]> granicus.if.org Git - python/log
python
10 years agoIssue #21704: Merge.
Richard Oudkerk [Mon, 28 Jul 2014 22:13:58 +0000 (23:13 +0100)]
Issue #21704: Merge.

10 years ago(Merge 3.4) Fix test_bytes when sys.stdin is None, for example on Windows when
Victor Stinner [Mon, 28 Jul 2014 22:07:56 +0000 (00:07 +0200)]
(Merge 3.4) Fix test_bytes when sys.stdin is None, for example on Windows when
using pythonw.exe instead of python.exe

10 years agoIssue #21704: Fix build error for _multiprocessing when semaphores
Richard Oudkerk [Mon, 28 Jul 2014 22:01:02 +0000 (23:01 +0100)]
Issue #21704: Fix build error for _multiprocessing when semaphores
are not available.  Patch by Arfrever Frehtes Taifersar Arahesis.

10 years agoFix test_bytes when sys.stdin is None, for example on Windows when using
Victor Stinner [Mon, 28 Jul 2014 20:07:07 +0000 (22:07 +0200)]
Fix test_bytes when sys.stdin is None, for example on Windows when using
pythonw.exe instead of python.exe

10 years ago(Merge 3.4) asyncio, tulip issue 196: ProactorIocp._register() now registers
Victor Stinner [Sun, 27 Jul 2014 22:19:36 +0000 (00:19 +0200)]
(Merge 3.4) asyncio, tulip issue 196: ProactorIocp._register() now registers
the overlapped in the _cache dictionary, even if we already got the result. We
need to keep a reference to the overlapped object, otherwise the memory may be
reused and GetQueuedCompletionStatus() may use random bytes and behaves badly.

There is still a hack for ConnectNamedPipe(): the overlapped object is not
register into _cache if the overlapped object completed directly.

Log also an error in debug mode in ProactorIocp._loop() if we get an unexpected
event.

Add a protection in ProactorIocp.close() to avoid blocking, even if it should
not happen. I still don't understand exactly why some the completion of some
overlapped objects are not notified.

10 years agoasyncio, tulip issue 196: ProactorIocp._register() now registers the overlapped
Victor Stinner [Sun, 27 Jul 2014 22:18:43 +0000 (00:18 +0200)]
asyncio, tulip issue 196: ProactorIocp._register() now registers the overlapped
in the _cache dictionary, even if we already got the result. We need to keep a
reference to the overlapped object, otherwise the memory may be reused and
GetQueuedCompletionStatus() may use random bytes and behaves badly.

There is still a hack for ConnectNamedPipe(): the overlapped object is not
register into _cache if the overlapped object completed directly.

Log also an error in debug mode in ProactorIocp._loop() if we get an unexpected
event.

Add a protection in ProactorIocp.close() to avoid blocking, even if it should
not happen. I still don't understand exactly why some the completion of some
overlapped objects are not notified.

10 years agoIssue #22076: Minor grammar fix.
Berker Peksag [Sun, 27 Jul 2014 20:22:34 +0000 (23:22 +0300)]
Issue #22076: Minor grammar fix.

Patch by Martin Matusiak.

10 years agoRerun AC, silence pointer conversion warnings.
Martin v. Löwis [Sun, 27 Jul 2014 15:29:17 +0000 (17:29 +0200)]
Rerun AC, silence pointer conversion warnings.

10 years agoIssue #20179: Apply Argument Clinic to bytes and bytearray.
Martin v. Löwis [Sun, 27 Jul 2014 14:25:09 +0000 (16:25 +0200)]
Issue #20179: Apply Argument Clinic to bytes and bytearray.
Patch by Tal Einat.

10 years agoArgument Clinic bugfix: Don't let the C preprocessor "Monitor" see lines
Larry Hastings [Sun, 27 Jul 2014 14:22:20 +0000 (16:22 +0200)]
Argument Clinic bugfix: Don't let the C preprocessor "Monitor" see lines
that we are scanning for the output marker.  If we don't find it, we will
scan them again, so it sees them twice, and it can get confused (like
thinking we're still in a comment).

10 years agoIssue #22082: Fix a compiler warning (function is not a prototype)
Victor Stinner [Sun, 27 Jul 2014 14:11:30 +0000 (16:11 +0200)]
Issue #22082: Fix a compiler warning (function is not a prototype)

10 years agoIssue #20173: Convert sha1, sha256, sha512 and md5 to ArgumentClinic.
Martin v. Löwis [Sun, 27 Jul 2014 12:20:23 +0000 (14:20 +0200)]
Issue #20173: Convert sha1, sha256, sha512 and md5 to ArgumentClinic.
Patch by Vajrasky Kok.

10 years agoMerge with 3.4
Terry Jan Reedy [Sun, 27 Jul 2014 07:01:29 +0000 (03:01 -0400)]
Merge with 3.4

10 years agoModernize turtledemo with conditional expressions; remove duplicate line.
Terry Jan Reedy [Sun, 27 Jul 2014 07:01:13 +0000 (03:01 -0400)]
Modernize turtledemo with conditional expressions; remove duplicate line.

10 years agoMerge with 3.4
Terry Jan Reedy [Sat, 26 Jul 2014 23:40:31 +0000 (19:40 -0400)]
Merge with 3.4

10 years agoReduce unnecessary 2.7 versus 3.4+ differences in idlelib/configDialog.py.
Terry Jan Reedy [Sat, 26 Jul 2014 23:40:16 +0000 (19:40 -0400)]
Reduce  unnecessary 2.7 versus 3.4+ differences in idlelib/configDialog.py.
2009-02-02 51562 (27c589f09a21) merged Mac support revisions into 3.0.
2009-03-04 52014 (dc3d24824b60) merged slightly improved version into 2.7.
Change 3.4+ to match the later 2.7 patch.  Buttons tested manually.

10 years agoDocument lock parameter for asyncio.Condition()
Andrew Svetlov [Sat, 26 Jul 2014 16:53:38 +0000 (19:53 +0300)]
Document lock parameter for asyncio.Condition()

10 years agoDocument lock parameter for asyncio.Condition()
Andrew Svetlov [Sat, 26 Jul 2014 16:50:37 +0000 (19:50 +0300)]
Document lock parameter for asyncio.Condition()

10 years agoIssue #22074: Fix Lib/test/make_ssl_certs.py
Antoine Pitrou [Sat, 26 Jul 2014 15:18:05 +0000 (11:18 -0400)]
Issue #22074: Fix Lib/test/make_ssl_certs.py

10 years agoIssue #22074: Fix Lib/test/make_ssl_certs.py
Antoine Pitrou [Sat, 26 Jul 2014 15:15:52 +0000 (11:15 -0400)]
Issue #22074: Fix Lib/test/make_ssl_certs.py

10 years agoAccept optional lock object in Condition ctor (tulip issue #198)
Andrew Svetlov [Sat, 26 Jul 2014 14:54:56 +0000 (17:54 +0300)]
Accept optional lock object in Condition ctor (tulip issue #198)

10 years agoAccept optional lock object in Condition ctor (tulip issue #198)
Andrew Svetlov [Sat, 26 Jul 2014 14:54:34 +0000 (17:54 +0300)]
Accept optional lock object in Condition ctor (tulip issue #198)

10 years agoIssue #22082: Clear interned strings in slotdefs.
Martin v. Löwis [Sat, 26 Jul 2014 14:44:07 +0000 (16:44 +0200)]
Issue #22082: Clear interned strings in slotdefs.

10 years agoslotdefs is no longer sorted.
Martin v. Löwis [Sat, 26 Jul 2014 13:25:04 +0000 (15:25 +0200)]
slotdefs is no longer sorted.

10 years ago(Merge 3.4) Fix repr(_socket.socket) on Windows 64-bit: don't fail with
Victor Stinner [Sat, 26 Jul 2014 12:37:57 +0000 (14:37 +0200)]
(Merge 3.4) Fix repr(_socket.socket) on Windows 64-bit: don't fail with
OverflowError on closed socket. repr(socket.socket) already works fine.

10 years agoFix repr(_socket.socket) on Windows 64-bit: don't fail with OverflowError
Victor Stinner [Sat, 26 Jul 2014 12:36:55 +0000 (14:36 +0200)]
Fix repr(_socket.socket) on Windows 64-bit: don't fail with OverflowError
on closed socket. repr(socket.socket) already works fine.

10 years ago(Merge 3.4) Tulip issue 196: _OverlappedFuture.set_result() now clears its
Victor Stinner [Fri, 25 Jul 2014 22:58:55 +0000 (00:58 +0200)]
(Merge 3.4) Tulip issue 196: _OverlappedFuture.set_result() now clears its
reference to the overlapped object. IocpProactor._poll() now also ignores false
alarms: GetQueuedCompletionStatus() returns the overlapped but it is still
pending.

10 years agoTulip issue 196: _OverlappedFuture.set_result() now clears its reference to the
Victor Stinner [Fri, 25 Jul 2014 22:58:34 +0000 (00:58 +0200)]
Tulip issue 196: _OverlappedFuture.set_result() now clears its reference to the
overlapped object. IocpProactor._poll() now also ignores false alarms:
GetQueuedCompletionStatus() returns the overlapped but it is still pending.

10 years agomerge
Raymond Hettinger [Fri, 25 Jul 2014 22:00:30 +0000 (15:00 -0700)]
merge

10 years agoIssue #22044: Fixed premature DECREF in call_tzinfo_method.
Raymond Hettinger [Fri, 25 Jul 2014 21:59:48 +0000 (14:59 -0700)]
Issue #22044: Fixed premature DECREF in call_tzinfo_method.

10 years ago(Merge 3.4) asyncio: sync with Tulip
Victor Stinner [Fri, 25 Jul 2014 20:40:12 +0000 (22:40 +0200)]
(Merge 3.4) asyncio: sync with Tulip

* Tulip issue #196: IocpProactor._poll() clears the reference to the
  overlapped operation when the operation is done. It would be better to clear
  the reference in a new _OverlappedFuture.set_result() method, but it cannot
  be done yet because of a weird bug.
* BaseSelectorEventLoop._write_to_self() now logs errors in debug mode.

10 years agoasyncio: sync with Tulip
Victor Stinner [Fri, 25 Jul 2014 20:36:05 +0000 (22:36 +0200)]
asyncio: sync with Tulip

* Tulip issue #196: IocpProactor._poll() clears the reference to the
  overlapped operation when the operation is done. It would be better to clear
  the reference in a new _OverlappedFuture.set_result() method, but it cannot
  be done yet because of a weird bug.
* BaseSelectorEventLoop._write_to_self() now logs errors in debug mode.

10 years agoIssue #22033: Reprs of most Python implemened classes now contain actual
Serhiy Storchaka [Fri, 25 Jul 2014 20:36:00 +0000 (23:36 +0300)]
Issue #22033: Reprs of most Python implemened classes now contain actual
class name instead of hardcoded one.

10 years agoIssue #21958: Merge with 3.4
Zachary Ware [Fri, 25 Jul 2014 19:47:29 +0000 (14:47 -0500)]
Issue #21958: Merge with 3.4

10 years agoIssue #18093: Create Programs directory in build location for
Ned Deily [Fri, 25 Jul 2014 19:41:31 +0000 (12:41 -0700)]
Issue #18093: Create Programs directory in build location for
out-of-tree builds.

10 years agoIssue #21958: Define HAVE_ROUND when building with VS 2013 and above.
Zachary Ware [Fri, 25 Jul 2014 19:34:19 +0000 (14:34 -0500)]
Issue #21958: Define HAVE_ROUND when building with VS 2013 and above.
Patch by Zachary Turner.

10 years agoCloses #22072: Merge typo fixes from 3.4
Zachary Ware [Fri, 25 Jul 2014 18:31:36 +0000 (13:31 -0500)]
Closes #22072: Merge typo fixes from 3.4

10 years agoIssue #22072: Fix a couple of SSL doc typos. Patch by Alex Gaynor.
Zachary Ware [Fri, 25 Jul 2014 18:30:50 +0000 (13:30 -0500)]
Issue #22072: Fix a couple of SSL doc typos.  Patch by Alex Gaynor.

10 years agoIssue #19875: Fix random test_getsockaddrarg() failure.
Charles-François Natali [Fri, 25 Jul 2014 17:45:28 +0000 (18:45 +0100)]
Issue #19875: Fix random test_getsockaddrarg() failure.

10 years agoMerge.
Charles-François Natali [Fri, 25 Jul 2014 17:45:02 +0000 (18:45 +0100)]
Merge.

10 years agoIssue #19875: Fix random test_getsockaddrarg() failure.
Charles-François Natali [Fri, 25 Jul 2014 17:44:30 +0000 (18:44 +0100)]
Issue #19875: Fix random test_getsockaddrarg() failure.

10 years agoIssue #18093: Give the Windows build _testembed.c's new location.
Zachary Ware [Fri, 25 Jul 2014 13:58:07 +0000 (08:58 -0500)]
Issue #18093: Give the Windows build _testembed.c's new location.

10 years agoIssue #21947: handle generator-iterator objects in dis
Nick Coghlan [Fri, 25 Jul 2014 13:02:56 +0000 (23:02 +1000)]
Issue #21947: handle generator-iterator objects in dis

Patch by Clement Rouault.

10 years ago(Merge 3.4) asyncio, test_subprocess: relax timings for slow builbots
Victor Stinner [Fri, 25 Jul 2014 12:05:19 +0000 (14:05 +0200)]
(Merge 3.4) asyncio, test_subprocess: relax timings for slow builbots

10 years agoasyncio, test_subprocess: relax timings for slow builbots
Victor Stinner [Fri, 25 Jul 2014 12:05:07 +0000 (14:05 +0200)]
asyncio, test_subprocess: relax timings for slow builbots

10 years agounicodeobject.c: fix a compiler warning on Windows 64 bits
Victor Stinner [Fri, 25 Jul 2014 12:03:03 +0000 (14:03 +0200)]
unicodeobject.c: fix a compiler warning on Windows 64 bits

10 years agoAdd missing NEWS entry for issue #18093
Nick Coghlan [Fri, 25 Jul 2014 11:55:07 +0000 (21:55 +1000)]
Add missing NEWS entry for issue #18093

10 years agoIssue #18093: Factor out the programs that embed the runtime
Nick Coghlan [Fri, 25 Jul 2014 11:52:14 +0000 (21:52 +1000)]
Issue #18093: Factor out the programs that embed the runtime

10 years ago(Merge 3.4) asyncio: sync with Tulip
Victor Stinner [Fri, 25 Jul 2014 11:05:43 +0000 (13:05 +0200)]
(Merge 3.4) asyncio: sync with Tulip

* Fix _WaitHandleFuture.cancel(): return the result of the parent cancel()
  method.
* _OverlappedFuture.cancel() now clears its reference to the overlapped object.
  Make also the _OverlappedFuture.ov attribute private.
* Check if _WaitHandleFuture completed before unregistering it in the callback.
  Add also _WaitHandleFuture._poll() and repr(_WaitHandleFuture).
* _WaitHandleFuture now unregisters its wait handler if WaitForSingleObject()
  raises an exception.
* _OverlappedFuture.set_exception() now cancels the overlapped operation.

10 years agoasyncio: sync with Tulip
Victor Stinner [Fri, 25 Jul 2014 11:05:20 +0000 (13:05 +0200)]
asyncio: sync with Tulip

* Fix _WaitHandleFuture.cancel(): return the result of the parent cancel()
  method.
* _OverlappedFuture.cancel() now clears its reference to the overlapped object.
  Make also the _OverlappedFuture.ov attribute private.
* Check if _WaitHandleFuture completed before unregistering it in the callback.
  Add also _WaitHandleFuture._poll() and repr(_WaitHandleFuture).
* _WaitHandleFuture now unregisters its wait handler if WaitForSingleObject()
  raises an exception.
* _OverlappedFuture.set_exception() now cancels the overlapped operation.

10 years agoIssue #22061: Restored empty obsolete methods removed in issue #4350 and
Serhiy Storchaka [Fri, 25 Jul 2014 09:29:40 +0000 (12:29 +0300)]
Issue #22061: Restored empty obsolete methods removed in issue #4350 and
added deprecation warnings to them.

10 years agoIssue #22061: Add deprecation warnings in empty obsolete methods.
Serhiy Storchaka [Fri, 25 Jul 2014 09:24:07 +0000 (12:24 +0300)]
Issue #22061: Add deprecation warnings in empty obsolete methods.

10 years agoIssue #22061: remove call of useless function slated for removal.
Terry Jan Reedy [Fri, 25 Jul 2014 07:08:14 +0000 (03:08 -0400)]
Issue #22061: remove call of useless function slated for removal.

10 years agoIssue #22061: remove call of useless function slated for removal.
Terry Jan Reedy [Fri, 25 Jul 2014 07:06:32 +0000 (03:06 -0400)]
Issue #22061: remove call of useless function slated for removal.

10 years agoIssue #22053: Make help work, after previous patch for this issue disabled it
Terry Jan Reedy [Fri, 25 Jul 2014 06:26:40 +0000 (02:26 -0400)]
Issue #22053: Make help work, after previous patch for this issue disabled it
by removing global 'demo'. Simple fix because 2.7/3.4 fix cannot merge.

10 years agoIssue #22053: Make help work, after previous patch for this issue disabled it
Terry Jan Reedy [Fri, 25 Jul 2014 05:56:24 +0000 (01:56 -0400)]
Issue #22053: Make help work, after previous patch for this issue disabled it
by removing global 'demo'. Refactor and remove duplicate code.

10 years ago(Merge 3.4) asyncio: sync with Tulip
Victor Stinner [Thu, 24 Jul 2014 22:55:23 +0000 (00:55 +0200)]
(Merge 3.4) asyncio: sync with Tulip

Improve stability of the proactor event loop, especially operations on
overlapped objects:

* Tulip issue 195: Don't call UnregisterWait() twice if a _WaitHandleFuture is
  cancelled twice to fix a crash.
* IocpProactor.close(): cancel futures to cancel overlapped operations, instead
  of cancelling directly overlapped operations. Future objects may not call
  ov.cancel() if the future was cancelled or if the overlapped was already
  cancelled. The cancel() method of the future may also catch exceptions. Log
  also errors on cancellation.
* tests: rename "f" to "fut"
* Add a __repr__() method to IocpProactor
* Add a destructor to IocpProactor which closes it
* _OverlappedFuture.cancel() doesn't cancel the overlapped anymore if it is
  done: if it is already cancelled or completed. Log also an error if the
  cancellation failed.
* Add the address of the overlapped object in repr(_OverlappedFuture)
* _OverlappedFuture truncates the source traceback to hide the call to the
  parent constructor (useless in debug).

10 years agoasyncio: sync with Tulip
Victor Stinner [Thu, 24 Jul 2014 22:54:53 +0000 (00:54 +0200)]
asyncio: sync with Tulip

Improve stability of the proactor event loop, especially operations on
overlapped objects:

* Tulip issue 195: Don't call UnregisterWait() twice if a _WaitHandleFuture is
  cancelled twice to fix a crash.
* IocpProactor.close(): cancel futures to cancel overlapped operations, instead
  of cancelling directly overlapped operations. Future objects may not call
  ov.cancel() if the future was cancelled or if the overlapped was already
  cancelled. The cancel() method of the future may also catch exceptions. Log
  also errors on cancellation.
* tests: rename "f" to "fut"
* Add a __repr__() method to IocpProactor
* Add a destructor to IocpProactor which closes it
* _OverlappedFuture.cancel() doesn't cancel the overlapped anymore if it is
  done: if it is already cancelled or completed. Log also an error if the
  cancellation failed.
* Add the address of the overlapped object in repr(_OverlappedFuture)
* _OverlappedFuture truncates the source traceback to hide the call to the
  parent constructor (useless in debug).

10 years agoIssue #22018: Fix test_set_wakeup_fd_result(), use assertEqual() not
Victor Stinner [Thu, 24 Jul 2014 20:55:12 +0000 (22:55 +0200)]
Issue #22018: Fix test_set_wakeup_fd_result(), use assertEqual() not
assertIs().

10 years agoBackout 42ced0d023cd: oops, i didn't want to push this changeset :-/
Victor Stinner [Thu, 24 Jul 2014 20:51:05 +0000 (22:51 +0200)]
Backout 42ced0d023cd: oops, i didn't want to push this changeset :-/

10 years ago(Merge 3.4) Issue #19884, readline: calling rl_variable_bind
Victor Stinner [Thu, 24 Jul 2014 20:11:55 +0000 (22:11 +0200)]
(Merge 3.4) Issue #19884, readline: calling rl_variable_bind
("enable-meta-key", "off") does crash on Mac OS X which uses libedit instead of
readline.

10 years agoIssue #19884, readline: calling rl_variable_bind ("enable-meta-key", "off")
Victor Stinner [Thu, 24 Jul 2014 20:11:38 +0000 (22:11 +0200)]
Issue #19884, readline: calling rl_variable_bind ("enable-meta-key", "off")
does crash on Mac OS X which uses libedit instead of readline.

10 years agotets
Victor Stinner [Thu, 24 Jul 2014 19:58:53 +0000 (21:58 +0200)]
tets

10 years ago(Merge 3.4) Issue #16133: The asynchat.async_chat.handle_read() method now
Victor Stinner [Thu, 24 Jul 2014 17:00:28 +0000 (19:00 +0200)]
(Merge 3.4) Issue #16133: The asynchat.async_chat.handle_read() method now
ignores BlockingIOError exceptions. Initial patch written by Xavier de Gaye.

Document also in asyncore documentation that recv() may raise BlockingIOError.

10 years agoIssue #16133: The asynchat.async_chat.handle_read() method now ignores
Victor Stinner [Thu, 24 Jul 2014 16:49:36 +0000 (18:49 +0200)]
Issue #16133: The asynchat.async_chat.handle_read() method now ignores
BlockingIOError exceptions. Initial patch written by Xavier de Gaye.

Document also in asyncore documentation that recv() may raise BlockingIOError.

10 years agoIssue #21813: Enhance documentation of the os.stat_result class.
Victor Stinner [Thu, 24 Jul 2014 10:44:07 +0000 (12:44 +0200)]
Issue #21813: Enhance documentation of the os.stat_result class.

10 years agoBackport os.rst documentation from Python 3.5.
Victor Stinner [Thu, 24 Jul 2014 10:42:45 +0000 (12:42 +0200)]
Backport os.rst documentation from Python 3.5.

10 years agoIssue #21813: Enhance documentation of the os.stat_result class.
Victor Stinner [Thu, 24 Jul 2014 10:42:16 +0000 (12:42 +0200)]
Issue #21813: Enhance documentation of the os.stat_result class.

10 years ago(Merge 3.4) Issue #19884: readline: Disable the meta modifier key if stdout is
Victor Stinner [Thu, 24 Jul 2014 10:24:45 +0000 (12:24 +0200)]
(Merge 3.4) Issue #19884: readline: Disable the meta modifier key if stdout is
not a terminal to not write the ANSI sequence "\033[1034h" into stdout. This
sequence is used on some terminal (ex: TERM=xterm-256color") to enable support
of 8 bit characters.

10 years agoIssue #19884: readline: Disable the meta modifier key if stdout is not a
Victor Stinner [Thu, 24 Jul 2014 10:23:56 +0000 (12:23 +0200)]
Issue #19884: readline: Disable the meta modifier key if stdout is not a
terminal to not write the ANSI sequence "\033[1034h" into stdout. This sequence
is used on some terminal (ex: TERM=xterm-256color") to enable support of 8 bit
characters.

10 years agotest_gettext: use support.rmtree() instead of shutil.rmtree()
Victor Stinner [Thu, 24 Jul 2014 10:08:20 +0000 (12:08 +0200)]
test_gettext: use support.rmtree() instead of shutil.rmtree()

10 years agotest_gettext: use support.rmtree() instead of shutil.rmtree()
Victor Stinner [Thu, 24 Jul 2014 10:07:45 +0000 (12:07 +0200)]
test_gettext: use support.rmtree() instead of shutil.rmtree()

10 years ago(Merge 3.4) asyncio tests: relax timings for slow buildbots
Victor Stinner [Thu, 24 Jul 2014 10:04:35 +0000 (12:04 +0200)]
(Merge 3.4) asyncio tests: relax timings for slow buildbots

10 years agoasyncio tests: relax timings for slow buildbots
Victor Stinner [Thu, 24 Jul 2014 10:04:22 +0000 (12:04 +0200)]
asyncio tests: relax timings for slow buildbots

10 years ago(Merge 3.4) Issue #20055: Fix BaseEventLoop.stop() docstring, incomplete
Victor Stinner [Thu, 24 Jul 2014 09:34:26 +0000 (11:34 +0200)]
(Merge 3.4) Issue #20055: Fix BaseEventLoop.stop() docstring, incomplete
sentence. Patch written by Saimadhav Heblikar.

10 years agoIssue #20055: Fix BaseEventLoop.stop() docstring, incomplete sentence.
Victor Stinner [Thu, 24 Jul 2014 09:34:11 +0000 (11:34 +0200)]
Issue #20055: Fix BaseEventLoop.stop() docstring, incomplete sentence.
Patch written by Saimadhav Heblikar.

10 years agoMerge 3.4: Fix incomplete sentence in asyncio docs.
Andrew Svetlov [Thu, 24 Jul 2014 08:36:54 +0000 (11:36 +0300)]
Merge 3.4: Fix incomplete sentence in asyncio docs.

10 years agoFix incomplete sentence in asyncio docs.
Andrew Svetlov [Thu, 24 Jul 2014 08:36:33 +0000 (11:36 +0300)]
Fix incomplete sentence in asyncio docs.

10 years agoMerge with 3.4
Terry Jan Reedy [Thu, 24 Jul 2014 06:59:17 +0000 (02:59 -0400)]
Merge with 3.4

10 years agoAsyncio doc fixes: spelling, grammar, duplication.
Terry Jan Reedy [Thu, 24 Jul 2014 06:59:02 +0000 (02:59 -0400)]
Asyncio doc fixes: spelling, grammar, duplication.

10 years agoMerge with 3.4
Terry Jan Reedy [Thu, 24 Jul 2014 06:33:32 +0000 (02:33 -0400)]
Merge with 3.4

10 years agoEnglich grammar nit.
Terry Jan Reedy [Thu, 24 Jul 2014 06:33:14 +0000 (02:33 -0400)]
Englich grammar nit.

10 years agomerge 3.4 (#22036)
Benjamin Peterson [Thu, 24 Jul 2014 04:39:59 +0000 (21:39 -0700)]
merge 3.4 (#22036)

10 years agos/stringobject/bytesobject/ (closes #22036)
Benjamin Peterson [Thu, 24 Jul 2014 04:39:37 +0000 (21:39 -0700)]
s/stringobject/bytesobject/ (closes #22036)

Patch by Martin Matusiak.

10 years agoMerge with 3.4
Terry Jan Reedy [Thu, 24 Jul 2014 00:30:41 +0000 (20:30 -0400)]
Merge with 3.4

10 years agoFix doc example bug reported on python-list by Akshay Verma.
Terry Jan Reedy [Thu, 24 Jul 2014 00:30:29 +0000 (20:30 -0400)]
Fix doc example bug reported on python-list by Akshay Verma.

10 years agoMerge with 3.4
Terry Jan Reedy [Wed, 23 Jul 2014 21:28:15 +0000 (17:28 -0400)]
Merge with 3.4

10 years agoIssue #22053: Cleanup turtledemo start and stop and fix debug shutdown warning.
Terry Jan Reedy [Wed, 23 Jul 2014 21:27:57 +0000 (17:27 -0400)]
Issue #22053: Cleanup turtledemo start and stop and fix debug shutdown warning.

10 years agoIssue #22042: Avoid dangerous C cast in socket.setblocking()
Victor Stinner [Wed, 23 Jul 2014 20:56:55 +0000 (22:56 +0200)]
Issue #22042: Avoid dangerous C cast in socket.setblocking()

Avoid cast from (int*) to (u_long*), even if sizeof(int) == sizeof(u_long).

10 years agoMerge with 3.4
Zachary Ware [Wed, 23 Jul 2014 19:40:27 +0000 (14:40 -0500)]
Merge with 3.4

10 years agoIssue #19493: Fix two uses of ctypes.test.requires (it's not a decorator)
Zachary Ware [Wed, 23 Jul 2014 19:39:50 +0000 (14:39 -0500)]
Issue #19493: Fix two uses of ctypes.test.requires (it's not a decorator)
and skip test_win32.FunctionCallTestCase.test_SEH when Python was compiled
in debug configuration or by a non-MSC compiler.

10 years agoIssue #4350: Removed a number of out-of-dated and non-working for a long time
Serhiy Storchaka [Wed, 23 Jul 2014 19:33:50 +0000 (22:33 +0300)]
Issue #4350: Removed a number of out-of-dated and non-working for a long time
Tkinter methods.

10 years agoNull merge
Serhiy Storchaka [Wed, 23 Jul 2014 19:13:37 +0000 (22:13 +0300)]
Null merge

10 years agoMerge heads
Serhiy Storchaka [Wed, 23 Jul 2014 19:13:09 +0000 (22:13 +0300)]
Merge heads

10 years agoMerge heads
Serhiy Storchaka [Wed, 23 Jul 2014 19:12:53 +0000 (22:12 +0300)]
Merge heads

10 years agoNull merge
Serhiy Storchaka [Wed, 23 Jul 2014 19:09:45 +0000 (22:09 +0300)]
Null merge

10 years agoIssue #6167: Backported tests for Scrollbar.activate() and Scrollbar.set()
Serhiy Storchaka [Wed, 23 Jul 2014 19:08:45 +0000 (22:08 +0300)]
Issue #6167: Backported tests for Scrollbar.activate() and Scrollbar.set()
from 6ae34a948cb4.

10 years agoMerge with 3.4
Terry Jan Reedy [Wed, 23 Jul 2014 19:01:29 +0000 (15:01 -0400)]
Merge with 3.4

10 years agoIssue #21597: Turtledemo text pane can now be widened to view or copy complete
Terry Jan Reedy [Wed, 23 Jul 2014 19:01:12 +0000 (15:01 -0400)]
Issue #21597: Turtledemo text pane can now be widened to view or copy complete
lines or narrowed for small screens.
Issie #19132: Turtledemo buttons no longer disappear when window is shrun.
Patch mostly by Lita Cho (21597) using idea from patch by Jan Kanis (18132).

10 years agoIssue #6167: Scrollbar.activate() now returns the name of active element if
Serhiy Storchaka [Wed, 23 Jul 2014 19:00:44 +0000 (22:00 +0300)]
Issue #6167: Scrollbar.activate() now returns the name of active element if
the argument is not specified.  Scrollbar.set() now always accepts only 2
arguments.  Added tests for Scrollbar.activate() and Scrollbar.set().