d41102d | Mark Dickinson | 03 August 2013, 16:14:50 UTC | Remove debugging print. | 03 August 2013, 16:14:50 UTC |
fb0ffa1 | doko@ubuntu.com | 03 August 2013, 14:18:55 UTC | - Fix a fcntl test case on KFreeBSD, Debian #708653 (Petr Salinger). | 03 August 2013, 14:18:55 UTC |
1dfb918 | doko@ubuntu.com | 03 August 2013, 14:12:33 UTC | - Fix a fcntl test case on KFreeBSD, Debian #708653 (Petr Salinger). | 03 August 2013, 14:12:33 UTC |
1337130 | Nick Coghlan | 03 August 2013, 13:03:27 UTC | Merge #18396 from 3.3 | 03 August 2013, 13:03:27 UTC |
60b3ac7 | Nick Coghlan | 03 August 2013, 12:56:30 UTC | Close #18396: fix spurious test_signal failure on Windows signal.getsignal returns None for some signals if faulthandler is enabled (Patch by Jeremy Kloth) | 03 August 2013, 12:56:30 UTC |
02515f7 | Mark Dickinson | 03 August 2013, 11:08:22 UTC | Minor consistency fixes for some longobject.c exception messages: - replace 'long int' / 'long' by 'int' - fix capitalization of "Python" in PyLong_AsUnsignedLong - "is too large" -> "too large", for consistency with other messages. | 03 August 2013, 11:08:22 UTC |
d36fc43 | Larry Hastings | 03 August 2013, 09:49:53 UTC | Fix minor documentation markup error. | 03 August 2013, 09:49:53 UTC |
582583b | Ned Deily | 03 August 2013, 01:05:31 UTC | Issue #17046: merge from 3.3 | 03 August 2013, 01:05:31 UTC |
e92dfbf | Ned Deily | 03 August 2013, 01:02:21 UTC | Issue #17046: Fix test_subprocess test_executable_without_cwd broken test case. | 03 August 2013, 01:02:21 UTC |
95db2e7 | Antoine Pitrou | 02 August 2013, 18:39:46 UTC | Backout 62658d9d8926 (issue #10241): it causes a crash at shutdown when deallocating a Tkapp object. | 02 August 2013, 18:39:46 UTC |
865d12a | Christian Heimes | 02 August 2013, 09:10:51 UTC | Fix warning: ‘ptr2’ may be used uninitialized in this function | 02 August 2013, 09:10:51 UTC |
1d7ec8f | Charles-François Natali | 02 August 2013, 08:22:07 UTC | Issue #18325: Fix a test_kqueue failure on OpenBSD: kevent's data and event members are integers. | 02 August 2013, 08:22:07 UTC |
6ddd265 | Charles-François Natali | 02 August 2013, 08:21:20 UTC | Issue #18325: Fix a test_kqueue failure on OpenBSD: kevent's data and event members are integers. | 02 August 2013, 08:21:20 UTC |
b467d15 | Ned Deily | 02 August 2013, 06:53:24 UTC | Issue #18517: merge from 3.3 | 02 August 2013, 06:53:24 UTC |
cd3d8fb | Ned Deily | 02 August 2013, 06:51:27 UTC | Issue #18517: Move definition of "xxlimited" extension to detect_modules(). | 02 August 2013, 06:51:27 UTC |
970a87f | Ned Deily | 02 August 2013, 05:13:30 UTC | Closes #18627: merge from 3.3 | 02 August 2013, 05:13:30 UTC |
bd143c3 | Ned Deily | 02 August 2013, 05:12:29 UTC | Issue #18627: Fix typo noticed by Vajrasky Kok. | 02 August 2013, 05:12:29 UTC |
ffb3289 | Ned Deily | 02 August 2013, 04:37:17 UTC | Issue #17557: merge from 3.3 | 02 August 2013, 04:37:17 UTC |
b5dd6d2 | Ned Deily | 02 August 2013, 04:21:15 UTC | Issue #17557: Fix os.getgroups() to work with the modified behavior of getgroups(2) on OS X 10.8. Original patch by Mateusz Lenik. | 02 August 2013, 04:21:15 UTC |
2e3e593 | Larry Hastings | 02 August 2013, 02:34:46 UTC | Issue #17899: Fix rare file descriptor leak in os.listdir(). (Done as separate patch from trunk as the code has diverged quite a bit.) | 02 August 2013, 02:34:46 UTC |
4dbc95e | Larry Hastings | 02 August 2013, 01:18:56 UTC | Issue #17899: Fix rare file descriptor leak in os.listdir(). | 02 August 2013, 01:18:56 UTC |
ffff763 | Victor Stinner | 01 August 2013, 23:48:10 UTC | Issue #18519: Fix test_sqlite on old versions of libsqlite3 With old SQLite versions, _sqlite3_result_error() sets a new Python exception, so don't restore the previous exception. | 01 August 2013, 23:48:10 UTC |
044c516 | Victor Stinner | 01 August 2013, 22:47:47 UTC | Issue #18609, #18408: _ctypes_add_traceback() now clears the current exception while adding the traceback, because it may call indirectly a Python function and Python functions must not be called with an exception set. In the case of the issue #18609, _ctypes_add_traceback() called the iso8859-1 decoder which is implemented in Python. Python has a ISO-8859-1 codec implemented in C. It is not used because PyUnicode_Decode() only uses the C codec for other names (aliases) of this codec ("latin-1", "latin1" and "iso-8859-1"). | 01 August 2013, 22:47:47 UTC |
362532b | Ned Deily | 01 August 2013, 22:39:47 UTC | Fix OS X installer build script to handle the pretty-printed _sysconfigdata.py when removing references to temporary build files. This is not an elegant solution but it does isolate changes to just build-installer.py. | 01 August 2013, 22:39:47 UTC |
284e5ce | Antoine Pitrou | 01 August 2013, 22:08:51 UTC | Fix the RLIMIT_CORE resource lowering logic in test_subprocess. We must only lower the soft limit, since lowering the hard limit makes it impossible to raise it back at the end. (this could prevent core dumps from being generated when the test suite crashes) | 01 August 2013, 22:08:51 UTC |
84f31a5 | Antoine Pitrou | 01 August 2013, 20:07:06 UTC | Issue #10241: Clear extension module dict copies at interpreter shutdown. Patch by Neil Schemenauer, minimally modified. | 01 August 2013, 20:07:06 UTC |
a68cbfa | Antoine Pitrou | 01 August 2013, 19:14:43 UTC | Issue #18589: fix hyperlinking of type slots (tp_*) | 01 August 2013, 19:14:43 UTC |
39668f5 | Antoine Pitrou | 01 August 2013, 19:12:45 UTC | Issue #18589: fix hyperlinking of type slots (tp_*) | 01 August 2013, 19:12:45 UTC |
932ff83 | Antoine Pitrou | 01 August 2013, 19:04:50 UTC | Issue #18608: Avoid keeping a strong reference to the locale module inside the _io module. | 01 August 2013, 19:04:50 UTC |
2d350fd | Antoine Pitrou | 01 August 2013, 18:56:12 UTC | Issue #18619: Fix atexit leaking callbacks registered from sub-interpreters, and make it GC-aware. | 01 August 2013, 18:56:12 UTC |
7a2572c | Antoine Pitrou | 01 August 2013, 18:43:26 UTC | test_capi: make a specific test case for the subinterpreter test (it was wrongly classified in the pending calls test case) | 01 August 2013, 18:43:26 UTC |
f93c7b8 | Antoine Pitrou | 01 August 2013, 17:46:04 UTC | Remove Lib/site.py hack to unregister patched builtins. It creates a refleak in subinterpreters, as atexit callbacks aren't triggered at their end. | 01 August 2013, 17:46:04 UTC |
4ed328c | Antoine Pitrou | 01 August 2013, 17:20:31 UTC | Add a test for module weakrefability | 01 August 2013, 17:20:31 UTC |
5284f80 | Martin v. Löwis | 01 August 2013, 16:31:06 UTC | Issue #18569: The installer now adds .py to the PATHEXT variable when extensions are registered. Patch by Paul Moore. | 01 August 2013, 16:31:06 UTC |
85a4847 | Martin v. Löwis | 01 August 2013, 15:34:42 UTC | Add UUID for 3.4. | 01 August 2013, 15:34:42 UTC |
0df35b0 | doko@ubuntu.com | 01 August 2013, 13:32:49 UTC | - Issue #18257: Fix readlink usage in python-config. Install the python version again on Darwin. | 01 August 2013, 13:32:49 UTC |
b2fcebb | Tim Golden | 01 August 2013, 12:58:58 UTC | issue9035: Prevent Windows-specific tests from running on non-Windows platforms | 01 August 2013, 12:58:58 UTC |
6b52806 | Tim Golden | 01 August 2013, 11:44:00 UTC | Issue #9035: os.path.ismount now recognises volumes mounted below a drive root on Windows. Original patch by Atsuo Ishimoto. | 01 August 2013, 11:44:00 UTC |
536ffe1 | R David Murray | 01 August 2013, 00:48:26 UTC | #17616: Improve context manager tests, fix bugs in close method and mode docs. 'mode' docs fix: the file must always be opened in binary in Python3. Bug in Wave_write.close: when the close method calls the check that the header exists and it raises an error, the _file attribute never gets set to None, so the next close tries to close the file again and we get an ignored traceback in the __del__ method. The fix is to set _file to None in a finally clause. This represents a behavior change...in theory a program could be checking for the error on close and then doing a recovery action on the still open file and closing it again. But this change will only go into 3.4, so I think that behavior change is acceptable given that it would be pretty weird and unlikely logic to begin with. | 01 August 2013, 00:48:26 UTC |
abe639f | Christian Heimes | 31 July 2013, 22:12:06 UTC | Initialize utime with 0. It fixes a couple of compiler warnung: warning: 'utime.mtime_ns' may be used uninitialized in this function [-Wmaybe-uninitialized] | 31 July 2013, 22:12:06 UTC |
b3c8724 | Christian Heimes | 31 July 2013, 22:08:16 UTC | Initialize utime with 0. It fixes a couple of compiler warnung: warning: 'utime.mtime_ns' may be used uninitialized in this function [-Wmaybe-uninitialized] | 31 July 2013, 22:08:16 UTC |
6b6a5ab | Christian Heimes | 31 July 2013, 21:50:03 UTC | merge | 31 July 2013, 21:50:03 UTC |
294b93d | Christian Heimes | 31 July 2013, 21:49:48 UTC | merge | 31 July 2013, 21:49:48 UTC |
1289565 | Christian Heimes | 31 July 2013, 21:48:04 UTC | Silence warning about set but unused variable inside compile_atom() in non-debug builds | 31 July 2013, 21:48:04 UTC |
5e4d372 | Christian Heimes | 31 July 2013, 21:47:56 UTC | Silence warning about set but unused variable inside compile_atom() in non-debug builds | 31 July 2013, 21:47:56 UTC |
a4617cd | Antoine Pitrou | 31 July 2013, 21:15:37 UTC | Fix whitespace | 31 July 2013, 21:15:37 UTC |
dcedaf6 | Antoine Pitrou | 31 July 2013, 21:14:08 UTC | Issue #18214: Improve finalization of Python modules to avoid setting their globals to None, in most cases. | 31 July 2013, 21:14:08 UTC |
c27cd71 | Antoine Pitrou | 31 July 2013, 19:54:18 UTC | Merge | 31 July 2013, 19:54:18 UTC |
5c30a75 | Antoine Pitrou | 31 July 2013, 19:52:53 UTC | Issue #15699: The readline module now uses PEP 3121-style module initialization, so as to reclaim allocated resources (Python callbacks) at shutdown. Original patch by Robin Schreiber. | 31 July 2013, 19:52:53 UTC |
c91d5ee | R David Murray | 31 July 2013, 17:46:08 UTC | #17616: wave.open now supports the 'with' statement. Feature and tests by ClClaudiu.Popa, I added the doc changes. | 31 July 2013, 17:46:08 UTC |
0ce642e | Christian Heimes | 31 July 2013, 11:32:40 UTC | yet another WITH_THREADS typo | 31 July 2013, 11:32:40 UTC |
7cd97fc | Christian Heimes | 31 July 2013, 09:58:41 UTC | merge | 31 July 2013, 09:58:41 UTC |
6ffa5ce | Christian Heimes | 31 July 2013, 09:58:18 UTC | Fix _sha3 module to actually release the GIL around its update function. gcov is great. | 31 July 2013, 09:58:18 UTC |
92e2a22 | Terry Jan Reedy | 31 July 2013, 02:31:30 UTC | Merge with 3.3 | 31 July 2013, 02:31:30 UTC |
778cba7 | Terry Jan Reedy | 31 July 2013, 02:31:06 UTC | Issue #18573: More copy-paste fixes to assertWarns entry. | 31 July 2013, 02:31:06 UTC |
4efdb41 | Christian Heimes | 31 July 2013, 00:36:43 UTC | Test Py_IncRef() and Py_DecRef() C functions | 31 July 2013, 00:36:43 UTC |
72f455e | Christian Heimes | 30 July 2013, 23:33:50 UTC | Fix use of uninitialized scalar variable, see 3f994367a979 CID 1058763 | 30 July 2013, 23:33:50 UTC |
49e52f9 | Christian Heimes | 30 July 2013, 22:55:18 UTC | Issue #18481: Add C coverage reporting with gcov and lcov. A new make target "coverage-report" creates an instrumented Python build, runs unit tests and creates a HTML. The report can be updated with "make coverage-lcov". | 30 July 2013, 22:55:18 UTC |
3b998d1 | Ned Deily | 30 July 2013, 21:32:52 UTC | Issue #15494: merge from 3.3 | 30 July 2013, 21:32:52 UTC |
cdc75d9 | Ned Deily | 30 July 2013, 21:30:15 UTC | Issue #15494: Install new test/support directory. | 30 July 2013, 21:30:15 UTC |
bc47d7b | R David Murray | 30 July 2013, 19:53:30 UTC | Merge: #16273: Fix tutorial discussion of seek/tell (opaque text-mode values). | 30 July 2013, 19:53:30 UTC |
1c4e443 | R David Murray | 30 July 2013, 19:51:57 UTC | #16273: Fix tutorial discussion of seek/tell (opaque text-mode values). Patch by Sijin Joseph. | 30 July 2013, 19:51:57 UTC |
f392c60 | R David Murray | 30 July 2013, 19:37:43 UTC | Merge: #18601: fix error made when difflib example was converted to use 'with'. | 30 July 2013, 19:37:43 UTC |
96433f8 | R David Murray | 30 July 2013, 19:37:11 UTC | #18601: fix error made when difflib example was converted to use 'with'. | 30 July 2013, 19:37:11 UTC |
9026262 | Ethan Furman | 30 July 2013, 19:24:25 UTC | fixed examples to work with changed attribute names | 30 July 2013, 19:24:25 UTC |
32d483c | Antoine Pitrou | 30 July 2013, 19:01:23 UTC | Fix style | 30 July 2013, 19:01:23 UTC |
20ac14f | R David Murray | 30 July 2013, 18:43:10 UTC | Merge #18584: s/testcleanup/testsetup/ until we switch to Sphinx 1.1. | 30 July 2013, 18:43:10 UTC |
11bfd32 | R David Murray | 30 July 2013, 18:42:40 UTC | #18584: s/testcleanup/testsetup/ until we switch to Sphinx 1.1. testcleanup directive is new as of 1.1, and we are currently running 1.0.7. But using testsetup works just as well, and avoids the unknown directive error when building the docs. | 30 July 2013, 18:42:40 UTC |
8547a0e | Antoine Pitrou | 30 July 2013, 18:09:36 UTC | Merge doc fix | 30 July 2013, 18:09:36 UTC |
3ab7b0a | Antoine Pitrou | 30 July 2013, 18:09:03 UTC | Simplify example of PyErr_Fetch() use | 30 July 2013, 18:09:03 UTC |
796564c | Antoine Pitrou | 30 July 2013, 17:59:21 UTC | Issue #18112: PEP 442 implementation (safe object finalization). | 30 July 2013, 17:59:21 UTC |
c5d95b1 | Christian Heimes | 30 July 2013, 13:54:39 UTC | Add simple test for repr(lock) | 30 July 2013, 13:54:39 UTC |
8ad91cc | Christian Heimes | 30 July 2013, 13:44:24 UTC | Add simple test for resource.getpagesize() | 30 July 2013, 13:44:24 UTC |
4ebc929 | Christian Heimes | 30 July 2013, 13:44:13 UTC | Add simple test for resource.getpagesize() | 30 July 2013, 13:44:13 UTC |
d49a371 | Christian Heimes | 30 July 2013, 13:35:54 UTC | Issue #18599: Fix name attribute of _sha1.sha1() object. It now returns 'SHA1' instead of 'SHA'. Add more tests for hashlib and hash object attributes | 30 July 2013, 13:35:54 UTC |
65aa573 | Christian Heimes | 30 July 2013, 13:33:30 UTC | Add more tests for hashlib and hash object attributes | 30 July 2013, 13:33:30 UTC |
4fec431 | Christian Heimes | 30 July 2013, 13:32:57 UTC | Issue #18599: Fix name attribute of _sha1.sha1() object. It now returns 'SHA1' instead of 'SHA'. | 30 July 2013, 13:32:57 UTC |
580373c | Eli Bendersky | 30 July 2013, 13:13:01 UTC | Issue #18593: fix typo in comment | 30 July 2013, 13:13:01 UTC |
25f043b | Eli Bendersky | 30 July 2013, 13:12:49 UTC | Issue #18593: fix typo in comment | 30 July 2013, 13:12:49 UTC |
5bcbdf1 | Terry Jan Reedy | 30 July 2013, 05:37:28 UTC | Merge with 3.3 | 30 July 2013, 05:37:28 UTC |
4bd3129 | Terry Jan Reedy | 30 July 2013, 05:36:48 UTC | Make all idle test case names end with 'Test'. | 30 July 2013, 05:36:48 UTC |
299a221 | R David Murray | 29 July 2013, 19:56:26 UTC | Merge #18584: make doctest examples in email documentation pass. | 29 July 2013, 19:56:26 UTC |
fdfb005 | R David Murray | 29 July 2013, 19:49:58 UTC | #18584: make doctest examples in email documentation pass. With the exception of the 'as_string' call in the policy docs. That one is a separate issue. Note that when building the docs sphinx is complaining about .. testcleanup:: being an invalid directive. I don't know why this is, as I'm following the sphinx docs...but fortunately the action is to omit the text in the body, so the generated documentation is correct. | 29 July 2013, 19:49:58 UTC |
00dc61c | Nick Coghlan | 29 July 2013, 07:52:21 UTC | Merge #15415 from 3.3 (again) | 29 July 2013, 07:52:21 UTC |
c06c0ae | Nick Coghlan | 29 July 2013, 07:51:16 UTC | Third attempt to fix #15415 on Windows With help from jkloth on IRC, so it will hopefully work this time :) | 29 July 2013, 07:51:16 UTC |
4bcb1ec | Nick Coghlan | 29 July 2013, 05:19:29 UTC | Merge #15415 Windows buildbot change from 3.3 | 29 July 2013, 05:19:29 UTC |
2f253e8 | Nick Coghlan | 29 July 2013, 05:18:09 UTC | Attempt to fix #15415 on Windows | 29 July 2013, 05:18:09 UTC |
7e6cd29 | Terry Jan Reedy | 28 July 2013, 20:40:07 UTC | Merge with 3.3 | 28 July 2013, 20:40:07 UTC |
015b453 | Terry Jan Reedy | 28 July 2013, 20:39:44 UTC | Update gui section of idle test README. | 28 July 2013, 20:39:44 UTC |
a315a97 | Terry Jan Reedy | 28 July 2013, 20:25:52 UTC | Merge with 3.3 | 28 July 2013, 20:25:52 UTC |
a9d16c8 | Terry Jan Reedy | 28 July 2013, 20:25:16 UTC | Issue #18441: Remove check from test_text.py in 3.3,4 (already done in 2.7). | 28 July 2013, 20:25:16 UTC |
cf67b2a | Nick Coghlan | 28 July 2013, 12:25:25 UTC | Merge #15415 from 3.3 | 28 July 2013, 12:25:25 UTC |
5517596 | Nick Coghlan | 28 July 2013, 12:11:50 UTC | Close #15415: Factor out temp dir helpers to test.support Patch by Chris Jerdonek | 28 July 2013, 12:11:50 UTC |
b8de598 | Nick Coghlan | 28 July 2013, 11:40:20 UTC | Null merge from 3.3 | 28 July 2013, 11:40:20 UTC |
69e3bda | Nick Coghlan | 28 July 2013, 11:06:50 UTC | Issue #15494: test.support is now a package rather than a module Also including this change in 3.3 to help avoid spurious conflicts between the two most active branches. (Initial patch by Indra Talip) | 28 July 2013, 11:06:50 UTC |
fb15aa1 | Nick Coghlan | 28 July 2013, 10:56:19 UTC | Close #15494: test.support is now a package rather than a module Initial patch by Indra Talip | 28 July 2013, 10:56:19 UTC |
e8c45d6 | Nick Coghlan | 28 July 2013, 10:00:01 UTC | Close #13266: Add inspect.unwrap Initial patch by Daniel Urban and Aaron Iles | 28 July 2013, 10:00:01 UTC |
7757820 | Raymond Hettinger | 28 July 2013, 09:39:49 UTC | Restore the data block size to 62. The former block size traded away good fit within cache lines in order to gain faster division in deque_item(). However, compilers are getting smarter and can now replace the slow division operation with a fast integer multiply and right shift. Accordingly, it makes sense to go back to a size that lets blocks neatly fill entire cache-lines. GCC-4.8 and CLANG 4.0 both compute "x // 62" with something roughly equivalent to "x * 9520900167075897609 >> 69". | 28 July 2013, 09:39:49 UTC |
1f1d0a5 | Terry Jan Reedy | 28 July 2013, 04:00:47 UTC | Merge with 3.3 | 28 July 2013, 04:00:47 UTC |
725d9dd | Terry Jan Reedy | 28 July 2013, 04:00:20 UTC | Issue #18441: Make test.support.requires('gui') skip when it should. | 28 July 2013, 04:00:20 UTC |