6921c13bbb
Minor nit. Make the rotate() success/fail tests consistent.
2015-03-21 02:03:40 -07:00
ac13ad6a32
For safety, wait to decref deleted values until the deque state has been restored.
2015-03-21 01:53:16 -07:00
0f6f94778a
Fix minor formatting nits and remove unnecessary comment.
2015-03-21 01:42:10 -07:00
32ea16577d
Issue 23704: Add index(), copy(), and insert() to deques. Register deques as a MutableSequence.
2015-03-21 01:37:37 -07:00
39dadf7abf
Issue 23705: Improve the performance of __contains__ checks for deques.
2015-03-20 16:38:56 -07:00
17d3a58e39
Issue #22832 : Tweaked parameter names for fcntl module to better match
...
official POSIX documentation. Updated the documenttion for Python 3.
Patch by Alex Shkop.
2015-03-20 20:04:21 +02:00
a453cd8d85
Issue #23715 : signal.sigwaitinfo() and signal.sigtimedwait() are now retried
...
when interrupted by a signal not in the *sigset* parameter, if the signal
handler does not raise an exception. signal.sigtimedwait() recomputes the
timeout with a monotonic clock when it is retried.
Remove test_signal.test_sigwaitinfo_interrupted() because sigwaitinfo() doesn't
raise InterruptedError anymore if it is interrupted by a signal not in its
sigset parameter.
2015-03-20 12:54:28 +01:00
f50e187724
Fix compiler warnings: comparison between signed and unsigned numbers
2015-03-20 11:32:24 +01:00
fbd6f9ed12
Issue #23696 : Chain ZipImportError to the OSError
2015-03-20 10:52:25 +01:00
99953006df
Issue #23709 , #23001 : ossaudiodev now uses Py_ssize_t for sizes instead of int
...
The module is now also "SSIZE_T clean" (for PyArg_Parse...() functions) since
it switched to Py_buffer ("y*" argument format).
2015-03-20 10:37:34 +01:00
716a74ea22
Issue #23001 : Fix typo
2015-03-20 10:24:18 +01:00
8490f5acfe
Issue #23001 : Few functions in modules mmap, ossaudiodev, socket, ssl, and
...
codecs, that accepted only read-only bytes-like object now accept writable
bytes-like object too.
2015-03-20 09:00:36 +02:00
0eac13052c
Issue #23646 : Fix test_threading on Windows
2015-03-20 03:06:12 +01:00
9a8089b32a
Issue #23646 : Enhance precision of time.sleep() and socket timeout when
...
interrupted by a signal
Add a new _PyTime_AddDouble() function and remove _PyTime_ADD_SECONDS() macro.
The _PyTime_ADD_SECONDS only supported an integer number of seconds, the
_PyTime_AddDouble() has subsecond resolution.
2015-03-20 01:42:20 +01:00
4fa99cdb4c
Issue #23709 : The ossaudiodev module now retries read/write when interrupted by
...
a signal (PEP 475).
Use he new _Py_read() and _Py_write() functions.
2015-03-20 00:27:28 +01:00
54799672da
Issue #23708 : select.devpoll now retries its internal write() when interrupted
...
by a signal (EINTR).
Modify devpoll_flush() to use _Py_write() instead of calling directly write().
2015-03-19 23:33:09 +01:00
66aab0c4b5
Issue #23708 : Add _Py_read() and _Py_write() functions to factorize code handle
...
EINTR error and special cases for Windows.
These functions now truncate the length to PY_SSIZE_T_MAX to have a portable
and reliable behaviour. For example, read() result is undefined if counter is
greater than PY_SSIZE_T_MAX on Linux.
2015-03-19 22:53:20 +01:00
79d68f929d
Issue #23646 : If time.sleep() is interrupted by a signal, the sleep is now
...
retried with the recomputed delay, except if the signal handler raises an
exception (PEP 475).
Modify also test_signal to use a monotonic clock instead of the system clock.
2015-03-19 21:54:09 +01:00
009b811d67
Removed unintentional trailing spaces in non-external and non-generated C files.
2015-03-18 21:53:15 +02:00
cb7e5f6f08
Merge 3.4 (tracemalloc typo)
2015-03-18 16:05:32 +01:00
444f124fcf
_tracemalloc.c: Fix typo
2015-03-18 16:05:18 +01:00
e371b3d21a
Fix compiler warning in mmapmodule.c (compare signed/unsigned integers)
2015-03-18 15:04:34 +01:00
5200f55024
Issue #19428 : Handle PyMarshal_Read*() errors in run_pyc_file()
...
Detect also earlier PyMarshal_Read*() errors in zipimport.
2015-03-18 13:56:25 +01:00
e42ccd2bfd
Issue #23694 : Enhance _Py_fopen(), it now raises an exception on error
...
* If fopen() fails, OSError is raised with the original filename object.
* The GIL is now released while calling fopen()
2015-03-18 01:39:23 +01:00
a555cfcb73
Issue #23694 : Enhance _Py_open(), it now raises exceptions
...
* _Py_open() now raises exceptions on error. If open() fails, it raises an
OSError with the filename.
* _Py_open() now releases the GIL while calling open()
* Add _Py_open_noraise() when _Py_open() cannot be used because the GIL is not
held
2015-03-18 00:22:14 +01:00
f024d263b4
Issue #23685 : Fix usage of PyMODINIT_FUNC in _json, _scproxy, nis, pyexpat
...
_codecs_cn, _codecs_hk, _codecs_iso2022, _codecs_jp, _codecs_kr and _codecs_tw
modules.
pyexpat.c doesn't need to redeclare PyMODINIT_FUNC, it's already declared in
Include/pyport.h.
2015-03-17 17:48:27 +01:00
0c2fd89777
Revert changeset d927047b1d8eb87738676980a24930d053ba2150
...
Sorry, it was a mistake, the patch is still under review: issue #23646 .
2015-03-17 10:49:17 +01:00
a60c2fe480
Issue #23641 : Cleaned out legacy dunder names from tests and docs.
...
Fixed 2 to 3 porting bug in pynche.ColorDB.
2015-03-12 21:56:08 +02:00
945c82eea3
test
2015-03-12 16:19:01 +01:00
c2ccce791c
Issue #23641 : Cleaned out legacy dunder names from tests and docs.
...
Fixed 2 to 3 porting bug in pynche.ColorDB.
Added few tests for __truediv__, __floordiv__ and __matmul__.
2015-03-12 22:01:30 +02:00
95bb714ff7
Issue #23566 : enable(), register(), dump_traceback() and dump_traceback_later()
...
functions of faulthandler now accept file descriptors. Patch by Wei Wu.
2015-03-12 15:32:03 +01:00
8acde7dcce
Issue #23524 : Change back to using Windows errors for _Py_fstat instead of the errno shim.
2015-03-07 18:14:07 -08:00
35a97c0bed
Issue #22524 : Fix os.scandir() for platforms which don't have a d_type field in
...
the dirent structure (ex: OpenIndiana).
2015-03-08 02:59:09 +01:00
6036e4431d
Issue #22524 : New os.scandir() function, part of the PEP 471: "os.scandir()
...
function -- a better and faster directory iterator". Patch written by Ben
Hoyt.
2015-03-08 01:58:04 +01:00
4a7cc88472
Issue #23571 : PyObject_Call(), PyCFunction_Call() and call_function() now
...
raise a SystemError if a function returns a result and raises an exception.
The SystemError is chained to the previous exception.
Refactor also PyObject_Call() and PyCFunction_Call() to make them more readable.
Remove some checks which became useless (duplicate checks).
Change reviewed by Serhiy Storchaka.
2015-03-06 23:35:27 +01:00
d81431f587
Issue #23524 : Replace _PyVerify_fd function with calling _set_thread_local_invalid_parameter_handler on every thread.
2015-03-06 14:47:02 -08:00
e37a1946c7
Fix regression introduced by changeset 7c6e3358221a that caused compile
...
errors of _testcapimodule.c with older versions of gcc.
2015-03-05 15:47:10 -08:00
caa01f861c
Fixed GCC version testing.
2015-03-05 23:42:24 +02:00
a4c727131e
Fix "GCC diagnostic" in socketmodule.c
...
Fix regression of changeset 7c6e3358221a on GCC < 4.4. The _socket module
cannot be compiled on "x86 FreeBSD 7.2 3.x" buildbot anymore.
2015-03-05 14:04:03 +01:00
de8eca4638
merge 3.4
2015-03-04 22:50:25 -05:00
990fcaac3c
expose X509_V_FLAG_TRUSTED_FIRST
2015-03-04 22:49:41 -05:00
b64ae7bf2d
merge 3.4 ( #23476 )
2015-03-04 22:11:48 -05:00
fdb1971587
enable X509_V_FLAG_TRUSTED_FIRST when possible ( closes #23476 )
2015-03-04 22:11:12 -05:00
0bfd0a4048
Issue #23576 : Avoid stalling in SSL reads when EOF has been reached in the SSL layer but the underlying connection hasn't been closed.
2015-03-04 20:54:57 +01:00
f7f3b0a14a
Issue #23576 : Avoid stalling in SSL reads when EOF has been reached in the SSL layer but the underlying connection hasn't been closed.
2015-03-04 20:51:55 +01:00
9672da7bb4
Issue #23285 : Fix handling of EINTR in fileio.c
...
Fix handling of EINTR: don't return None if PyErr_CheckSignals() raised an
exception.
Initialize also the length outside the loop to only initialize it once.
2015-03-04 18:40:10 +01:00
87e691240b
Minor neatening-up. Make assignments in same order a struct fields. Line-up comments.
2015-03-02 23:32:02 -08:00
f9d9c79aa8
Switch the state variable to unsigned for defined wrap-around behavior.
2015-03-02 22:47:46 -08:00
30c9074b96
Minor beautification. Move struct definitions to the top. Fix-up a comment.
2015-03-02 22:31:35 -08:00
f30f5b9ba6
Minor code beautification. Replace macro with in-lineable functions.
2015-03-02 22:23:37 -08:00
3c186ba441
Beautify and better document the use of the size_t cast for bounds checking.
2015-03-02 21:45:02 -08:00
83d8fc265a
merge 3.4 ( #23367 )
2015-03-02 11:18:56 -05:00
5061e67f0f
merge 3.3 ( #23367 )
2015-03-02 11:18:40 -05:00
b779bfba45
fix possible overflow bugs in unicodedata ( closes #23367 )
2015-03-02 11:17:05 -05:00
3e96f324dc
Issue #23451 : Update pyconfig.h for Windows to require Vista headers and remove unnecessary version checks.
2015-03-02 08:01:10 -08:00
e49af34151
Issue #7830 : Flatten nested functools.partial.
2015-03-01 15:08:17 -05:00
e2e178e081
Closes issue #22791 : Improved datetime from timestamp methods documentation.
...
Original patch by Akira Li.
2015-03-01 14:52:07 -05:00
7f9ea7543e
Issue #23553 : Use an unsigned cast to tighten-up the bounds checking logic.
2015-03-01 00:38:00 -08:00
90295b450c
Merge heads
2015-03-01 10:03:46 +02:00
c20830804d
Need a (size_t) cast instead of (unsigned) to be big enough for a Py_ssize_t.
2015-02-28 23:29:16 -08:00
50adb9fc32
Issue #20204 : Added the __module__ attribute to _tkinter classes.
2015-03-01 09:07:10 +02:00
efdc16f978
Issue #20204 : Added the __module__ attribute to _tkinter classes.
2015-03-01 09:06:29 +02:00
a473b9da15
Use unsigned division and modulo for item assignment as well.
2015-02-28 17:49:47 -08:00
63d1ff2a0b
Convert one more division to unsigned arithmetic to speed-up deque_item().
2015-02-28 07:41:30 -08:00
7e8c7956a7
Line missed in last checkin
2015-02-27 16:59:29 -08:00
da2850f932
Since the index is always non-negative, use faster unsigned division and modulo.
2015-02-27 12:42:54 -08:00
daf57f25e5
Bump the blocksize up from 62 to 64 to speed up the modulo calculation.
...
Remove the old comment suggesting that it was desireable to have
blocksize+2 as a multiple of the cache line length. That would
have made sense only if the block structure start point was always
aligned to a cache line boundary. However, the memory allocations
are 16 byte aligned, so we don't really have control over whether
the struct spills across cache line boundaries.
2015-02-26 23:21:29 -08:00
b48af340b9
Silenced minor GCC warnings.
2015-02-26 15:27:57 +02:00
e71258a0e6
Issue #15955 : Add an option to limit the output size in bz2.decompress().
...
Patch by Nikolaus Rath.
2015-02-26 13:08:07 +01:00
f63dab5a84
Back-out wcstok deprecation suppression and updates calls to use wcstok_s.
2015-02-25 20:48:01 -08:00
06a13f81ea
Issue #23152 : Move declarations back to posixmodule.c.
...
Declarations of Windows-specific auxilary functions need Windows types
from windows.h. Instead of including windows.h in Python.h and making
it available to all Windows users, it is simpler and safer just move
declarations to the single file that needs them.
2015-02-22 21:34:54 +02:00
12ebbc74f1
Issue #23152 : Move declaration into a header and exclude from stable API.
2015-02-22 19:39:36 +02:00
bf1f376b3e
Issue #23152 : Renames time_t_to_FILE_TIME to _Py_time_t_to_FILE_TIME, removes unused struct win32_stat and return value
2015-02-21 15:26:02 -08:00
a2af1a5a5a
Issue #23152 : Renames attribute_data_to_stat to _Py_attribute_data_to_stat
2015-02-21 10:04:10 -08:00
f2f373f593
Issue #23152 : Implement _Py_fstat() to support files larger than 2 GB on Windows.
...
fstat() may fail with EOVERFLOW on files larger than 2 GB because the file size type is an signed 32-bit integer.
2015-02-21 08:44:05 -08:00
df80706f14
Issue #23215 : Multibyte codecs with custom error handlers that ignores errors
...
consumed too much memory and raised SystemError or MemoryError.
Original patch by Aleksi Torhamo.
2015-02-21 01:21:08 +02:00
a1543cdcd6
Issue #23215 : Multibyte codecs with custom error handlers that ignores errors
...
consumed too much memory and raised SystemError or MemoryError.
Original patch by Aleksi Torhamo.
2015-02-21 01:19:58 +02:00
254dd59068
Issue #5700 : io.FileIO() called flush() after closing the file.
...
flush() was not called in close() if closefd=False.
2015-02-21 00:35:53 +02:00
a3712a9a6c
Issue #5700 : io.FileIO() called flush() after closing the file.
...
flush() was not called in close() if closefd=False.
2015-02-21 00:35:09 +02:00
79d8f3f123
Regenerated Argument Clinic checksums.
2015-02-20 12:46:11 +02:00
7a4f43a8dd
Improve struct cache locality by bring commonly accessed fields close together.
2015-02-20 00:50:04 -08:00
483405bcca
Issue #22883 : Got rid of outdated references to PyInt and PyString in comments.
2015-02-17 10:14:30 +02:00
3584056ca5
Shoould be Py_MIN, not Py_MAX.
2015-02-16 20:54:03 +02:00
26861b0b29
Issue #23450 : Fixed possible integer overflows.
2015-02-16 20:52:17 +02:00
4d0d982985
Issue #23446 : Use PyMem_New instead of PyMem_Malloc to avoid possible integer
...
overflows. Added few missed PyErr_NoMemory().
2015-02-16 13:33:32 +02:00
1a1ff29659
Issue #23446 : Use PyMem_New instead of PyMem_Malloc to avoid possible integer
...
overflows. Added few missed PyErr_NoMemory().
2015-02-16 13:28:22 +02:00
53fa8b2a4b
Fixed few compiler warnings.
2015-02-16 09:40:12 +02:00
c86ca26d32
Issue #23096 : Pickle representation of floats with protocol 0 now is the same
...
for both Python and C implementations.
2015-02-15 14:18:32 +02:00
9ab1cc4fe4
Issue #13637 : Improve exception message of a2b_* functions.
...
Patch by Vajrasky Kok.
2015-02-15 00:31:26 +02:00
3cd30c2cee
Issue #13637 : Improve exception message of a2b_* functions.
...
Patch by Vajrasky Kok.
2015-02-15 00:31:00 +02:00
cf40a9e640
Issue #23450 : Fix signal.set_wakeup_fd() on Windows
...
Detect integer overflow on the file descriptor of the socket on 64-bit Python.
2015-02-12 16:34:54 +01:00
385efb4d99
Merge 3.4 (faulthandler)
2015-02-11 14:23:46 +01:00
7a5567a92c
Issue #23433 : Fix faulthandler._stack_overflow()
...
Fix undefined behaviour: don't compare pointers. Use Py_uintptr_t type instead
of void*. It fixes test_faulthandler on Fedora 22 which now uses GCC 5.
2015-02-11 14:23:35 +01:00
eb6b554fbc
Update copyright.
2015-02-10 22:37:22 -06:00
5ef01e9b93
merge 3.4 ( #23361 )
2015-02-09 21:00:00 -05:00
22ef9f722e
merge 3.3 ( #23361 )
2015-02-09 20:58:52 -05:00
8ce6806498
add overflow checking ( closes #23361 )
2015-02-09 20:58:12 -05:00
6e6c59b508
Issue #23285 : PEP 475 -- Retry system calls failing with EINTR.
2015-02-07 13:27:50 +00:00
6cfb61f163
Issue #23392 : Added tests for marshal C API that works with FILE*.
2015-02-06 09:00:44 +02:00
b51813403f
Issue #23392 : Added tests for marshal C API that works with FILE*.
2015-02-06 08:58:56 +02:00
a7559c04e0
Issue #14203 : Temporary fix for the compile failure on Windows.
2015-02-03 22:27:21 +01:00
650c1e818d
Issue #14203 : Remove obsolete support for view==NULL in bytesiobuf_getbuffer()
...
and array_buffer_getbuf().
2015-02-03 21:43:23 +01:00
38c30e6c8e
Issue #15381 : Fixed a bug in BytesIO.write().
...
It was expected that string_size == PyBytes_GET_SIZE(buf) if the buffer is
shared, but truncate() and __setstate__() can set string_size without
unsharing the buffer.
2015-02-03 18:51:58 +02:00
5178d91be0
Issue #14203 : Remove obsolete support for view==NULL in PyBuffer_FillInfo()
...
and bytearray_getbuffer(). Both functions now raise BufferError in that
case.
2015-02-03 16:57:21 +01:00
b9765eec5c
Issue #15381 : Try to fix refcount bug. Empty and 1-byte buffers are always shared.
2015-02-03 14:57:49 +02:00
87d0b45485
Issue #15381 : Optimized io.BytesIO to make less allocations and copyings.
2015-02-03 11:30:10 +02:00
83e802796c
Issue #22818 : Splitting on a pattern that could match an empty string now
...
raises a warning. Patterns that can only match empty strings are now
rejected.
2015-02-03 11:04:19 +02:00
32ca3dcb97
Issue #23099 : Closing io.BytesIO with exported buffer is rejected now to
...
prevent corrupting exported buffer.
2015-02-03 09:30:51 +02:00
c057c3859c
Issue #23099 : Closing io.BytesIO with exported buffer is rejected now to
...
prevent corrupting exported buffer.
2015-02-03 02:00:18 +02:00
7a27c97216
Issues #23363 , #23364 , #23365 , #23366 : Fixed itertools overflow tests.
...
Used PyMem_New to check overflow.
2015-02-03 01:50:31 +02:00
b5e8e57555
Issues #23363 , #23364 , #23365 , #23366 : Fixed itertools overflow tests.
...
Used PyMem_New to check overflow.
2015-02-03 01:35:10 +02:00
dee948b359
Issues #23363 , #23364 , #23365 , #23366 : Fixed itertools overflow tests.
...
Used PyMem_New to check overflow.
2015-02-03 01:34:09 +02:00
3dd3e26680
Issue #22896 : Avoid to use PyObject_AsCharBuffer(), PyObject_AsReadBuffer()
...
and PyObject_AsWriteBuffer().
2015-02-03 01:25:42 +02:00
4fdb68491e
Issue #22896 : Avoid to use PyObject_AsCharBuffer(), PyObject_AsReadBuffer()
...
and PyObject_AsWriteBuffer().
2015-02-03 01:21:08 +02:00
0e259f18f7
Optimization guides suggest copying memory in an ascending direction when possible.
2015-02-01 22:53:41 -08:00
5d0bb852a2
merge 3.4 ( #23364 , #23363 )
2015-02-01 21:36:01 -05:00
c468b537cd
merge 3.3 ( #23364 , #23363 )
2015-02-01 21:35:34 -05:00
0eaabf1c05
check for overflows in permutations() and product() ( closes #23363 , closes #23364 )
2015-02-01 21:34:07 -05:00
38d9772622
merge 3.4 ( #23365 )
2015-02-01 21:11:54 -05:00
f635dc32b1
merge 3.3 ( #23365 )
2015-02-01 21:11:39 -05:00
6f082297b2
check for overflow in combinations_with_replacement ( closes #23365 )
2015-02-01 21:10:47 -05:00
4db56d5002
merge 3.4 ( #23366 )
2015-02-01 21:01:43 -05:00
819c4e9bc4
merge 3.3 ( #23366 )
2015-02-01 21:00:15 -05:00
4b40eeb339
detect overflow in combinations ( closes #23366 )
2015-02-01 20:59:00 -05:00
71cf91cc09
merge 3.4 ( #23369 )
2015-02-01 18:00:19 -05:00
3675cd9db1
merge 3.3 ( #23369 )
2015-02-01 17:59:49 -05:00
e3bfe19358
fix possible overflow in encode_basestring_ascii ( closes #23369 )
2015-02-01 17:53:53 -05:00
2e6658b4a7
Merge from 3.4.
2015-02-01 19:19:49 +01:00
7213fcc27e
Issue #23370 : Fix off-by-one error for non-contiguous buffers.
2015-02-01 16:19:23 +01:00
363af44a4a
Issue #22445 : PyBuffer_IsContiguous() now implements precise contiguity
...
tests, compatible with NumPy's NPY_RELAXED_STRIDES_CHECKING compilation
flag. Previously the function reported false negatives for corner cases.
2015-02-01 14:53:54 +01:00
bd4976bbd3
Always #define _PyLong_FromDev as we always need it to compile rather than
...
only defining it when HAVE_MKNOD && HAVE_MAKEDEV are true.
This "oops" issue reported by John E. Malmberg on core-mentorship.
(what kinds of systems don't HAVE_MKNOD && HAVE_MAKEDEV?)
2015-01-28 16:08:07 -08:00
702dada2c5
Always #define _PyLong_FromDev as we always need it to compile rather than
...
only defining it when HAVE_MKNOD && HAVE_MAKEDEV are true.
This "oops" issue reported by John E. Malmberg on core-mentorship.
(what kinds of systems don't HAVE_MKNOD && HAVE_MAKEDEV?)
2015-01-28 16:07:52 -08:00
c54de47759
ifdef our way to compatibility with old openssl ( closes #23335 )
2015-01-28 12:06:39 -05:00
07f0515667
disable ALPN on LibreSSL, which has a large version number, but not ALPN support ( closes #23329 )
2015-01-27 11:10:18 -05:00
3bf8684c5e
Merge 3.4 (asyncio)
2015-01-26 22:43:52 +01:00
498b1f62db
asyncio, _overlapped.ConnectPipe(): release the GIL
2015-01-26 22:43:39 +01:00
c27a8afcb0
Merge 3.4 (asyncio)
2015-01-26 22:31:03 +01:00
41063d2a59
asyncio, Tulip issue 204: Fix IocpProactor.recv()
...
If ReadFile() fails with ERROR_BROKEN_PIPE, the operation is not pending: don't
register the overlapped.
I don't know if WSARecv() can fail with ERROR_BROKEN_PIPE. Since
Overlapped.WSARecv() already handled ERROR_BROKEN_PIPE, let me guess that it
has the same behaviour than ReadFile().
2015-01-26 22:30:49 +01:00
47efb4a5dc
Issue #19361 : JSON decoder now raises JSONDecodeError instead of ValueError.
2015-01-26 13:16:30 +02:00
7d0e0c9f7b
Closes #23253 : Delay-load ShellExecute
2015-01-24 08:18:24 -08:00
8861502e07
prefer server alpn ordering over the client's
2015-01-23 17:30:26 -05:00
cca2732a82
add support for ALPN ( closes #20188 )
2015-01-23 16:35:37 -05:00
5da0549154
Merge 3.4 (asyncio)
2015-01-22 23:50:34 +01:00
2b77c5467f
asyncio, Tulip issue 204: Fix IocpProactor.accept_pipe()
...
Overlapped.ConnectNamedPipe() now returns a boolean: True if the pipe is
connected (if ConnectNamedPipe() failed with ERROR_PIPE_CONNECTED), False if
the connection is in progress.
This change removes multiple hacks in IocpProactor.
2015-01-22 23:50:03 +01:00
58c85144db
Merge 3.4 (asyncio)
2015-01-22 22:55:31 +01:00
7ffa2c5fdd
Issue #23293 , asyncio: Rewrite IocpProactor.connect_pipe()
...
Add _overlapped.ConnectPipe() which tries to connect to the pipe for
asynchronous I/O (overlapped): call CreateFile() in a loop until it doesn't
fail with ERROR_PIPE_BUSY. Use an increasing delay between 1 ms and 100 ms.
Remove Overlapped.WaitNamedPipeAndConnect() which is no more used.
2015-01-22 22:55:08 +01:00
91d11bbb2d
Merge 3.4 (asyncio)
2015-01-21 23:40:19 +01:00
d0a28dee78
Issue #23095 , asyncio: Rewrite _WaitHandleFuture.cancel()
...
This change fixes a race conditon related to _WaitHandleFuture.cancel() leading
to Python crash or "GetQueuedCompletionStatus() returned an unexpected event"
logs. Before, the overlapped object was destroyed too early, it was possible
that the wait completed whereas the overlapped object was already destroyed.
Sometimes, a different overlapped was allocated at the same address, leading to
unexpected completition.
_WaitHandleFuture.cancel() now waits until the wait is cancelled to clear its
reference to the overlapped object. To wait until the cancellation is done,
UnregisterWaitEx() is used with an event instead of UnregisterWait().
To wait for this event, a new _WaitCancelFuture class was added. It's a
simplified version of _WaitCancelFuture. For example, its cancel() method calls
UnregisterWait(), not UnregisterWaitEx(). _WaitCancelFuture should not be
cancelled.
The overlapped object is kept alive in _WaitHandleFuture until the wait is
unregistered.
Other changes:
* Add _overlapped.UnregisterWaitEx()
* Remove fast-path in IocpProactor.wait_for_handle() to immediatly set the
result if the wait already completed. I'm not sure that it's safe to
call immediatly UnregisterWaitEx() before the completion was signaled.
* Add IocpProactor._unregistered() to forget an overlapped which may never be
signaled, but may be signaled for the next loop iteration. It avoids to
block forever IocpProactor.close() if a wait was cancelled, and it may also
avoid some "... unexpected event ..." warnings.
2015-01-21 23:39:51 +01:00
b4826c3fd1
Merge with 3.4 ( closes #23280 )
2015-01-20 14:11:38 -06:00
b176d40398
Issue #23280 : Fix docstrings for binascii.(un)hexlify
2015-01-20 13:59:46 -06:00
Raymond Hettinger
Serhiy Storchaka
Victor Stinner
Steve Dower
Ned Deily
Benjamin Peterson
Antoine Pitrou
Alexander Belopolsky
Berker Peksag
Charles-François Natali
Stefan Krah
Gregory P. Smith
Zachary Ware