Thomas Wouters [Tue, 7 Mar 2006 13:38:14 +0000 (13:38 +0000)]
Coverity-found bug: don't use temp->next *before* checking it for NULL. Also
return rather than use it again.
Georg Brandl [Tue, 7 Mar 2006 12:48:03 +0000 (12:48 +0000)]
Bug #
1432525: os.listdir now releases the GIL while calling
readdir().
Martin v. Löwis [Tue, 7 Mar 2006 12:08:51 +0000 (12:08 +0000)]
Change int to Py_ssize_t in several places.
Add (int) casts to silence compiler warnings.
Raise Python exceptions for overflows.
Thomas Wouters [Tue, 7 Mar 2006 11:58:50 +0000 (11:58 +0000)]
Don't DECREF a borrowed reference.
Barry Warsaw [Tue, 7 Mar 2006 09:46:03 +0000 (09:46 +0000)]
SF patch #
1443865; gc.get_count() added and optional argument 'generation'
added to gc.collect(). Updated docs, unit test, and NEWS entry.
(Also, fixed a typo in NEWS.)
Neal Norwitz [Tue, 7 Mar 2006 05:01:00 +0000 (05:01 +0000)]
Add a note about the bug fixes
Neal Norwitz [Tue, 7 Mar 2006 04:48:24 +0000 (04:48 +0000)]
Thanks to Coverity, these were all reported by their Prevent tool.
All of these (except _lsprof.c) should be backported. Particularly
the hotshot change which validates sys.path. Can someone backport?
Neal Norwitz [Mon, 6 Mar 2006 23:31:56 +0000 (23:31 +0000)]
Fix warnings on x86 (32-bit) and support Win64.
Neal Norwitz [Mon, 6 Mar 2006 23:31:27 +0000 (23:31 +0000)]
Fix warnings on x86 (32-bit).
Neal Norwitz [Mon, 6 Mar 2006 23:07:34 +0000 (23:07 +0000)]
Try to cleanup the error handling a bit so there aren't false positives
from static analysis. v was already checked for NULL above, so we don't
need a second check.
Neal Norwitz [Mon, 6 Mar 2006 23:04:04 +0000 (23:04 +0000)]
Ignore leaks from the socket test too. Sigh, this list is getting long. :-(
Walter Dörwald [Mon, 6 Mar 2006 22:39:12 +0000 (22:39 +0000)]
If size is specified, try to read at least size characters.
This is a alternative version of patch #
1379332.
Martin v. Löwis [Mon, 6 Mar 2006 16:32:05 +0000 (16:32 +0000)]
lower-case time_t.
Martin v. Löwis [Mon, 6 Mar 2006 16:30:25 +0000 (16:30 +0000)]
Patch #
1437769: notice that time_t is a 64-bit type in VS2005
Hye-Shik Chang [Mon, 6 Mar 2006 07:51:19 +0000 (07:51 +0000)]
Check NULL if Py_InitModule fails.
Martin v. Löwis [Sun, 5 Mar 2006 16:47:11 +0000 (16:47 +0000)]
Use devenv.com to better obtain output.
Martin v. Löwis [Sun, 5 Mar 2006 16:18:44 +0000 (16:18 +0000)]
Add buildbot support scripts.
Martin v. Löwis [Sun, 5 Mar 2006 14:04:26 +0000 (14:04 +0000)]
Add _msi.pyd.
Add warning on Win9x.
Martin v. Löwis [Sun, 5 Mar 2006 13:52:20 +0000 (13:52 +0000)]
Open dll file in binary.
Martin v. Löwis [Sun, 5 Mar 2006 13:39:10 +0000 (13:39 +0000)]
Update copyright notice.
Martin v. Löwis [Sun, 5 Mar 2006 13:36:04 +0000 (13:36 +0000)]
Import bdist_msi
Martin v. Löwis [Sun, 5 Mar 2006 05:33:54 +0000 (05:33 +0000)]
Use %Id for size_t-ish things on Win64.
Neal Norwitz [Sun, 5 Mar 2006 02:16:12 +0000 (02:16 +0000)]
Backout the last hack and add in this new one.
The failure definitely seems timing related. This change *seems* to work.
Since the failure isn't doesn't occur consistently, it's hard to tell.
Running these tests on Solaris in this order:
test_urllibnet test_operator test_cgi \
test_isinstance test_future test_ast test_logging
generally caused a failure (about 50% of the time) before the sleep.
I couldn't provoke the failure with the sleep.
This should really be cleaned up by using threading.Events or something
so it is not timing dependent and doesn't hang forever on failure.
Neal Norwitz [Sat, 4 Mar 2006 23:56:53 +0000 (23:56 +0000)]
Oops, urllib may or may not already be loaded.
Neal Norwitz [Sat, 4 Mar 2006 23:31:45 +0000 (23:31 +0000)]
Fix SF #
1442937, Bad Coroutine link
Neal Norwitz [Sat, 4 Mar 2006 23:13:41 +0000 (23:13 +0000)]
Fix spurious test failure of test_socket_ssl when run in this order:
test_codecmaps_tw test_importhooks test_socket_ssl
I don't completely understand the cause, but there's a lot of import magic
going on and this is the smallest change which fixes the problem.
Neal Norwitz [Sat, 4 Mar 2006 20:01:53 +0000 (20:01 +0000)]
Make PyGC_Collect() use Py_ssize_t.
Neal Norwitz [Sat, 4 Mar 2006 20:00:59 +0000 (20:00 +0000)]
Oops, forgot to include this in the last checkin.
Actually define Py_RefTotal as a Py_ssize_t.
Neal Norwitz [Sat, 4 Mar 2006 19:58:13 +0000 (19:58 +0000)]
Use Py_ssize_t for _Py_RefTotal.
I tried to handle Win64 properly, but please review.
Neal Norwitz [Sat, 4 Mar 2006 19:01:22 +0000 (19:01 +0000)]
Tabify
Neal Norwitz [Sat, 4 Mar 2006 18:52:26 +0000 (18:52 +0000)]
Get rid of run_err_mod(). It was only used in two places.
One place it wasn't necessary since mod was already checked.
Inline the check that mod != NULL for the other use.
Neal Norwitz [Sat, 4 Mar 2006 18:49:58 +0000 (18:49 +0000)]
Change some sequnce APIs to use Py_ssize_t.
Neal Norwitz [Sat, 4 Mar 2006 18:41:19 +0000 (18:41 +0000)]
Use Py_ssize_t for PySet_Size() like all the other Py*_Size() functions.
Neal Norwitz [Sat, 4 Mar 2006 18:35:47 +0000 (18:35 +0000)]
Remove test for timing (already not built since commented out in setup.py).
Add note to NEWS.
Hye-Shik Chang [Sat, 4 Mar 2006 16:08:19 +0000 (16:08 +0000)]
- Modernize code to use Py_ssize_t more intensively.
- Do some minor code clean-ups.
Barry Warsaw [Sat, 4 Mar 2006 04:55:02 +0000 (04:55 +0000)]
Actually this file should have svn:eol-style CRLF since it's specifically
testing line ending compatibility.
Tim Peters [Sat, 4 Mar 2006 02:57:23 +0000 (02:57 +0000)]
Set eol-style to native -- doesn't appear to be any reason
for it to claim it has mime-type application/octet-stream.
Tim Peters [Sat, 4 Mar 2006 02:43:44 +0000 (02:43 +0000)]
Whitespace normalization.
Neal Norwitz [Fri, 3 Mar 2006 23:11:42 +0000 (23:11 +0000)]
SF #
1442767, docs for os.statvfs miss f_bsize parameter. Will backport.
Neal Norwitz [Fri, 3 Mar 2006 21:53:14 +0000 (21:53 +0000)]
I think the test_logging failure on Solaris is timing related. We don't
want to wait forever if we don't receive the last message. But we also
don't want the test to fail if we shutdown too quickly. I can't reliably
reproduce this failure, so I'm kinda guessing this is the problem.
We'll see if this band-aid helps.
Neal Norwitz [Fri, 3 Mar 2006 20:29:11 +0000 (20:29 +0000)]
SF bug
1442442: LIST_APPEND optimization got lost in the AST merge.
Add it back.
Neal Norwitz [Fri, 3 Mar 2006 20:21:48 +0000 (20:21 +0000)]
Fix compiler breakage related to absolute imports
Neal Norwitz [Fri, 3 Mar 2006 19:13:57 +0000 (19:13 +0000)]
Remove commented code
Neal Norwitz [Fri, 3 Mar 2006 19:12:58 +0000 (19:12 +0000)]
Hopefully this removes all vestiges of CO_GENERATOR_ALLOWED that can
be removed in 2.x.
Thomas Wouters [Fri, 3 Mar 2006 18:16:20 +0000 (18:16 +0000)]
Add support for absolute/relative imports and if/else expressions:
- regenerate ast.py
- add future flags for absolute-import and with-statement so they
(hopefully) properly get set in code-object flags
- try out if/else expressions in actual code for the hell of it.
Seems to generate the same kind of bytecode as the normal compiler.
Thomas Wouters [Fri, 3 Mar 2006 18:11:37 +0000 (18:11 +0000)]
Update for absolute/relative imports and ifelse-expressions.
Neal Norwitz [Fri, 3 Mar 2006 17:25:57 +0000 (17:25 +0000)]
Provide checkin access to Ronald and Bob
Neal Norwitz [Fri, 3 Mar 2006 00:48:46 +0000 (00:48 +0000)]
Stop building timing module, it's old and deprecated
Neal Norwitz [Fri, 3 Mar 2006 00:43:06 +0000 (00:43 +0000)]
Add test_quopri to know spurious leakers
Brett Cannon [Thu, 2 Mar 2006 22:08:16 +0000 (22:08 +0000)]
Text moved to PEP 339.
Brett Cannon [Thu, 2 Mar 2006 22:07:40 +0000 (22:07 +0000)]
Fix minor docstring typo.
Tim Peters [Thu, 2 Mar 2006 21:41:18 +0000 (21:41 +0000)]
Document the purpose of the struct _block members.
Tim Peters [Thu, 2 Mar 2006 21:14:45 +0000 (21:14 +0000)]
Added words about what PyArena_Malloc() does.
Tim Peters [Thu, 2 Mar 2006 21:04:08 +0000 (21:04 +0000)]
Beefed up description of what this does; new XXX.
Tim Peters [Thu, 2 Mar 2006 20:48:25 +0000 (20:48 +0000)]
New XXX pointing out errors in the description of
PyArena_New(); unsure what the intent is.
Thomas Wouters [Thu, 2 Mar 2006 20:41:27 +0000 (20:41 +0000)]
Fix crashing bug in tokenizer, when tokenizing files with non-ASCII bytes
but without a specified encoding: decoding_fgets() (and decoding_feof()) can
return NULL and fiddle with the 'tok' struct, making tok->buf NULL. This is
okay in the other cases of calls to decoding_*(), it seems, but not in this
one.
This should get a test added, somewhere, but the testsuite doesn't seem to
test encoding anywhere (although plenty of tests use it.)
It seems to me that decoding errors in other places in the code (like at the
start of a token, instead of in the middle of one) make the code end up
adding small integers to NULL pointers, but happen to check for error states
before using the calculated new pointers. I haven't been able to trigger any
other crashes, in any case.
I would nominate this file for a comlete rewrite for Py3k. The whole
decoding trick is too bolted-on for my tastes.
Tim Peters [Thu, 2 Mar 2006 20:37:32 +0000 (20:37 +0000)]
Trimmed trailing whitespace.
Brett Cannon [Thu, 2 Mar 2006 18:34:57 +0000 (18:34 +0000)]
Fix refleak in PyErr_Display().
Thomas Wouters [Thu, 2 Mar 2006 17:58:27 +0000 (17:58 +0000)]
Explain why we use the unsigned int format for a signed int variable.
(Should 'code' be cast to the right pointer type?)
Brett Cannon [Thu, 2 Mar 2006 17:47:01 +0000 (17:47 +0000)]
Fix mismatch opening and closing quotes on a string.
Neal Norwitz [Thu, 2 Mar 2006 08:08:42 +0000 (08:08 +0000)]
Fix refleak if from __future__ import was not first
Neal Norwitz [Thu, 2 Mar 2006 07:54:28 +0000 (07:54 +0000)]
Use Py_ssize_t since we are working with list size below
Neal Norwitz [Thu, 2 Mar 2006 07:51:44 +0000 (07:51 +0000)]
Spell threading write, not thraeading
Thomas Wouters [Thu, 2 Mar 2006 05:05:17 +0000 (05:05 +0000)]
Properly fix Py_SAFE_DOWNCAST-triggerd bugs.
Thomas Wouters [Thu, 2 Mar 2006 04:48:27 +0000 (04:48 +0000)]
Py_SAFE_DOWNCAST isn't quite doing the right thing for going from Py_ssize_t
to an unsigned int (and back again) on 64-bit machines, even though the
actual value of the Py_ssize_t variable is way below 31 bits. I suspect
compiler-error.
Brett Cannon [Thu, 2 Mar 2006 04:31:55 +0000 (04:31 +0000)]
Add a missing Py_DECREF to BaseException__unicode__ .
Guido van Rossum [Thu, 2 Mar 2006 04:24:01 +0000 (04:24 +0000)]
Fix failure of test_compiler.py when compiling test_contextlib.py.
The culprit was an expression-less yield -- the first apparently in
the standard library. I added a unit test for this.
Also removed the hack to force compilation of test_with.py.
Neal Norwitz [Thu, 2 Mar 2006 04:06:10 +0000 (04:06 +0000)]
Update known issues to reflect reality
Neal Norwitz [Thu, 2 Mar 2006 04:03:44 +0000 (04:03 +0000)]
Remove duplicate entry
Brett Cannon [Thu, 2 Mar 2006 03:52:06 +0000 (03:52 +0000)]
Fix latex typos as spotted by George Yoshida.
Martin v. Löwis [Thu, 2 Mar 2006 00:31:27 +0000 (00:31 +0000)]
Fix memory leak on attributes.
Thomas Wouters [Thu, 2 Mar 2006 00:21:10 +0000 (00:21 +0000)]
Make Py_ssize_t-clean
Thomas Wouters [Wed, 1 Mar 2006 23:49:13 +0000 (23:49 +0000)]
Make Py_ssize_t clean.
Martin v. Löwis [Wed, 1 Mar 2006 23:24:34 +0000 (23:24 +0000)]
Reformat the exception message by going through a list.
Martin v. Löwis [Wed, 1 Mar 2006 23:10:49 +0000 (23:10 +0000)]
Remove UNLESS.
Thomas Wouters [Wed, 1 Mar 2006 23:10:05 +0000 (23:10 +0000)]
Use Py_ssize_t for arithmetic on Py_ssize_t's, instead of unsigned ints.
Tim Peters [Wed, 1 Mar 2006 23:02:57 +0000 (23:02 +0000)]
Whitespace normalization.
Thomas Wouters [Wed, 1 Mar 2006 22:54:36 +0000 (22:54 +0000)]
Fix gcc (4.0.x) warning about use of uninitialized variable.
Martin v. Löwis [Wed, 1 Mar 2006 22:49:05 +0000 (22:49 +0000)]
Patch #
1440601: Add col_offset attribute to AST nodes.
Thomas Wouters [Wed, 1 Mar 2006 22:45:36 +0000 (22:45 +0000)]
Rework channelnumber/samplesize detetion code's output variables a bit to
convince gcc (4.0.x) the variables are never used uninitialized (and raising
a proper exception if they ever are.)
Thomas Wouters [Wed, 1 Mar 2006 22:34:09 +0000 (22:34 +0000)]
Fix incompatible assignment warning from previous checkin.
Thomas Wouters [Wed, 1 Mar 2006 22:30:47 +0000 (22:30 +0000)]
Fix gcc (4.0.x) warning about use of uninitialized variables.
(PyMarshal_ReadShortFromFile() is only used in zipimport.c, I don't believe
the extra initializations will matter one way or another.)
Thomas Wouters [Wed, 1 Mar 2006 22:15:15 +0000 (22:15 +0000)]
Silence gcc (4.0.x) warning about use of uninitialized value.
Brett Cannon [Wed, 1 Mar 2006 22:10:49 +0000 (22:10 +0000)]
Document PEP 352 changes. Also added GeneratorExit.
Thomas Wouters [Wed, 1 Mar 2006 22:06:23 +0000 (22:06 +0000)]
Fix uninitialized value. (Why are we using bools instead of ints, like we do
everywhere else?)
Thomas Wouters [Wed, 1 Mar 2006 21:59:44 +0000 (21:59 +0000)]
Fix brainfart.
Thomas Wouters [Wed, 1 Mar 2006 21:58:30 +0000 (21:58 +0000)]
Remove gcc (4.0.x) warning about uninitialized value by explicitly setting
the sentinel value in the main function, rather than the helper. This
function could possibly do with an early-out if any of the helper calls ends
up with a len of 0, but I doubt it really matters (how common are malformed
hangul syllables, really?)
Thomas Wouters [Wed, 1 Mar 2006 21:50:07 +0000 (21:50 +0000)]
Make Py_ssize_t-clean.
Thomas Wouters [Wed, 1 Mar 2006 21:37:32 +0000 (21:37 +0000)]
Make Py_ssize_t-clean.
Martin v. Löwis [Wed, 1 Mar 2006 21:33:54 +0000 (21:33 +0000)]
Fix more memory leaks. Will backport to 2.4.
Thomas Wouters [Wed, 1 Mar 2006 21:31:21 +0000 (21:31 +0000)]
Fix C99-ism, and add XXX to comment
Martin v. Löwis [Wed, 1 Mar 2006 21:11:49 +0000 (21:11 +0000)]
Make failures in test cases print failing source file.
Brett Cannon [Wed, 1 Mar 2006 20:53:08 +0000 (20:53 +0000)]
Update for 'with' statement.
Guido van Rossum [Wed, 1 Mar 2006 17:10:01 +0000 (17:10 +0000)]
Fix a bug in nested() - if one of the sub-context-managers swallows the
exception, it should not be propagated up. With unit tests.
Martin v. Löwis [Wed, 1 Mar 2006 16:56:25 +0000 (16:56 +0000)]
Change GC refcount to Py_ssize_t.
Martin v. Löwis [Wed, 1 Mar 2006 16:55:42 +0000 (16:55 +0000)]
Reconst parameters that lost their const in the AST merge.
Martin v. Löwis [Wed, 1 Mar 2006 16:37:55 +0000 (16:37 +0000)]
Fix warning that texcheck complained about.
Jeremy Hylton [Wed, 1 Mar 2006 15:47:05 +0000 (15:47 +0000)]
Tabify and reflow some long lines.
Much of the peephole optimizer is now indented badly, but it's about
to be revised anyway.
Jeremy Hylton [Wed, 1 Mar 2006 15:02:24 +0000 (15:02 +0000)]
Add missing DECREF.
Fredrik Lundh [Wed, 1 Mar 2006 12:43:53 +0000 (12:43 +0000)]
markup glitch (spotted by George Yoshida)