]> granicus.if.org Git - python/log
python
9 years agoIssue #22117: Use the _PyTime_t API in _datetime.datetime() constructor
Victor Stinner [Sun, 29 Mar 2015 22:09:18 +0000 (00:09 +0200)]
Issue #22117: Use the _PyTime_t API in _datetime.datetime() constructor

* Remove _PyTime_gettimeofday()
* Add _PyTime_GetSystemClock()

9 years agoIssue #22390: test.regrtest now emits a warning if temporary files or
Serhiy Storchaka [Sun, 29 Mar 2015 22:30:22 +0000 (01:30 +0300)]
Issue #22390: test.regrtest now emits a warning if temporary files or
directories are left after running a test.

9 years agoIssue #22390: test.regrtest now emits a warning if temporary files or
Serhiy Storchaka [Sun, 29 Mar 2015 22:28:02 +0000 (01:28 +0300)]
Issue #22390: test.regrtest now emits a warning if temporary files or
directories are left after running a test.

9 years agoIssue #14260: The groupindex attribute of regular expression pattern object
Serhiy Storchaka [Sun, 29 Mar 2015 22:01:48 +0000 (01:01 +0300)]
Issue #14260: The groupindex attribute of regular expression pattern object
now is non-modifiable mapping.

9 years ago#2211: properly document the Morsel behavior changes.
R David Murray [Sun, 29 Mar 2015 21:09:21 +0000 (17:09 -0400)]
#2211: properly document the Morsel behavior changes.

Also deprecate the undocumented set argument instead of removing
it already in 3.5.

Initial patch by Demian Brecht.

9 years agomerge 3.4 (#23801)
Benjamin Peterson [Sun, 29 Mar 2015 20:45:19 +0000 (16:45 -0400)]
merge 3.4 (#23801)

9 years agoCloses #23801 - Ignore entire preamble to multipart in cgi.FieldStorage
Donald Stufft [Sun, 29 Mar 2015 20:43:23 +0000 (16:43 -0400)]
Closes #23801 - Ignore entire preamble to multipart in cgi.FieldStorage

9 years agoMerge: #23792: Ignore KeyboardInterrupt when the pydoc pager is active.
R David Murray [Sun, 29 Mar 2015 19:19:13 +0000 (15:19 -0400)]
Merge: #23792: Ignore KeyboardInterrupt when the pydoc pager is active.

9 years ago#23792: Ignore KeyboardInterrupt when the pydoc pager is active.
R David Murray [Sun, 29 Mar 2015 19:15:40 +0000 (15:15 -0400)]
#23792: Ignore KeyboardInterrupt when the pydoc pager is active.

Previously, if you hit ctl-c while the pager was active, the python that
launched the subprocess for the pager would see the KeyboardInterrupt in the
__exit__ method of the subprocess context manager where it was waiting for the
subprocess to complete, ending the wait.  This would leave the pager running,
while the interactive interpreter, after handling the exception by printing
it, would go back to trying to post a prompt...but the pager would generally
have the terminal in raw mode, and in any case would be still trying to read
from stdin.  On some systems, even exiting python at that point would not
restore the terminal mode.  The problem with raw mode could also happen if
ctl-C was hit when pydoc was called from the shell command line and the pager
was active.

Instead, we now wait on the subprocess in a loop, ignoring KeyboardInterrupt
just like the pager does, until the pager actually exits.

(Note: this was a regression relative to python2...in python2 the pager
is called via system, and system does not return until the pager exits.)

9 years agoAdded explicit tests for issue #23803.
Serhiy Storchaka [Sun, 29 Mar 2015 16:23:27 +0000 (19:23 +0300)]
Added explicit tests for issue #23803.

9 years agoIssue #23803: Fixed str.partition() and str.rpartition() when a separator
Serhiy Storchaka [Sun, 29 Mar 2015 16:21:02 +0000 (19:21 +0300)]
Issue #23803: Fixed str.partition() and str.rpartition() when a separator
is wider then partitioned string.

9 years agoRemoved unintentional trailing spaces in text files.
Serhiy Storchaka [Sun, 29 Mar 2015 16:12:58 +0000 (19:12 +0300)]
Removed unintentional trailing spaces in text files.

9 years agoMake some tests more frienly to MemoryError.
Serhiy Storchaka [Sat, 28 Mar 2015 18:39:51 +0000 (20:39 +0200)]
Make some tests more frienly to MemoryError.
Free memory, unlock hanging threads.

9 years agoMake some tests more frienly to MemoryError.
Serhiy Storchaka [Sat, 28 Mar 2015 18:38:37 +0000 (20:38 +0200)]
Make some tests more frienly to MemoryError.
Free memory, unlock hanging threads.

9 years agoIssue #22117: Fix _PyTime_GetMonotonicClock() and
Victor Stinner [Sat, 28 Mar 2015 04:24:19 +0000 (05:24 +0100)]
Issue #22117: Fix _PyTime_GetMonotonicClock() and
_PyTime_GetSystemClockWithInfo() to not raise an exception and return 0 on
error (it should never occur)

9 years agoIssue #22117: Add the new _PyTime_ROUND_FLOOR rounding method for the datetime
Victor Stinner [Sat, 28 Mar 2015 04:02:39 +0000 (05:02 +0100)]
Issue #22117: Add the new _PyTime_ROUND_FLOOR rounding method for the datetime
module. time.clock_settime() now uses this rounding method instead of
_PyTime_ROUND_DOWN to handle correctly dates before 1970.

9 years agoIssue #22117: Use the _PyTime_t API for time.clock_settime()
Victor Stinner [Sat, 28 Mar 2015 03:09:41 +0000 (04:09 +0100)]
Issue #22117: Use the _PyTime_t API for time.clock_settime()

Remove also the now unused _PyTime_AddDouble() function.

9 years agoIssue #22117: Use the new _PyTime_t API in the select module
Victor Stinner [Sat, 28 Mar 2015 04:07:51 +0000 (05:07 +0100)]
Issue #22117: Use the new _PyTime_t API in the select module

9 years agoIssue #22117: The thread module uses the new _PyTime_t timestamp API
Victor Stinner [Sat, 28 Mar 2015 02:52:05 +0000 (03:52 +0100)]
Issue #22117: The thread module uses the new _PyTime_t timestamp API

Add also a new _PyTime_AsMicroseconds() function.

threading.TIMEOUT_MAX is now be smaller: only 292 years instead of 292,271
years on 64-bit system for example. Sorry, your threads will hang a *little
bit* shorter. Call me if you want to ensure that your locks wait longer, I can
share some tricks with you.

9 years agoIssue #22117: Fix ssl to use _PyTime_t API on sock_timeout
Victor Stinner [Sat, 28 Mar 2015 02:00:46 +0000 (03:00 +0100)]
Issue #22117: Fix ssl to use _PyTime_t API on sock_timeout

I didn't notice that the ssl module uses private attributes of socket objects.

9 years agoIssue #22117: remove _PyTime_INTERVAL() macro
Victor Stinner [Sat, 28 Mar 2015 00:32:13 +0000 (01:32 +0100)]
Issue #22117: remove _PyTime_INTERVAL() macro

9 years agoIssue #22117: The socket module uses _PyTime_t timestamp for timeouts
Victor Stinner [Sat, 28 Mar 2015 00:18:54 +0000 (01:18 +0100)]
Issue #22117: The socket module uses _PyTime_t timestamp for timeouts

9 years agoIssue #22117: Write unit tests for _PyTime_AsTimeval()
Victor Stinner [Sat, 28 Mar 2015 00:26:47 +0000 (01:26 +0100)]
Issue #22117: Write unit tests for _PyTime_AsTimeval()

* _PyTime_AsTimeval() now ensures that tv_usec is always positive
* _PyTime_AsTimespec() now ensures that tv_nsec is always positive
* _PyTime_AsTimeval() now returns an integer on overflow instead of raising an
  exception

9 years agoIssue #23618, #22117: refactor socketmodule.c
Victor Stinner [Fri, 27 Mar 2015 21:59:32 +0000 (22:59 +0100)]
Issue #23618, #22117: refactor socketmodule.c

Move Py_BEGIN_ALLOW_THREADS/Py_END_ALLOW_THREADS inside internal_select_ex() to
prepare a switch to the _PyTime_t type and retry syscall on EINTR.

9 years agoIssue #22117: The signal modules uses the new _PyTime_t API
Victor Stinner [Fri, 27 Mar 2015 17:19:03 +0000 (18:19 +0100)]
Issue #22117: The signal modules uses the new _PyTime_t API

* Add _PyTime_AsTimespec()
* Add unit tests for _PyTime_AsTimespec()

9 years agoIssue #22117: The gc module now uses _PyTime_t timestamp
Victor Stinner [Fri, 27 Mar 2015 16:47:53 +0000 (17:47 +0100)]
Issue #22117: The gc module now uses _PyTime_t timestamp

9 years agoIssue #22117: time.time() now uses the new _PyTime_t API
Victor Stinner [Fri, 27 Mar 2015 17:16:17 +0000 (18:16 +0100)]
Issue #22117: time.time() now uses the new _PyTime_t API

* Add _PyTime_GetSystemClockWithInfo()

9 years agoIssue #22117: time.monotonic() now uses the new _PyTime_t API
Victor Stinner [Fri, 27 Mar 2015 21:27:24 +0000 (22:27 +0100)]
Issue #22117: time.monotonic() now uses the new _PyTime_t API

* Add _PyTime_FromNanoseconds()
* Add _PyTime_AsSecondsDouble()
* Add unit tests for _PyTime_AsSecondsDouble()

9 years agoformat .. note properly
Benjamin Peterson [Fri, 27 Mar 2015 20:07:35 +0000 (16:07 -0400)]
format .. note properly

9 years agoRemove a dead test for a never-launched API
Brett Cannon [Fri, 27 Mar 2015 18:21:26 +0000 (14:21 -0400)]
Remove a dead test for a never-launched API

9 years agoMerge
Brett Cannon [Fri, 27 Mar 2015 16:57:16 +0000 (12:57 -0400)]
Merge

9 years agoFix module deprecation warnings to have a useful stacklevel
Brett Cannon [Fri, 27 Mar 2015 16:56:57 +0000 (12:56 -0400)]
Fix module deprecation warnings to have a useful stacklevel

9 years agoIssue #22117: Fix rounding in _PyTime_FromSecondsObject()
Victor Stinner [Fri, 27 Mar 2015 16:12:45 +0000 (17:12 +0100)]
Issue #22117: Fix rounding in _PyTime_FromSecondsObject()

* Rename _PyTime_FromObject() to _PyTime_FromSecondsObject()
* Add _PyTime_AsNanosecondsObject() and _testcapi.pytime_fromsecondsobject()
* Add unit tests

9 years agoIssue #22117: Fix test_gdb for the new time.sleep()
Victor Stinner [Fri, 27 Mar 2015 14:42:37 +0000 (15:42 +0100)]
Issue #22117: Fix test_gdb for the new time.sleep()

Use time.gmtime() instead of time.sleep(), because time.sleep() is no more
declared with METH_VARARGS but with METH_O. time.gmtime() is still declared
with METH_VARARGS and so it is called with PyCFunction_Call() which is the
target of the test_gdb unit test.

9 years agoMerge 3.4 (test.support)
Victor Stinner [Fri, 27 Mar 2015 14:36:15 +0000 (15:36 +0100)]
Merge 3.4 (test.support)

9 years agoIssue #23445: Fix test.support.python_is_optimized() for CFLAGS=-Og
Victor Stinner [Fri, 27 Mar 2015 14:36:01 +0000 (15:36 +0100)]
Issue #23445: Fix test.support.python_is_optimized() for CFLAGS=-Og

-Og does not optimize the C code, it's just "fast debugging".

9 years agoMerge 3.4 (asyncio)
Victor Stinner [Fri, 27 Mar 2015 14:20:37 +0000 (15:20 +0100)]
Merge 3.4 (asyncio)

9 years agoasyncio: Fix _SelectorTransport.__repr__() if the event loop is closed
Victor Stinner [Fri, 27 Mar 2015 14:20:08 +0000 (15:20 +0100)]
asyncio: Fix _SelectorTransport.__repr__() if the event loop is closed

9 years agoIssue #23715: Fix test_sigtimedwait() of test_eintr
Victor Stinner [Fri, 27 Mar 2015 13:32:22 +0000 (14:32 +0100)]
Issue #23715: Fix test_sigtimedwait() of test_eintr

sigtimedwait([], timeout) fails with OSError(EINVAL) on OpenIndiana, wait for a
signal which will never be received instead.

9 years agoIssue #23451, #22117: Python 3.5 now requires Windows Vista or newer, so
Victor Stinner [Fri, 27 Mar 2015 13:12:08 +0000 (14:12 +0100)]
Issue #23451, #22117: Python 3.5 now requires Windows Vista or newer, so
GetTickCount64() is now always available.

9 years agoIssue #22117: Add a new Python timestamp format _PyTime_t to pytime.h
Victor Stinner [Fri, 27 Mar 2015 12:31:18 +0000 (13:31 +0100)]
Issue #22117: Add a new Python timestamp format _PyTime_t to pytime.h

In practice, _PyTime_t is a number of nanoseconds. Its C type is a 64-bit
signed number. It's integer value is in the range [-2^63; 2^63-1]. In seconds,
the range is around [-292 years; +292 years]. In term of Epoch timestamp
(1970-01-01), it can store a date between 1677-09-21 and 2262-04-11.

The API has a resolution of 1 nanosecond and use integer number. With a
resolution on 1 nanosecond, 64-bit IEEE 754 floating point numbers loose
precision after 194 days. It's not the case with this API. The drawback is
overflow for values outside [-2^63; 2^63-1], but these values are unlikely for
most Python modules, except of the datetime module.

New functions:

- _PyTime_GetMonotonicClock()
- _PyTime_FromObject()
- _PyTime_AsMilliseconds()
- _PyTime_AsTimeval()

This change uses these new functions in time.sleep() to avoid rounding issues.

The new API will be extended step by step, and the old API will be removed step
by step. Currently, some code is duplicated just to be able to move
incrementally, instead of pushing a large change at once.

9 years agoIssue #23648: Document the PEP 475 in the "Porting to Python 3.5" section and
Victor Stinner [Thu, 26 Mar 2015 22:50:57 +0000 (23:50 +0100)]
Issue #23648: Document the PEP 475 in the "Porting to Python 3.5" section and
add a version changed note in modified functions.

9 years ago_PyUnicodeWriter_WriteStr() now checks that the input string is consistent
Victor Stinner [Thu, 26 Mar 2015 08:37:23 +0000 (09:37 +0100)]
_PyUnicodeWriter_WriteStr() now checks that the input string is consistent
in debug mode to detect bugs earlier.

_PyUnicodeWriter_Finish() doesn't check if the read only string is consistent,
whereas it does check consistency for strings built by itself.

9 years agoIssue #23775: pprint() of OrderedDict now outputs the same representation
Serhiy Storchaka [Thu, 26 Mar 2015 06:51:33 +0000 (08:51 +0200)]
Issue #23775: pprint() of OrderedDict now outputs the same representation
as repr().

9 years agoIssue #23776: Removed asserts from pprint.PrettyPrinter constructor.
Serhiy Storchaka [Thu, 26 Mar 2015 06:43:21 +0000 (08:43 +0200)]
Issue #23776: Removed asserts from pprint.PrettyPrinter constructor.

9 years agoIssue #23765: Removed IsBadStringPtr calls in ctypes
Steve Dower [Thu, 26 Mar 2015 04:58:36 +0000 (21:58 -0700)]
Issue #23765: Removed IsBadStringPtr calls in ctypes

9 years agoIssue #22364: Improved some re error messages using regex for hints.
Serhiy Storchaka [Wed, 25 Mar 2015 19:03:47 +0000 (21:03 +0200)]
Issue #22364: Improved some re error messages using regex for hints.

9 years agoCheck that failed writerow() doesn't produce change a file.
Serhiy Storchaka [Wed, 25 Mar 2015 17:16:54 +0000 (19:16 +0200)]
Check that failed writerow() doesn't produce change a file.

9 years agoCheck that failed writerow() doesn't produce change a file.
Serhiy Storchaka [Wed, 25 Mar 2015 17:16:15 +0000 (19:16 +0200)]
Check that failed writerow() doesn't produce change a file.

9 years agoIssue #23742: ntpath.expandvars() no longer loses unbalanced single quotes.
Serhiy Storchaka [Wed, 25 Mar 2015 14:41:15 +0000 (16:41 +0200)]
Issue #23742: ntpath.expandvars() no longer loses unbalanced single quotes.

9 years agoIssue #23742: ntpath.expandvars() no longer loses unbalanced single quotes.
Serhiy Storchaka [Wed, 25 Mar 2015 14:40:15 +0000 (16:40 +0200)]
Issue #23742: ntpath.expandvars() no longer loses unbalanced single quotes.

9 years agoIssue #21717: The zipfile.ZipFile.open function now supports 'x' (exclusive
Serhiy Storchaka [Wed, 25 Mar 2015 08:09:41 +0000 (10:09 +0200)]
Issue #21717: The zipfile.ZipFile.open function now supports 'x' (exclusive
creation) mode.

9 years agoMissing brace and minor formatting in Windows installer
Steve Dower [Wed, 25 Mar 2015 06:46:55 +0000 (23:46 -0700)]
Missing brace and minor formatting in Windows installer

9 years agoIssue #23571: If io.TextIOWrapper constructor fails in _Py_DisplaySourceLine(),
Victor Stinner [Wed, 25 Mar 2015 01:25:25 +0000 (02:25 +0100)]
Issue #23571: If io.TextIOWrapper constructor fails in _Py_DisplaySourceLine(),
close the binary file to fix a resource warning.

9 years agoIssue #23571: Fix reentrant call to Py_FatalError()
Victor Stinner [Wed, 25 Mar 2015 00:54:46 +0000 (01:54 +0100)]
Issue #23571: Fix reentrant call to Py_FatalError()

Flushing sys.stdout and sys.stderr in Py_FatalError() can call again
Py_FatalError(). Add a reentrant flag to detect this case and just abort at the
second call.

9 years agoFixed bytes warnings when run tests with -vv.
Serhiy Storchaka [Tue, 24 Mar 2015 23:33:19 +0000 (01:33 +0200)]
Fixed bytes warnings when run tests with -vv.

9 years agoIssue #21802: The reader in BufferedRWPair now is closed even when closing
Serhiy Storchaka [Tue, 24 Mar 2015 21:21:57 +0000 (23:21 +0200)]
Issue #21802: The reader in BufferedRWPair now is closed even when closing
writer failed in BufferedRWPair.close().

9 years agoIssue #23671: string.Template now allows to specify the "self" parameter as
Serhiy Storchaka [Tue, 24 Mar 2015 20:28:43 +0000 (22:28 +0200)]
Issue #23671: string.Template now allows to specify the "self" parameter as
keyword argument.  string.Formatter now allows to specify the "self" and
the "format_string" parameters as keyword arguments.

9 years agoAdded tests for mixed kinds of Unicode strings.
Serhiy Storchaka [Tue, 24 Mar 2015 19:48:30 +0000 (21:48 +0200)]
Added tests for mixed kinds of Unicode strings.

9 years agoIssue #23583: Added tests for standard IO streams in IDLE.
Serhiy Storchaka [Tue, 24 Mar 2015 17:46:54 +0000 (19:46 +0200)]
Issue #23583: Added tests for standard IO streams in IDLE.

9 years agoIssue #23571: Py_FatalError() now tries to flush sys.stdout and sys.stderr
Victor Stinner [Tue, 24 Mar 2015 12:44:35 +0000 (13:44 +0100)]
Issue #23571: Py_FatalError() now tries to flush sys.stdout and sys.stderr

It should help to see exceptions when stderr if buffered: PyErr_Display() calls
sys.stderr.write(), it doesn't write into stderr file descriptor directly.

9 years agoIssue #23571: Enhance Py_FatalError()
Victor Stinner [Tue, 24 Mar 2015 10:24:06 +0000 (11:24 +0100)]
Issue #23571: Enhance Py_FatalError()

* Display the current Python stack if an exception was raised but the exception
  has no traceback
* Disable faulthandler if an exception was raised (before it was only disabled
  if no exception was raised)
* To display the current Python stack, call PyGILState_GetThisThreadState()
  which works even if the GIL was released

9 years ago#11468: improve unittest basic example. Initial patch by Florian Preinstorfer.
Ezio Melotti [Tue, 24 Mar 2015 10:42:41 +0000 (12:42 +0200)]
#11468: improve unittest basic example.  Initial patch by Florian Preinstorfer.

9 years agoIssue #23654: Fix faulthandler._stack_overflow() for the Intel C Compiler (ICC)
Victor Stinner [Mon, 23 Mar 2015 20:20:27 +0000 (21:20 +0100)]
Issue #23654: Fix faulthandler._stack_overflow() for the Intel C Compiler (ICC)

Issue #23654: Turn off ICC's tail call optimization for the stack_overflow
generator. ICC turns the recursive tail call into a loop.

Patch written by Matt Frank.

9 years agoIssue #21560: An attempt to write a data of wrong type no longer cause
Serhiy Storchaka [Mon, 23 Mar 2015 13:25:43 +0000 (15:25 +0200)]
Issue #21560: An attempt to write a data of wrong type no longer cause
GzipFile corruption.  Original patch by Wolfgang Maier.

9 years agoIssue 23729: Document ElementTree namespace handling and fix an omission in the...
Raymond Hettinger [Sun, 22 Mar 2015 22:29:09 +0000 (15:29 -0700)]
Issue 23729:  Document ElementTree namespace handling and fix an omission in the XPATH predicate table.

9 years ago#23647: Increase imaplib's MAXLINE to accommodate modern mailbox sizes.
R David Murray [Sun, 22 Mar 2015 20:17:46 +0000 (16:17 -0400)]
#23647: Increase imaplib's MAXLINE to accommodate modern mailbox sizes.

9 years ago#23539: Set Content-Length to 0 for PUT, POST, and PATCH if body is None.
R David Murray [Sun, 22 Mar 2015 19:18:23 +0000 (15:18 -0400)]
#23539: Set Content-Length to 0 for PUT, POST, and PATCH if body is None.

Some http servers will reject PUT, POST, and PATCH requests if they
do not have a Content-Length header.

Patch by James Rutherford, with additional cleaning up of the
'request' documentation by me.

9 years ago#23700: fix/improve comment
R David Murray [Sun, 22 Mar 2015 16:33:46 +0000 (12:33 -0400)]
#23700: fix/improve comment

9 years agoclarify behavior of shutil.move when destination exists (closes #22933)
Benjamin Peterson [Sun, 22 Mar 2015 14:11:54 +0000 (10:11 -0400)]
clarify behavior of shutil.move when destination exists (closes #22933)

Patch by Mike Short.

9 years agoIssue #22289: Prevent test_urllib2net failures due to ftp connection timeout.
Ned Deily [Sun, 22 Mar 2015 08:14:48 +0000 (01:14 -0700)]
Issue #22289: Prevent test_urllib2net failures due to ftp connection timeout.

9 years agoIssue #22079: Deprecation warning now is issued in PyType_Ready() instead of
Serhiy Storchaka [Sun, 22 Mar 2015 07:46:36 +0000 (09:46 +0200)]
Issue #22079: Deprecation warning now is issued in  PyType_Ready() instead of
raising TypeError when statically allocated type subclasses dynamically
allocated type

9 years agoIssue #22351: The nntplib.NNTP constructor no longer leaves the connection
Serhiy Storchaka [Sat, 21 Mar 2015 07:40:26 +0000 (09:40 +0200)]
Issue #22351: The nntplib.NNTP constructor no longer leaves the connection
and socket open until the garbage collector cleans them up.  Patch by
Martin Panter.

9 years agoCloses #9445: Removes detection of GetFinalPathNameByHandle
Steve Dower [Sat, 21 Mar 2015 04:49:12 +0000 (21:49 -0700)]
Closes #9445: Removes detection of GetFinalPathNameByHandle

9 years agoIssue #23765: Remove IsBadStringPtr calls in ctypes
Steve Dower [Wed, 25 Mar 2015 06:28:54 +0000 (23:28 -0700)]
Issue #23765: Remove IsBadStringPtr calls in ctypes
Also renames a local to avoid warnings about shadowing

9 years agoFixes UI labels and ability to add/remove features.
Steve Dower [Wed, 25 Mar 2015 06:20:57 +0000 (23:20 -0700)]
Fixes UI labels and ability to add/remove features.

9 years agoAdds pydoc shortcut, Edit with IDLE menu, moves redist folder into externals and...
Steve Dower [Wed, 25 Mar 2015 04:25:16 +0000 (21:25 -0700)]
Adds pydoc shortcut, Edit with IDLE menu, moves redist folder into externals and fixes test script.

9 years agoFix typo
Raymond Hettinger [Wed, 25 Mar 2015 04:12:57 +0000 (21:12 -0700)]
Fix typo

9 years agoMinor code and comment cleanups.
Raymond Hettinger [Wed, 25 Mar 2015 04:01:50 +0000 (21:01 -0700)]
Minor code and comment cleanups.

9 years agoIssue #23573: Fix bytes.rfind() and bytearray.rfind() on Windows
Victor Stinner [Wed, 25 Mar 2015 02:16:32 +0000 (03:16 +0100)]
Issue #23573: Fix bytes.rfind() and bytearray.rfind() on Windows

Windows has no memrchr() function.

This change is only a workaround, the optimization must be reenabled on other
platforms.

9 years agoMerge 3.4 (traceback)
Victor Stinner [Wed, 25 Mar 2015 01:30:01 +0000 (02:30 +0100)]
Merge 3.4 (traceback)

9 years ago(Merge 3.4) Issue #23571: Fix reentrant call to Py_FatalError()
Victor Stinner [Wed, 25 Mar 2015 00:55:14 +0000 (01:55 +0100)]
(Merge 3.4) Issue #23571: Fix reentrant call to Py_FatalError()

Flushing sys.stdout and sys.stderr in Py_FatalError() can call again
Py_FatalError(). Add a reentrant flag to detect this case and just abort at the
second call.

9 years agoFixed bytes warnings when run tests with -vv.
Serhiy Storchaka [Tue, 24 Mar 2015 23:34:03 +0000 (01:34 +0200)]
Fixed bytes warnings when run tests with -vv.

9 years agoFixed using deprecated escaping in regular expression in _strptime.py (issue23622).
Serhiy Storchaka [Tue, 24 Mar 2015 23:21:50 +0000 (01:21 +0200)]
Fixed using deprecated escaping in regular expression in _strptime.py (issue23622).

9 years agoIssue #21802: The reader in BufferedRWPair now is closed even when closing
Serhiy Storchaka [Tue, 24 Mar 2015 21:23:28 +0000 (23:23 +0200)]
Issue #21802: The reader in BufferedRWPair now is closed even when closing
writer failed in BufferedRWPair.close().

9 years agoIssue #23622: Unknown escapes in regular expressions that consist of ``'\'``
Serhiy Storchaka [Tue, 24 Mar 2015 20:58:14 +0000 (22:58 +0200)]
Issue #23622: Unknown escapes in regular expressions that consist of ``'\'``
and ASCII letter now raise a deprecation warning and will be forbidden in
Python 3.6.

9 years agoIssue #4727: Fixed issue number in Misc/NEWS.
Serhiy Storchaka [Tue, 24 Mar 2015 20:37:08 +0000 (22:37 +0200)]
Issue #4727: Fixed issue number in Misc/NEWS.

9 years agoIssue #23671: string.Template now allows to specify the "self" parameter as
Serhiy Storchaka [Tue, 24 Mar 2015 20:30:46 +0000 (22:30 +0200)]
Issue #23671: string.Template now allows to specify the "self" parameter as
keyword argument.  string.Formatter now allows to specify the "self" and
the "format_string" parameters as keyword arguments.

9 years agoIssue #23573: Increased performance of string search operations (str.find,
Serhiy Storchaka [Tue, 24 Mar 2015 19:55:47 +0000 (21:55 +0200)]
Issue #23573: Increased performance of string search operations (str.find,
str.index, str.count, the in operator, str.split, str.partition) with
arguments of different kinds (UCS1, UCS2, UCS4).

9 years agoAdded tests for mixed kinds of Unicode strings.
Serhiy Storchaka [Tue, 24 Mar 2015 19:49:53 +0000 (21:49 +0200)]
Added tests for mixed kinds of Unicode strings.

9 years agoIssue #23583: Added tests for standard IO streams in IDLE.
Serhiy Storchaka [Tue, 24 Mar 2015 17:49:23 +0000 (19:49 +0200)]
Issue #23583: Added tests for standard IO streams in IDLE.

9 years agoIssue #23502: The pprint module now supports mapping proxies.
Serhiy Storchaka [Tue, 24 Mar 2015 17:31:50 +0000 (19:31 +0200)]
Issue #23502: The pprint module now supports mapping proxies.
In particular the __dict__ attributes of building types.

9 years agoIssue #17530: pprint now wraps long bytes objects and bytearrays.
Serhiy Storchaka [Tue, 24 Mar 2015 17:22:37 +0000 (19:22 +0200)]
Issue #17530: pprint now wraps long bytes objects and bytearrays.

9 years agoIssue #23741: Slightly refactor the pprint module to make it a little more
Serhiy Storchaka [Tue, 24 Mar 2015 16:45:23 +0000 (18:45 +0200)]
Issue #23741: Slightly refactor the pprint module to make it a little more
extesible.  No public API is added.

9 years agoIssue #22687: Fixed some corner cases in breaking words in tetxtwrap.
Serhiy Storchaka [Tue, 24 Mar 2015 16:32:27 +0000 (18:32 +0200)]
Issue #22687: Fixed some corner cases in breaking words in tetxtwrap.
Got rid of quadratic complexity in breaking long words.

9 years agochange Σ to ν for obscure joke reasons
Benjamin Peterson [Tue, 24 Mar 2015 16:12:44 +0000 (12:12 -0400)]
change Σ to ν for obscure joke reasons

https://twitter.com/ncoghlan_dev/status/579173053793353728

9 years agoIssue #20289: The copy module now uses pickle protocol 4 (PEP 3154) and
Serhiy Storchaka [Tue, 24 Mar 2015 16:06:42 +0000 (18:06 +0200)]
Issue #20289: The copy module now uses pickle protocol 4 (PEP 3154) and
supports copying of instances of classes whose __new__ method takes
keyword-only arguments.

9 years agoIssue #23571: Enhance _Py_CheckFunctionResult()
Victor Stinner [Tue, 24 Mar 2015 15:28:52 +0000 (16:28 +0100)]
Issue #23571: Enhance _Py_CheckFunctionResult()

Too bad, sometimes Py_FatalError() is unable to write the exception into
sys.stderr (on "AMD64 OpenIndiana 3.x" buildbot, the buildbot was probably out
of memory).

Call Py_FatalError() with a different message for the two cases (result+error,
or no result and no error).

9 years agoIssue #23571: Update test_capi
Victor Stinner [Tue, 24 Mar 2015 13:01:32 +0000 (14:01 +0100)]
Issue #23571: Update test_capi

9 years ago(Merge 3.4) Issue #23571: Py_FatalError() now tries to flush sys.stdout and
Victor Stinner [Tue, 24 Mar 2015 12:46:18 +0000 (13:46 +0100)]
(Merge 3.4) Issue #23571: Py_FatalError() now tries to flush sys.stdout and
sys.stderr

It should help to see exceptions when stderr if buffered: PyErr_Display() calls
sys.stderr.write(), it doesn't write into stderr file descriptor directly.