]>
granicus.if.org Git - python/log
Vinay Sajip [Wed, 25 Apr 2012 09:51:35 +0000 (10:51 +0100)]
Merged fixed test from 3.2.
Vinay Sajip [Wed, 25 Apr 2012 09:47:00 +0000 (10:47 +0100)]
Fixed test to work on Windows.
Marc-Andre Lemburg [Wed, 25 Apr 2012 08:54:48 +0000 (10:54 +0200)]
Issue #14605 and #14642:
Issue a warning in case Python\importlib.h needs to be rebuilt,
but there's no Python interpreter around to freeze the bootstrap
script.
Marc-Andre Lemburg [Wed, 25 Apr 2012 00:31:37 +0000 (02:31 +0200)]
Issue #14605: Rename _SourcelessFileLoader to SourcelessFileLoader.
This time also recreating the Python/importlib.h file to make
make happy. See the ticket for details.
Marc-Andre Lemburg [Wed, 25 Apr 2012 00:11:07 +0000 (02:11 +0200)]
Issue #14605: Revert renaming of _SourcelessFileLoader, since it caused
the buildbots to fail.
Marc-Andre Lemburg [Tue, 24 Apr 2012 23:36:48 +0000 (01:36 +0200)]
Issue #14605: Rename _SourcelessFileLoader to SourcelessFileLoader
Vinay Sajip [Tue, 24 Apr 2012 22:33:33 +0000 (23:33 +0100)]
Issue #14632: Updated WatchedFileHandler to deal with race condition. Thanks to John Mulligan for the problem report and patch.
Vinay Sajip [Tue, 24 Apr 2012 22:25:30 +0000 (23:25 +0100)]
Issue #14632: Updated WatchedFileHandler to deal with race condition. Thanks to John Mulligan for the problem report and patch.
Antoine Pitrou [Tue, 24 Apr 2012 20:56:57 +0000 (22:56 +0200)]
Issue #4892: multiprocessing Connections can now be transferred over multiprocessing Connections.
Patch by Richard Oudkerk (sbt).
Lars Gustäbel [Tue, 24 Apr 2012 19:09:17 +0000 (21:09 +0200)]
Merge with 3.2: Issue #14160: TarFile.extractfile() failed to resolve symbolic
links when the links were not located in an archive subdirectory.
Lars Gustäbel [Tue, 24 Apr 2012 19:04:40 +0000 (21:04 +0200)]
Issue #14160: TarFile.extractfile() failed to resolve symbolic links
when the links were not located in an archive subdirectory.
Jesus Cea [Tue, 24 Apr 2012 18:59:17 +0000 (20:59 +0200)]
Closes Issue #14661: posix module: add O_EXEC, O_SEARCH, O_TTY_INIT (I add some Solaris constants too)
Jesus Cea [Tue, 24 Apr 2012 18:44:40 +0000 (20:44 +0200)]
Issue #14661: posix module: add O_EXEC, O_SEARCH, O_TTY_INIT
Benjamin Peterson [Tue, 24 Apr 2012 18:44:18 +0000 (14:44 -0400)]
distiguish between refusing to creating shared keys and error (#13903)
Sandro Tosi [Tue, 24 Apr 2012 17:46:29 +0000 (19:46 +0200)]
Issue #14448: merge with 3.2
Sandro Tosi [Tue, 24 Apr 2012 17:46:06 +0000 (19:46 +0200)]
Issue #14448: mention pytz; patch by Andrew Svetlov
Georg Brandl [Tue, 24 Apr 2012 17:16:24 +0000 (19:16 +0200)]
Fix location of versionaddeds and empty lines.
Martin v. Loewis [Tue, 24 Apr 2012 17:13:57 +0000 (19:13 +0200)]
Account for shared keys in type's __sizeof__ (#13903).
Sandro Tosi [Tue, 24 Apr 2012 16:39:53 +0000 (18:39 +0200)]
Issue #14554: merge with 3.2
Sandro Tosi [Tue, 24 Apr 2012 16:39:24 +0000 (18:39 +0200)]
Issue #14554: correct example for captured_stdout(); patch by Tshepang Lekhonkhobe
Sandro Tosi [Tue, 24 Apr 2012 16:12:13 +0000 (18:12 +0200)]
Issue #13478: merge with 3.2
Sandro Tosi [Tue, 24 Apr 2012 16:11:46 +0000 (18:11 +0200)]
Issue #13478: document timeit.default_timer()
Sandro Tosi [Tue, 24 Apr 2012 15:37:02 +0000 (17:37 +0200)]
Issue #13587: merge with 3.2
Sandro Tosi [Tue, 24 Apr 2012 15:36:41 +0000 (17:36 +0200)]
Issue #13587: use the right RFC2617 name for WWW-Authenticate; patch by Aaron Maenpaa
Benjamin Peterson [Tue, 24 Apr 2012 15:09:20 +0000 (11:09 -0400)]
merge 3.2 (#14658)
Benjamin Peterson [Tue, 24 Apr 2012 15:06:25 +0000 (11:06 -0400)]
don't use a slot wrapper from a different special method (closes #14658)
This also alters the fix to #11603. Specifically, setting __repr__ to
object.__str__ now raises a recursion RuntimeError when str() or repr() is
called instead of silently bypassing the recursion. I believe this behavior is
more correct.
Benjamin Peterson [Tue, 24 Apr 2012 14:32:57 +0000 (10:32 -0400)]
fix dict gc tracking (#13903)
Antoine Pitrou [Tue, 24 Apr 2012 11:55:35 +0000 (13:55 +0200)]
Skip long repr tests when the cached pyc file's path length would exceed the maximum Win32 path length.
This should fix a failure on one of the XP buildbots.
Benjamin Peterson [Tue, 24 Apr 2012 01:45:40 +0000 (21:45 -0400)]
make pointer arith c89
Benjamin Peterson [Mon, 23 Apr 2012 22:07:01 +0000 (18:07 -0400)]
use correct base ptr
Benjamin Peterson [Mon, 23 Apr 2012 22:00:25 +0000 (18:00 -0400)]
simplify and reformat
Victor Stinner [Mon, 23 Apr 2012 21:36:38 +0000 (23:36 +0200)]
Close #14648: Compute correctly maxchar in str.format() for substrin
R David Murray [Mon, 23 Apr 2012 18:45:45 +0000 (14:45 -0400)]
Merge #14640: Fix typos/syntax in pyporting.rst.
Patch by Dionysios Kalofonos.
R David Murray [Mon, 23 Apr 2012 18:44:00 +0000 (14:44 -0400)]
#14640: Fix typos/syntax in pyporting.rst.
Patch by Dionysios Kalofonos.
Sandro Tosi [Mon, 23 Apr 2012 18:07:15 +0000 (20:07 +0200)]
Issue #14650: fix typo in shutil.disk_usage() docstring; patch by Hobson Lane
Benjamin Peterson [Mon, 23 Apr 2012 17:50:05 +0000 (13:50 -0400)]
merge heads
Sandro Tosi [Mon, 23 Apr 2012 17:45:45 +0000 (19:45 +0200)]
Issue #14641: merge with 3.2
Sandro Tosi [Mon, 23 Apr 2012 17:45:07 +0000 (19:45 +0200)]
Issue #14641: minor fixes to sockets Howto; patch by Dionysios Kalofonos
Benjamin Peterson [Mon, 23 Apr 2012 17:44:32 +0000 (13:44 -0400)]
fix instance dicts with str subclasses (#13903)
R David Murray [Mon, 23 Apr 2012 17:26:44 +0000 (13:26 -0400)]
merge #14638: pydoc now treats non-str __name__ as None instead of raising
Original patch by Peter Otten.
R David Murray [Mon, 23 Apr 2012 17:23:57 +0000 (13:23 -0400)]
#14638: pydoc now treats non-str __name__ as None instead of raising
Original patch by Peter Otten.
Senthil Kumaran [Mon, 23 Apr 2012 15:56:36 +0000 (23:56 +0800)]
merge heads
Senthil Kumaran [Mon, 23 Apr 2012 15:55:50 +0000 (23:55 +0800)]
news for issue13684
Senthil Kumaran [Mon, 23 Apr 2012 15:53:51 +0000 (23:53 +0800)]
3.2 - Fix for Issue13684 - httplib tunnel infinite loop
Senthil Kumaran [Mon, 23 Apr 2012 15:53:16 +0000 (23:53 +0800)]
news for issue13684
Benjamin Peterson [Mon, 23 Apr 2012 15:50:47 +0000 (11:50 -0400)]
don't make shared keys with dict subclasses
Senthil Kumaran [Mon, 23 Apr 2012 15:50:07 +0000 (23:50 +0800)]
3.2 - Fix for Issue13684 - httplib tunnel infinite loop
Benjamin Peterson [Mon, 23 Apr 2012 15:25:49 +0000 (11:25 -0400)]
merge 3.2
Benjamin Peterson [Mon, 23 Apr 2012 15:25:32 +0000 (11:25 -0400)]
add Mark Shannon
Benjamin Peterson [Mon, 23 Apr 2012 15:24:50 +0000 (11:24 -0400)]
Implement PEP 412: Key-sharing dictionaries (closes #13903)
Patch from Mark Shannon.
Benjamin Peterson [Mon, 23 Apr 2012 14:55:29 +0000 (10:55 -0400)]
inherit maxchar of field value where needed (closes #14648)
Benjamin Peterson [Mon, 23 Apr 2012 14:08:20 +0000 (10:08 -0400)]
merge 3.2
Benjamin Peterson [Mon, 23 Apr 2012 14:08:14 +0000 (10:08 -0400)]
sleep here
Brett Cannon [Sun, 22 Apr 2012 23:58:33 +0000 (19:58 -0400)]
Issue #14605: Expose importlib.abc.FileLoader and
importlib.machinery.(FileFinder, SourceFileLoader,
_SourcelessFileLoader, ExtensionFileLoader).
This exposes all of importlib's mechanisms that will become public on
the sys module.
Brett Cannon [Sun, 22 Apr 2012 17:30:07 +0000 (13:30 -0400)]
merge
Brett Cannon [Sun, 22 Apr 2012 17:29:47 +0000 (13:29 -0400)]
Change tests for imp.cache_from_source() to follow os.path.join/split
semantics.
Vinay Sajip [Sun, 22 Apr 2012 17:19:50 +0000 (18:19 +0100)]
Issue #14644: Increased default timeout for SMTPHandler. Note: last commit message referred to the wrong issue number.
Vinay Sajip [Sun, 22 Apr 2012 17:16:14 +0000 (18:16 +0100)]
Issue #14622: Increased default timeout for SMTPHandler.
Brett Cannon [Sun, 22 Apr 2012 17:02:31 +0000 (13:02 -0400)]
merge
Brett Cannon [Sun, 22 Apr 2012 15:45:07 +0000 (11:45 -0400)]
Issue #13959: Continue to try to accomodate altsep in importlib by not
ignoring altsep if it already exists on a path when doing a join.
Charles-François Natali [Sun, 22 Apr 2012 13:55:43 +0000 (15:55 +0200)]
Use os.path.samestat() instead of reinventing the wheel.
Nick Coghlan [Sun, 22 Apr 2012 07:16:39 +0000 (17:16 +1000)]
Resolve #14026 (Merge from 3.2)
Nick Coghlan [Sun, 22 Apr 2012 07:11:33 +0000 (17:11 +1000)]
Close issue #14026 by better testing sys.argv handling in test_cmd_line_script (patch by Jason Yeo)
Brett Cannon [Sun, 22 Apr 2012 06:08:59 +0000 (02:08 -0400)]
Mention the backwards-incompatible change to imp.cache_from_source()/source_from_cache().
Brett Cannon [Sun, 22 Apr 2012 06:06:48 +0000 (02:06 -0400)]
merge
Brett Cannon [Sun, 22 Apr 2012 06:06:23 +0000 (02:06 -0400)]
Revert to os.path.join() semantics for path manipulation in importlib
which is different than what imp.cache_from_source() operates on.
Senthil Kumaran [Sun, 22 Apr 2012 02:31:52 +0000 (10:31 +0800)]
issue2193 - Update docs about the legal characters allowed in Cookie name
Senthil Kumaran [Sun, 22 Apr 2012 02:30:17 +0000 (10:30 +0800)]
issue2193 - Update 3.2 docs about legal characters allowed in Cookie name
Brett Cannon [Sun, 22 Apr 2012 01:47:08 +0000 (21:47 -0400)]
Update importlib.h
Brett Cannon [Sun, 22 Apr 2012 01:46:32 +0000 (21:46 -0400)]
Continue the good fight to get Windows to like importlib by fixing a
variable name.
Brett Cannon [Sun, 22 Apr 2012 01:21:54 +0000 (21:21 -0400)]
merge
Brett Cannon [Sun, 22 Apr 2012 01:21:27 +0000 (21:21 -0400)]
Have importlib look for pre-existing path separators when joining
paths.
Senthil Kumaran [Sun, 22 Apr 2012 01:19:49 +0000 (09:19 +0800)]
merge heads
Senthil Kumaran [Sun, 22 Apr 2012 01:19:04 +0000 (09:19 +0800)]
Fix Issue2193 - Allow ":" character in Cookie NAME values
Brett Cannon [Sun, 22 Apr 2012 01:15:28 +0000 (21:15 -0400)]
Fix whitespace.
Brett Cannon [Sun, 22 Apr 2012 01:09:46 +0000 (21:09 -0400)]
Issue #13959: Re-implement imp.find_module() in Lib/imp.py.
Thanks to Eric Snow for taking an initial stab at the implementation.
Brett Cannon [Sat, 21 Apr 2012 23:11:58 +0000 (19:11 -0400)]
Issue #14637: Fix the UNC import test under Windows to actually use
the UNC path. Also clean up sys.path and invalidate finder caches.
Thanks to Vinay Sajip for spotting the use of the wrong path.
Brett Cannon [Sat, 21 Apr 2012 22:55:51 +0000 (18:55 -0400)]
Don't worry about moving imp.get_tag() over to Lib/imp.py.
Brett Cannon [Sat, 21 Apr 2012 22:53:14 +0000 (18:53 -0400)]
merge
Brett Cannon [Sat, 21 Apr 2012 22:52:52 +0000 (18:52 -0400)]
Issue #13959: Re-implement imp.source_from_cache() in Lib/imp.py.
Michael Foord [Sat, 21 Apr 2012 17:32:56 +0000 (18:32 +0100)]
Remove incorrect comment
Michael Foord [Sat, 21 Apr 2012 17:22:28 +0000 (18:22 +0100)]
Closes issue 14634. unittest.mock.create_autospec now supports keyword only arguments.
Michael Foord [Sat, 21 Apr 2012 14:52:11 +0000 (15:52 +0100)]
Closes issue 14636. mock objects raise exceptions from an iterable side_effect
Brett Cannon [Sat, 21 Apr 2012 01:44:46 +0000 (21:44 -0400)]
Issue #13959: Re-implement imp.cache_from_source() in Lib/imp.py.
Brett Cannon [Sat, 21 Apr 2012 01:19:53 +0000 (21:19 -0400)]
Make path manipulation more robust for platforms with alternative path
separators.
Brett Cannon [Fri, 20 Apr 2012 22:04:03 +0000 (18:04 -0400)]
Issue #13959: Keep imp.get_magic() in C code, but cache in importlib
for performance. While get_magic() could move to Lib/imp.py, having to
support PyImport_GetMagicNumber() would lead to equal, if not more, C
code than sticking with the status quo.
Brett Cannon [Fri, 20 Apr 2012 21:34:59 +0000 (17:34 -0400)]
Fix a cleanup.
Brett Cannon [Fri, 20 Apr 2012 21:19:14 +0000 (17:19 -0400)]
Merge
Mark Dickinson [Fri, 20 Apr 2012 20:44:09 +0000 (21:44 +0100)]
Issue #14630: Merge fix from 3.2.
Mark Dickinson [Fri, 20 Apr 2012 20:42:49 +0000 (21:42 +0100)]
Issue #14630: Fix an incorrect access of ob_digit[0] for a zero instance of an int subclass.
Brett Cannon [Fri, 20 Apr 2012 20:29:39 +0000 (16:29 -0400)]
Use a skipUnless decorator instead of conditional renaming.
Mark Dickinson [Fri, 20 Apr 2012 20:21:24 +0000 (21:21 +0100)]
Issue #14339: Improve speed of bin, oct and hex builtins. Patch by Serhiy Storchaka (with minor modifications).
Brett Cannon [Fri, 20 Apr 2012 19:57:46 +0000 (15:57 -0400)]
Issue #14628: Document the fact that import always returns the module
as found in sys.modules and not as what the loader returns (even
though it is required to by PEP 302).
Brett Cannon [Fri, 20 Apr 2012 19:52:17 +0000 (15:52 -0400)]
Issue #14585: test_import now runs all tests under
importlib.test.import_ using builtins.__import__() instead of just the
relative import tests.
Brett Cannon [Fri, 20 Apr 2012 19:31:11 +0000 (15:31 -0400)]
Issue #14599: Support ImportError.path on AIX and HPUX when loading
extension modules.
Brett Cannon [Fri, 20 Apr 2012 19:23:11 +0000 (15:23 -0400)]
merge
Brett Cannon [Fri, 20 Apr 2012 19:22:50 +0000 (15:22 -0400)]
Issue #14599: Generalize a test for ImportError.path and add support
in Python/dynload_shlibs.c.
This should fix the remaining importlib test failure on Windows.
Support in AIX and HP-UX will be in a separate checkin.
Stefan Krah [Fri, 20 Apr 2012 19:00:31 +0000 (21:00 +0200)]
Defensive programming: mpd_isspecial(r) already implies mpd_isspecial(q), but
this is more readable.
Stefan Krah [Fri, 20 Apr 2012 17:59:20 +0000 (19:59 +0200)]
The divmod function for large numbers now has an ACL2 proof. Related changes:
1) Rename _mpd_qbarrett_divmod into _mpd_base_ndivmod: The function is
only marginally related to either Barrett's algorithm or to the version
in Hasselstrom's paper.
2) In places where the proof assumes exact operations, use new versions of
add/sub/multiply that set NaN/Invalid_operation if this condition is
not met. According to the proof this cannot happen, so this should be
regarded as an extra safety net.
3) Raise Division_impossible for operands with a number of digits greater
than MPD_MAX_PREC. This facilitates the audit of the function and can
practically only occur in the 32-bit version under conditions where
a MemoryError is already imminent.
4) Use _mpd_qmul() in places where the result can exceed MPD_MAX_PREC in
a well defined manner.
5) Test for mpd_isspecial(qq) in a place where the addition of one
can theoretically trigger a Malloc_error.
6) Remove redundant code in _mpd_qdivmod().
7) Add many comments.
Brett Cannon [Fri, 20 Apr 2012 17:34:35 +0000 (13:34 -0400)]
Try to debug a Windows failure on the buildbots.