT in all tp_traverse methods, instead of traversing manually or
using a custom, nearly-identical macro. This probably changes how some of
these functions are compiled, which may result in fractionally slower (or
faster) execution. Considering the nature of traversal, visiting much of the
address space in unpredictable patterns, I'd argue the code readability and
maintainability is well worth it ;P

18 years ago - Whitespace normalization
Thomas Wouters [Sat, 15 Apr 2006 21:41:56 +0000 (21:41 +0000)]
 - Whitespace normalization
 - In functions where we already hold the same object in differently typed
   pointers, use the correctly typed pointer instead of casting the other
   pointer a second time.

18 years agoFix typo.
Thomas Heller [Sat, 15 Apr 2006 20:43:22 +0000 (20:43 +0000)]
Fix typo.

18 years agoFix for FreeBSD. Fixes http://python.org/sf/1470353.
Thomas Heller [Sat, 15 Apr 2006 20:23:54 +0000 (20:23 +0000)]
Fix for FreeBSD.  Fixes http://python.org/sf/1470353.
Looks like a libffi bug.

18 years agoZap ZAP.
Martin v. Löwis [Sat, 15 Apr 2006 18:14:21 +0000 (18:14 +0000)]
Zap ZAP.

18 years agoPatch #1470875: Building Python with MS Free Compiler.
Martin v. Löwis [Sat, 15 Apr 2006 18:06:54 +0000 (18:06 +0000)]
Patch #1470875: Building Python with MS Free Compiler.

18 years agoRemove partial change (don't edit, commit and think at the same time :P)
Thomas Wouters [Sat, 15 Apr 2006 17:36:42 +0000 (17:36 +0000)]
Remove partial change (don't edit, commit and think at the same time :P)

18 years agoRe-instate backward compatibility by defining Py_CLEAR if it isn't
Thomas Wouters [Sat, 15 Apr 2006 17:33:14 +0000 (17:33 +0000)]
Re-instate backward compatibility by defining Py_CLEAR if it isn't
available.

18 years agoUse Py_CLEAR instead of in-place DECREF/XDECREF or custom macros, for
Thomas Wouters [Sat, 15 Apr 2006 17:28:34 +0000 (17:28 +0000)]
Use Py_CLEAR instead of in-place DECREF/XDECREF or custom macros, for
tp_clear methods.

18 years agoClear dummy and emptyfrozenset, so that we don't have
Martin v. Löwis [Sat, 15 Apr 2006 12:47:23 +0000 (12:47 +0000)]
Clear dummy and emptyfrozenset, so that we don't have
dangling references in case of a Py_Initialize/Py_Finalize
cycle.

18 years agoFix sys.getobjects(0): we get a reference to the
Martin v. Löwis [Sat, 15 Apr 2006 12:46:09 +0000 (12:46 +0000)]
Fix sys.getobjects(0): we get a reference to the
arena's "private" list of objects, so there might
be two references to that list.

18 years agoUnlink the structseq type from the global list of
Martin v. Löwis [Sat, 15 Apr 2006 12:45:05 +0000 (12:45 +0000)]
Unlink the structseq type from the global list of
objects before initializing it. It might be linked
already if there was a Py_Initialize/Py_Finalize
cycle earlier; not unlinking it would break the global
list.

18 years agoPatch #1191700: Adjust column alignment in bdb breakpoint lists.
Martin v. Löwis [Sat, 15 Apr 2006 08:41:11 +0000 (08:41 +0000)]
Patch #1191700: Adjust column alignment in bdb breakpoint lists.
Backported to 2.4.

18 years agoPatch #1191065: Fix preprocessor problems on systems where recvfrom
Martin v. Löwis [Sat, 15 Apr 2006 08:35:59 +0000 (08:35 +0000)]
Patch #1191065: Fix preprocessor problems on systems where recvfrom
is a macro.

18 years agoPatch #1161914: Add python-config.
Martin v. Löwis [Sat, 15 Apr 2006 08:13:05 +0000 (08:13 +0000)]
Patch #1161914: Add python-config.

18 years agoframe_clear(): Explain why it's important to make the frame
Tim Peters [Sat, 15 Apr 2006 03:30:08 +0000 (03:30 +0000)]
frame_clear():  Explain why it's important to make the frame
look dead right at the start.  Use Py_CLEAR for four more
frame members.

18 years agoframe_traverse(): Use the standard Py_VISIT macro.
Tim Peters [Sat, 15 Apr 2006 03:22:46 +0000 (03:22 +0000)]
frame_traverse():  Use the standard Py_VISIT macro.

Py_VISIT:  cast the `op` argument to PyObject* when calling
`visit()`.  Else the caller has to pay too much attention to
this silly detail (e.g., frame_traverse needs to traverse
`struct _frame *` and `PyCodeObject *` pointers too).

18 years agoTrimmed trailing whitespace.
Tim Peters [Sat, 15 Apr 2006 03:15:24 +0000 (03:15 +0000)]
Trimmed trailing whitespace.

18 years agoThere were no comments explaining what Py_CLEAR() did or
Tim Peters [Sat, 15 Apr 2006 02:14:03 +0000 (02:14 +0000)]
There were no comments explaining what Py_CLEAR() did or
why it's important.  Now there are ;-)

If someone else hasn't already, I'll add a Py_CLEAR cleanup
task to the TODO Wiki next.

18 years agoChanged comments to make sense now that the LazyList-based
Tim Peters [Sat, 15 Apr 2006 01:48:57 +0000 (01:48 +0000)]
Changed comments to make sense now that the LazyList-based
examples no longer require any explicit closing to avoid
leaking.

That the tee-based examples still do is (I think) still a
mystery.  Part of the mystery is that gc.garbage remains
empty:  if it were the case that some generator in a trash
cycle said it needed finalization, suppressing collection
of that cycle, that generator _would_ show up in gc.garbage.

So this is acting more like, e.g., some tp_traverse slot
isn't visiting all the pointers it should (in which case
the skipped pointer(s) would act like an external root,
silently suppressing collection of everything reachable
from it(them)).

18 years agoFix SF#1470508: crash in generator cycle finalization. There were two
Phillip J. Eby [Sat, 15 Apr 2006 01:02:17 +0000 (01:02 +0000)]
Fix SF#1470508: crash in generator cycle finalization.  There were two
problems: first, PyGen_NeedsFinalizing() had an off-by-one bug that
prevented it from ever saying a generator didn't need finalizing, and
second, frame objects cleared themselves in a way that caused their
owning generator to think they were still executable, causing a double
deallocation of objects on the value stack if there was still a loop
on the block stack.  This revision also removes some unnecessary
close() operations from test_generators that are now appropriately
handled by the cycle collector.

18 years agoCoverity-found bug: datetime_strptime() failed to check for NULL return from
Thomas Wouters [Fri, 14 Apr 2006 21:23:42 +0000 (21:23 +0000)]
Coverity-found bug: datetime_strptime() failed to check for NULL return from
PySequence_GetItem of the time.strptime() result. Not a high probability
bug, but not inconceivable either, considering people can provide their own
'time' module.

18 years agoAdd an item; better crediting; fix error in SQL example; minor edits
Andrew M. Kuchling [Fri, 14 Apr 2006 20:35:17 +0000 (20:35 +0000)]
Add an item; better crediting; fix error in SQL example; minor edits

18 years agoTypo fix
Andrew M. Kuchling [Fri, 14 Apr 2006 20:32:36 +0000 (20:32 +0000)]
Typo fix

18 years agoWhitespace normalization.
Tim Peters [Fri, 14 Apr 2006 18:34:14 +0000 (18:34 +0000)]
Whitespace normalization.

18 years agoAdd a BufferedIncrementalEncoder class that can be used for implementing
Walter Dörwald [Fri, 14 Apr 2006 18:25:39 +0000 (18:25 +0000)]
Add a BufferedIncrementalEncoder class that can be used for implementing
an incremental encoder that must retain part of the data between calls
to the encode() method.

Fix the incremental encoder and decoder for the IDNA encoding.

This closes SF patch #1453235.

18 years agoMake error message less misleading for u"a..b".encode("idna").
Walter Dörwald [Fri, 14 Apr 2006 17:00:36 +0000 (17:00 +0000)]
Make error message less misleading for u"a..b".encode("idna").

18 years agoPatch #702933: Undocument PyObject_NEW, PyObject_NEW_VAR,
Martin v. Löwis [Fri, 14 Apr 2006 15:58:03 +0000 (15:58 +0000)]
Patch #702933: Undocument PyObject_NEW, PyObject_NEW_VAR,
and PyObject_DEL.

18 years agoFix wrong attribute name.
Walter Dörwald [Fri, 14 Apr 2006 15:40:54 +0000 (15:40 +0000)]
Fix wrong attribute name.

18 years agoMake raise statements PEP 8 compatible.
Walter Dörwald [Fri, 14 Apr 2006 15:22:27 +0000 (15:22 +0000)]
Make raise statements PEP 8 compatible.

18 years agoPatch #1045620: Prepend Modules/ before Setup in stdout.
Martin v. Löwis [Fri, 14 Apr 2006 15:16:15 +0000 (15:16 +0000)]
Patch #1045620: Prepend Modules/ before Setup in stdout.

18 years agoPatch #1470300: Port _hotshot to QNX4.25 port.
Martin v. Löwis [Fri, 14 Apr 2006 15:07:46 +0000 (15:07 +0000)]
Patch #1470300: Port _hotshot to QNX4.25 port.

18 years agoPatch #1355883: Build Python-ast.c and Python-ast.h
Martin v. Löwis [Fri, 14 Apr 2006 15:02:32 +0000 (15:02 +0000)]
Patch #1355883: Build Python-ast.c and Python-ast.h
independently. Fixes #1355883.

18 years agoShow case: reference cycles involving only the ob_type field are rather
Armin Rigo [Fri, 14 Apr 2006 14:58:30 +0000 (14:58 +0000)]
Show case: reference cycles involving only the ob_type field are rather
uncommon but possible.  Inspired by SF bug 1469629.

18 years agoTypo fix
Andrew M. Kuchling [Fri, 14 Apr 2006 14:54:18 +0000 (14:54 +0000)]
Typo fix

18 years agoPatch #1324762: Change --with-cxx to --with-cxx-main.
Martin v. Löwis [Fri, 14 Apr 2006 14:34:26 +0000 (14:34 +0000)]
Patch #1324762: Change --with-cxx to --with-cxx-main.

18 years agoCall encode()/decode() with final==True as the last call in the
Walter Dörwald [Fri, 14 Apr 2006 14:03:55 +0000 (14:03 +0000)]
Call encode()/decode() with final==True as the last call in the
incremental codec tests.

18 years agoTypo fixes
Andrew M. Kuchling [Fri, 14 Apr 2006 12:42:09 +0000 (12:42 +0000)]
Typo fixes

18 years agoAdd more items
Andrew M. Kuchling [Fri, 14 Apr 2006 12:41:19 +0000 (12:41 +0000)]
Add more items

18 years agoTypo fix
Andrew M. Kuchling [Fri, 14 Apr 2006 12:07:41 +0000 (12:07 +0000)]
Typo fix

18 years agoAdd an item and a reminder
Andrew M. Kuchling [Fri, 14 Apr 2006 10:29:55 +0000 (10:29 +0000)]
Add an item and a reminder

18 years agoMake Py_BuildValue, PyObject_CallFunction and
Martin v. Löwis [Fri, 14 Apr 2006 09:08:42 +0000 (09:08 +0000)]
Make Py_BuildValue, PyObject_CallFunction and
PyObject_CallMethod aware of PY_SSIZE_T_CLEAN.

18 years agoCopy note from leakers README here too. We want to keep all test cases.
Neal Norwitz [Fri, 14 Apr 2006 06:35:46 +0000 (06:35 +0000)]
Copy note from leakers README here too.  We want to keep all test cases.

18 years agoRestore test tee with some modifications.
Neal Norwitz [Fri, 14 Apr 2006 06:33:24 +0000 (06:33 +0000)]
Restore test tee with some modifications.
The test case came from test_generators, not test_itertools.
Ensure there's no cyclic garbage we are counting.

This is weird because it leaks, then reaches a limit:

python.exe -i test_tee.py
>>> leak()
0
[26633 refs]
>>> leak()
0
[26658 refs]
>>> leak()
0
[26683 refs]
>>> leak()
0
[26708 refs]
>>> leak()
0
[26708 refs]
>>> leak()
0
[26708 refs]
>>> leak()
0

18 years agoMove the old test_generator_cycle.py which leaked but was removed into the test
Neal Norwitz [Fri, 14 Apr 2006 06:11:08 +0000 (06:11 +0000)]
Move the old test_generator_cycle.py which leaked but was removed into the test

18 years agoUpdate notes to address cyclic garbage and what should happen when a test is fixed.
Neal Norwitz [Fri, 14 Apr 2006 05:35:39 +0000 (05:35 +0000)]
Update notes to address cyclic garbage and what should happen when a test is fixed.

18 years agoSF Bug #1454485, array.array('u') could crash the interpreter when
Neal Norwitz [Fri, 14 Apr 2006 05:20:28 +0000 (05:20 +0000)]
SF Bug #1454485, array.array('u') could crash the interpreter when
passing a string.  Martin already fixed the actual crash by ensuring
Py_UNICODE is unsigned.  As discussed on python-dev, this fix
removes the possibility of creating a unicode string from a raw buffer.

There is an outstanding question of how to fix the crash in 2.4.

18 years agoAdded George Yoshida.
Tim Peters [Fri, 14 Apr 2006 04:54:58 +0000 (04:54 +0000)]
Added George Yoshida.

18 years agoALIGNMENT_SHIFT is not used
Neal Norwitz [Fri, 14 Apr 2006 03:53:34 +0000 (03:53 +0000)]
ALIGNMENT_SHIFT is not used

18 years agoWhen using -R, lots of "*** DocTestRunner.merge:" nuisance messages
Tim Peters [Thu, 13 Apr 2006 23:12:24 +0000 (23:12 +0000)]
When using -R, lots of "*** DocTestRunner.merge:" nuisance messages
appear.  Get rid of them by nuking doctest's default DocTestRunner
instance as part of cleanup().  Also cleanup() before running the
first test repetition (the test was run once before we get into
the -R branch).

18 years agoWhitespace normalization.
Tim Peters [Thu, 13 Apr 2006 22:58:42 +0000 (22:58 +0000)]
Whitespace normalization.

18 years agoraise the minimum supported BerkeleyDB version to 3.3 and add notes to
Gregory P. Smith [Thu, 13 Apr 2006 19:19:01 +0000 (19:19 +0000)]
raise the minimum supported BerkeleyDB version to 3.3 and add notes to
news about this and a couple other recent fixes.

18 years agoForce 8-alignment of memory blocks, as needed on
Martin v. Löwis [Thu, 13 Apr 2006 19:16:13 +0000 (19:16 +0000)]
Force 8-alignment of memory blocks, as needed on
64-bit machines that require pointers to be aligned (e.g. IA64)

18 years agoFix type errors.
Martin v. Löwis [Thu, 13 Apr 2006 13:08:58 +0000 (13:08 +0000)]
Fix type errors.