]>
granicus.if.org Git - python/log
Serhiy Storchaka [Fri, 11 Nov 2016 10:07:48 +0000 (12:07 +0200)]
Issue #19398: Extra slash no longer added to sys.path components in case of
empty compile-time PYTHONPATH components. This fixes some tests in -S or -I
modes.
Serhiy Storchaka [Fri, 11 Nov 2016 10:06:38 +0000 (12:06 +0200)]
Issue #19398: Extra slash no longer added to sys.path components in case of
empty compile-time PYTHONPATH components. This fixes some tests in -S or -I
modes.
Serhiy Storchaka [Fri, 11 Nov 2016 10:05:01 +0000 (12:05 +0200)]
Issue #19398: Extra slash no longer added to sys.path components in case of
empty compile-time PYTHONPATH components. This fixes some tests in -S or -I
modes.
Serhiy Storchaka [Fri, 11 Nov 2016 09:48:22 +0000 (11:48 +0200)]
Issue #23839: Various caches now are cleared before running every test file.
Serhiy Storchaka [Fri, 11 Nov 2016 09:46:44 +0000 (11:46 +0200)]
Issue #23839: Various caches now are cleared before running every test file.
Serhiy Storchaka [Fri, 11 Nov 2016 09:42:47 +0000 (11:42 +0200)]
Issue #23839: Various caches now are cleared before running every test file.
Victor Stinner [Fri, 11 Nov 2016 01:13:35 +0000 (02:13 +0100)]
Issue #28618: Make hot functions using __attribute__((hot))
When Python is not compiled with PGO, the performance of Python on call_simple
and call_method microbenchmarks depend highly on the code placement. In the
worst case, the performance slowdown can be up to 70%.
The GCC __attribute__((hot)) attribute helps to keep hot code close to reduce
the risk of such major slowdown. This attribute is ignored when Python is
compiled with PGO.
The following functions are considered as hot according to statistics collected
by perf record/perf report:
* _PyEval_EvalFrameDefault()
* call_function()
* _PyFunction_FastCall()
* PyFrame_New()
* frame_dealloc()
* PyErr_Occurred()
Victor Stinner [Fri, 11 Nov 2016 00:43:56 +0000 (01:43 +0100)]
Use PyThreadState_GET() in performance critical code
It seems like _PyThreadState_UncheckedGet() is not inlined as expected, even
when using gcc -O3.
Yury Selivanov [Thu, 10 Nov 2016 20:39:46 +0000 (15:39 -0500)]
Merge 3.6 (issue #28635)
Yury Selivanov [Thu, 10 Nov 2016 20:39:27 +0000 (15:39 -0500)]
Issue #28635: what's new in 3.6: remove mentions of backported fixes.
Patch by Elvis Pranskevichus.
Yury Selivanov [Thu, 10 Nov 2016 18:28:03 +0000 (13:28 -0500)]
Merge 3.6 (issue #28635)
Yury Selivanov [Thu, 10 Nov 2016 18:27:22 +0000 (13:27 -0500)]
Issue #28635: What's New in Python 3.6 updates
Patch by Elvis Pranskevichus.
Yury Selivanov [Thu, 10 Nov 2016 18:26:06 +0000 (13:26 -0500)]
Merge 3.6 (issue #28635)
Yury Selivanov [Thu, 10 Nov 2016 18:25:26 +0000 (13:25 -0500)]
Issue #28635: Fix a couple of missing/incorrect versionchanged tags
Patch by Elvis Pranskevichus.
Guido van Rossum [Thu, 10 Nov 2016 16:29:20 +0000 (08:29 -0800)]
Issue #28649: fix second issue with _ForwardRef (#328) (3.6->3.7)
Guido van Rossum [Thu, 10 Nov 2016 16:29:19 +0000 (08:29 -0800)]
Issue #28649: fix second issue with _ForwardRef (#328) (3.5->3.6)
Guido van Rossum [Thu, 10 Nov 2016 16:29:18 +0000 (08:29 -0800)]
Issue #28649: fix second issue with _ForwardRef (#328)
Guido van Rossum [Thu, 10 Nov 2016 16:27:50 +0000 (08:27 -0800)]
Issue #28649: fix first issue with _ForwardRef (#327) (3.6->3.7)
Guido van Rossum [Thu, 10 Nov 2016 16:26:19 +0000 (08:26 -0800)]
Issue #28649: fix first issue with _ForwardRef (#327) (3.5->3.6)
Guido van Rossum [Thu, 10 Nov 2016 16:24:06 +0000 (08:24 -0800)]
Issue #28649: fix first issue with _ForwardRef (#327)
Yury Selivanov [Wed, 9 Nov 2016 23:57:00 +0000 (18:57 -0500)]
Merge 3.6 (issue #28653)
Yury Selivanov [Wed, 9 Nov 2016 23:56:26 +0000 (18:56 -0500)]
Merge 3.6 (issue #28653)
Yury Selivanov [Wed, 9 Nov 2016 23:55:45 +0000 (18:55 -0500)]
Issue #28653: Fix a refleak in functools.lru_cache.
Serhiy Storchaka [Wed, 9 Nov 2016 21:52:40 +0000 (23:52 +0200)]
Issue #28649: Clear the typing module caches when search for reference leaks.
Serhiy Storchaka [Wed, 9 Nov 2016 21:51:54 +0000 (23:51 +0200)]
Issue #28649: Clear the typing module caches when search for reference leaks.
Serhiy Storchaka [Wed, 9 Nov 2016 21:49:02 +0000 (23:49 +0200)]
Issue #28649: Clear the typing module caches when search for reference leaks.
Guido van Rossum [Wed, 9 Nov 2016 21:23:06 +0000 (13:23 -0800)]
Issue #28649: typing-clear-caches.patch (3.6->3.7)
Guido van Rossum [Wed, 9 Nov 2016 21:23:05 +0000 (13:23 -0800)]
Issue #28649: typing-clear-caches.patch (3.5->3.6)
Guido van Rossum [Wed, 9 Nov 2016 21:23:04 +0000 (13:23 -0800)]
Issue #28649: typing-clear-caches.patch
Guido van Rossum [Wed, 9 Nov 2016 21:19:01 +0000 (13:19 -0800)]
Issue #28649: fix-typing-test-v2.diff (3.6->3.7)
Guido van Rossum [Wed, 9 Nov 2016 21:19:00 +0000 (13:19 -0800)]
Issue #28649: fix-typing-test-v2.diff (3.5->3.6)
Guido van Rossum [Wed, 9 Nov 2016 21:18:59 +0000 (13:18 -0800)]
Issue #28649: fix-typing-test-v2.diff
Guido van Rossum [Wed, 9 Nov 2016 21:12:56 +0000 (13:12 -0800)]
Issue #28556: More typing.py updates from upstream. (3.6->3.7)
Guido van Rossum [Wed, 9 Nov 2016 21:12:53 +0000 (13:12 -0800)]
Issue #28556: More typing.py updates from upstream. (3.5->3.6)
Guido van Rossum [Wed, 9 Nov 2016 21:12:51 +0000 (13:12 -0800)]
Issue #28556: More typing.py updates from upstream.
Steve Dower [Wed, 9 Nov 2016 20:58:31 +0000 (12:58 -0800)]
Issue #19717: Makes Path.resolve() succeed on paths that do not exist (patch by Vajrasky Kok)
Steve Dower [Wed, 9 Nov 2016 20:58:17 +0000 (12:58 -0800)]
Issue #19717: Makes Path.resolve() succeed on paths that do not exist (patch by Vajrasky Kok)
Yury Selivanov [Wed, 9 Nov 2016 20:48:22 +0000 (15:48 -0500)]
Merge 3.6 (issue #28652)
Yury Selivanov [Wed, 9 Nov 2016 20:47:47 +0000 (15:47 -0500)]
Merge 3.5 (issue #28652)
Yury Selivanov [Wed, 9 Nov 2016 20:47:00 +0000 (15:47 -0500)]
Issue #28652: Make loop methods reject socket kinds they do not support.
Yury Selivanov [Wed, 9 Nov 2016 14:43:37 +0000 (09:43 -0500)]
Merge 3.6 (issue #27942)
Yury Selivanov [Wed, 9 Nov 2016 14:43:08 +0000 (09:43 -0500)]
Merge 3.5 (issue #27942)
Yury Selivanov [Wed, 9 Nov 2016 14:42:14 +0000 (09:42 -0500)]
ssue #27942: Fix memory leak in codeobject.c
Andrew Kuchling [Wed, 9 Nov 2016 03:53:59 +0000 (22:53 -0500)]
Merge 3.6
Andrew Kuchling [Wed, 9 Nov 2016 03:38:54 +0000 (22:38 -0500)]
Merge from 3.5
Andrew Kuchling [Wed, 9 Nov 2016 03:33:31 +0000 (22:33 -0500)]
Use http instead of https, due to certificate error
Andrew Kuchling [Wed, 9 Nov 2016 03:14:39 +0000 (22:14 -0500)]
Merge from 3.5
Andrew Kuchling [Wed, 9 Nov 2016 03:07:43 +0000 (22:07 -0500)]
Update URL for requests documentation after IRC request
Yury Selivanov [Wed, 9 Nov 2016 00:59:57 +0000 (19:59 -0500)]
Merge 3.6 (issue #28639)
Yury Selivanov [Wed, 9 Nov 2016 00:59:29 +0000 (19:59 -0500)]
Merge 3.5 (issue #28639)
Yury Selivanov [Wed, 9 Nov 2016 00:57:44 +0000 (19:57 -0500)]
Issue #28639: Fix inspect.isawaitable to always return bool
Patch by Justin Mayfield.
Yury Selivanov [Wed, 9 Nov 2016 00:46:41 +0000 (19:46 -0500)]
Merge 3.6 (issue #28003)
Yury Selivanov [Wed, 9 Nov 2016 00:46:22 +0000 (19:46 -0500)]
Issue #28003: Make WrappedVal, ASend and AThrow GC types
Yury Selivanov [Wed, 9 Nov 2016 00:20:08 +0000 (19:20 -0500)]
Merge 3.6
Yury Selivanov [Wed, 9 Nov 2016 00:19:52 +0000 (19:19 -0500)]
Merge 3.5
Yury Selivanov [Wed, 9 Nov 2016 00:19:28 +0000 (19:19 -0500)]
genobject: Remove unnecessary tp_free slots from aiter_wrapper and coro_wrapper
Yury Selivanov [Wed, 9 Nov 2016 00:16:37 +0000 (19:16 -0500)]
Merge 3.6 (asyncio)
Yury Selivanov [Wed, 9 Nov 2016 00:16:15 +0000 (19:16 -0500)]
Merge 3.5 (asyncio)
Yury Selivanov [Wed, 9 Nov 2016 00:16:01 +0000 (19:16 -0500)]
asyncio: Fix _format_coroutine for coroutine-like objects w/o __name__
Some built-in coroutine-like objects might not have __name__ or
__qualname__. A good example of such are 'asend', 'aclose' and
'athrow' coroutine methods of asynchronous generators.
Yury Selivanov [Wed, 9 Nov 2016 00:05:42 +0000 (19:05 -0500)]
Merge 3.6 (issue #26081)
Yury Selivanov [Wed, 9 Nov 2016 00:04:57 +0000 (19:04 -0500)]
Issue #26081: Fix refleak in _asyncio.Future.__iter__().throw.
Yury Selivanov [Tue, 8 Nov 2016 21:57:40 +0000 (16:57 -0500)]
Merge 3.6 (typo in NEWS)
Yury Selivanov [Tue, 8 Nov 2016 21:57:22 +0000 (16:57 -0500)]
news: Fix a typo
Yury Selivanov [Tue, 8 Nov 2016 21:54:39 +0000 (16:54 -0500)]
Merge 3.6 (issue #26182)
Yury Selivanov [Tue, 8 Nov 2016 21:54:18 +0000 (16:54 -0500)]
Issue #26182: Fix ia refleak in code that raises DeprecationWarning.
Serhiy Storchaka [Tue, 8 Nov 2016 21:14:00 +0000 (23:14 +0200)]
Fixed possible abort in ceval loop if _PyUnicode_FromId() fails.
Every opcode should end with DISPATCH() or goto error.
Serhiy Storchaka [Tue, 8 Nov 2016 21:13:36 +0000 (23:13 +0200)]
Fixed possible abort in ceval loop if _PyUnicode_FromId() fails.
Every opcode should end with DISPATCH() or goto error.
Serhiy Storchaka [Tue, 8 Nov 2016 21:12:46 +0000 (23:12 +0200)]
Fixed possible abort in ceval loop if _PyUnicode_FromId() fails.
Every opcode should end with DISPATCH() or goto error.
Yury Selivanov [Tue, 8 Nov 2016 20:15:42 +0000 (15:15 -0500)]
Merge 3.6 (issue #27243)
Yury Selivanov [Tue, 8 Nov 2016 20:13:07 +0000 (15:13 -0500)]
Issue #27243: Change PendingDeprecationWarning -> DeprecationWarning.
As it was agreed in the issue, __aiter__ returning an awaitable
should result in PendingDeprecationWarning in 3.5 and in
DeprecationWarning in 3.6.
Serhiy Storchaka [Tue, 8 Nov 2016 19:28:30 +0000 (21:28 +0200)]
Issue #28563: Fixed possible DoS and arbitrary code execution when handle
plural form selections in the gettext module. The expression parser now
supports exact syntax supported by GNU gettext.
Serhiy Storchaka [Tue, 8 Nov 2016 19:27:17 +0000 (21:27 +0200)]
Issue #28563: Fixed possible DoS and arbitrary code execution when handle
plural form selections in the gettext module. The expression parser now
supports exact syntax supported by GNU gettext.
Serhiy Storchaka [Tue, 8 Nov 2016 19:26:14 +0000 (21:26 +0200)]
Issue #28563: Fixed possible DoS and arbitrary code execution when handle
plural form selections in the gettext module. The expression parser now
supports exact syntax supported by GNU gettext.
Serhiy Storchaka [Tue, 8 Nov 2016 19:20:09 +0000 (21:20 +0200)]
Issue #28563: Fixed possible DoS and arbitrary code execution when handle
plural form selections in the gettext module. The expression parser now
supports exact syntax supported by GNU gettext.
Serhiy Storchaka [Tue, 8 Nov 2016 19:17:46 +0000 (21:17 +0200)]
Issue #28563: Fixed possible DoS and arbitrary code execution when handle
plural form selections in the gettext module. The expression parser now
supports exact syntax supported by GNU gettext.
Serhiy Storchaka [Tue, 8 Nov 2016 18:34:22 +0000 (20:34 +0200)]
Issue #28621: Sped up converting int to float by reusing faster bits counting
implementation. Patch by Adrian Wielgosik.
Serhiy Storchaka [Tue, 8 Nov 2016 18:28:43 +0000 (20:28 +0200)]
Issue #28585: Restored docstring of os._isdir().
Serhiy Storchaka [Tue, 8 Nov 2016 18:26:18 +0000 (20:26 +0200)]
Issue #28585: Restored docstring of os._isdir().
Serhiy Storchaka [Tue, 8 Nov 2016 18:21:22 +0000 (20:21 +0200)]
Issue #28585: Restored docstring of os._isdir().
Serhiy Storchaka [Tue, 8 Nov 2016 18:18:20 +0000 (20:18 +0200)]
Issue #28637: No longer use re in site.py.
This makes Python startup from a virtual environment a little faster.
Serhiy Storchaka [Tue, 8 Nov 2016 18:17:35 +0000 (20:17 +0200)]
Issue #28637: No longer use re in site.py.
This makes Python startup from a virtual environment a little faster.
Yury Selivanov [Tue, 8 Nov 2016 17:23:26 +0000 (12:23 -0500)]
Merge 3.6 (docs/inspect)
Yury Selivanov [Tue, 8 Nov 2016 17:23:09 +0000 (12:23 -0500)]
docs/inspect: clarify iscoroutinefunction; add docs for isasyncgen*
Terry Jan Reedy [Tue, 8 Nov 2016 04:15:10 +0000 (23:15 -0500)]
Merge with 3.6
Terry Jan Reedy [Tue, 8 Nov 2016 04:14:53 +0000 (23:14 -0500)]
Issue #28572: Use system-specific values for configdialog font test
Yury Selivanov [Tue, 8 Nov 2016 00:01:27 +0000 (19:01 -0500)]
Merge 3.6 (issue #28613)
Yury Selivanov [Tue, 8 Nov 2016 00:01:14 +0000 (19:01 -0500)]
Merge 3.5 (issue #28613)
Yury Selivanov [Tue, 8 Nov 2016 00:00:46 +0000 (19:00 -0500)]
Issue #28613: Expose asyncio._get_running_loop() and _set_running_loop()
Victor Stinner [Mon, 7 Nov 2016 23:44:53 +0000 (00:44 +0100)]
Merge 3.6
Victor Stinner [Mon, 7 Nov 2016 23:42:46 +0000 (00:42 +0100)]
Issue #28637: Revert issue #28082, don't import enum in re
Importing the enum module in the re module slows down Python startup by 34%
when Python is run from a virtual environment, or more generally when the re
module is imported at startup but not the enum module.
Eric V. Smith [Mon, 7 Nov 2016 22:57:48 +0000 (17:57 -0500)]
Merge from 3.6.
Eric V. Smith [Mon, 7 Nov 2016 22:54:01 +0000 (17:54 -0500)]
Fixed issue #28633: segfault when concatenating bytes literal and f-string.
Terry Jan Reedy [Mon, 7 Nov 2016 22:15:16 +0000 (17:15 -0500)]
Merge with 3.6
Terry Jan Reedy [Mon, 7 Nov 2016 22:15:01 +0000 (17:15 -0500)]
Issue #28572: Add 10% to coverage of IDLE's test_configdialog.
Update and augment description of the configuration system.
Yury Selivanov [Mon, 7 Nov 2016 21:44:27 +0000 (16:44 -0500)]
Merge 3.6 (issue #28635)
Yury Selivanov [Mon, 7 Nov 2016 21:40:20 +0000 (16:40 -0500)]
whatsnew: Inital pass on "What's New in Python 3.6"
Patch by Elvis Pranskevichus.
Yury Selivanov [Mon, 7 Nov 2016 21:07:58 +0000 (16:07 -0500)]
Merge 3.6 (issue #28634)
Yury Selivanov [Mon, 7 Nov 2016 21:07:30 +0000 (16:07 -0500)]
Merge 3.5 (issue #28634)
Yury Selivanov [Mon, 7 Nov 2016 21:00:50 +0000 (16:00 -0500)]
Issue #28634: Fix asyncio.isfuture() to support mocks
Yury Selivanov [Mon, 7 Nov 2016 20:36:03 +0000 (15:36 -0500)]
Merge 3.6 (issue #27392)