xdegaye [Sun, 12 Nov 2017 16:31:07 +0000 (17:31 +0100)]
bpo-28759: Skip some tests on PermissionError raised by Android (GH-4350)
Access to mkfifo(), mknod() and hard link creation is controled
by SELinux on Android.
Also remove test.support.android_not_root.
xdegaye [Sun, 12 Nov 2017 15:50:48 +0000 (16:50 +0100)]
bpo-30696: Fix the REPL looping endlessly when no memory (GH-4160)
xdegaye [Sun, 12 Nov 2017 11:45:59 +0000 (12:45 +0100)]
bpo-28180: Fix the implementation of PEP 538 on Android (GH-4334)
Serhiy Storchaka [Sat, 11 Nov 2017 17:18:28 +0000 (19:18 +0200)]
Move comments in configure.ac to more appropriate place. (#4371)
Edward Betts [Sat, 11 Nov 2017 16:40:26 +0000 (16:40 +0000)]
The termios man page is in section 3 (GH-2450)
Berker Peksag [Sat, 11 Nov 2017 14:55:05 +0000 (17:55 +0300)]
Remove redundant 'exc = True' line (GH-4357)
It can be removed after https://github.com/python/peps/commit/
c28890fb421c906241da6da718f9eacc5a3109ee
Serhiy Storchaka [Sat, 11 Nov 2017 14:41:32 +0000 (16:41 +0200)]
bpo-31572: Get rid of _PyObject_HasAttrId() in the ASDL parser. (#3725)
Silence only expected AttributeError.
Serhiy Storchaka [Sat, 11 Nov 2017 14:19:56 +0000 (16:19 +0200)]
bpo-31572: Get rid of _PyObject_HasAttrId() in dict and OrderedDict. (#3728)
Silence only AttributeError when get "key" and "items" attributes in
the constructor and the update() method of dict and OrderedDict .
Serhiy Storchaka [Sat, 11 Nov 2017 13:51:42 +0000 (15:51 +0200)]
bpo-31572: Silence only AttributeError when get the __copy__ attribute in itertools.tee(). (#3724)
Serhiy Storchaka [Sat, 11 Nov 2017 13:19:47 +0000 (15:19 +0200)]
bpo-31572: Don't silence unexpected errors in the _warnings module. (#3731)
Get rid of _PyObject_HasAttrId() and PyDict_GetItemString().
Silence only expected AttributeError, KeyError and ImportError when
get an attribute, look up in a dict or import a module.
Serhiy Storchaka [Sat, 11 Nov 2017 11:06:26 +0000 (13:06 +0200)]
Add the const qualifier to "char *" variables that refer to literal strings. (#4370)
Pablo Galindo [Fri, 10 Nov 2017 23:05:12 +0000 (23:05 +0000)]
bpo-31824: Document default value of 'errors' parameters (GH-4328)
benfogle [Fri, 10 Nov 2017 21:03:40 +0000 (16:03 -0500)]
bpo-31976: Fix race condition when flushing a file is slow. (#4331)
Yury Selivanov [Fri, 10 Nov 2017 20:34:17 +0000 (15:34 -0500)]
Rewrite asyncio test to be more meaningful (#4363)
Brian Curtin [Fri, 10 Nov 2017 16:38:25 +0000 (11:38 -0500)]
bpo-31985: Deprecate openfp in aifc, sunau, and wave (#4344)
The openfp functions of aifp, sunau, and wave had pointed to the open
function of each module since 1993 as a matter of backwards
compatibility. In the case of aifc.openfp, it was both undocumented
and untested. This change begins the formal deprecation of those
openfp functions, with their removal coming in 3.9.
This additionally adds a TODO in test_pyclbr around using aifc.openfp,
though it shouldn't be changed until removal in 3.9.
Serhiy Storchaka [Fri, 10 Nov 2017 10:09:39 +0000 (12:09 +0200)]
bpo-31999: Fix test_venv in case the zlib module is not available. (#4359)
Serhiy Storchaka [Fri, 10 Nov 2017 10:09:24 +0000 (12:09 +0200)]
bpo-31998: Fix test_zipapp in case the zlib module is not available. (#4358)
Paul Ganssle [Thu, 9 Nov 2017 21:34:29 +0000 (16:34 -0500)]
bpo-31222: Make (datetime|date|time).replace return subclass type in Pure Python (#4176)
Cody Scott [Thu, 9 Nov 2017 18:58:59 +0000 (13:58 -0500)]
Fix phrasing in Doc/whatsnew/3.7.rst (GH-4318)
Ned Batchelder [Thu, 9 Nov 2017 16:55:34 +0000 (11:55 -0500)]
Correct the location of a function mentioned in a comment (GH-4327)
Serhiy Storchaka [Thu, 9 Nov 2017 16:00:38 +0000 (18:00 +0200)]
bpo-31927: Fix bugs in socketmodule.c on NetBSD and other issues. (#4235)
* Fix compilation of the socket module on NetBSD 8.
* Fix the assertion failure or reading arbitrary data when parse
a AF_BLUETOOTH address on NetBSD and DragonFly BSD.
* Fix other potential errors and make the code more reliable.
Antoine Pitrou [Thu, 9 Nov 2017 14:33:43 +0000 (15:33 +0100)]
Silence error output in test_concurrent_futures (bpo-21423) (#4347)
* Silence error output in test_concurrent_futures (bpo-21423)
Victor Stinner [Wed, 8 Nov 2017 22:44:44 +0000 (14:44 -0800)]
Replace KB unit with KiB (#4293)
kB (*kilo* byte) unit means 1000 bytes, whereas KiB ("kibibyte")
means 1024 bytes. KB was misused: replace kB or KB with KiB when
appropriate.
Same change for MB and GB which become MiB and GiB.
Change the output of Tools/iobench/iobench.py.
Round also the size of the documentation from 5.5 MB to 5 MiB.
Berker Peksag [Wed, 8 Nov 2017 21:43:14 +0000 (00:43 +0300)]
bpo-11063: Use more reliable way to check if uuid function exists (GH-4343)
Berker Peksag [Wed, 8 Nov 2017 20:09:16 +0000 (23:09 +0300)]
bpo-11063: Add a configure check for uuid_generate_time_safe (GH-4287)
Tom Floyer [Wed, 8 Nov 2017 17:31:26 +0000 (20:31 +0300)]
bpo-31810: added missing keywords to docs. (#4140)
async and await keywords has been merged into upstream, but they are
all missing in the lexical analysis docs. This change adds them to the
appropriate keywords section in documentation.
Petr Viktorin [Wed, 8 Nov 2017 15:59:20 +0000 (16:59 +0100)]
bpo-29179: Document the Py_UNUSED macro (#4341)
Py_UNUSED has a public name, and is used in the wild outside CPython,
but was not documented. Rectify that.
The macro was added in bpo-19976 and referenced in bpo-26179.
xdegaye [Wed, 8 Nov 2017 15:03:58 +0000 (16:03 +0100)]
bpo-31934: Abort when building out of a not clean source tree (GH-4255)
James [Wed, 8 Nov 2017 14:18:59 +0000 (14:18 +0000)]
bpo-31884 subprocess: add Windows constants for process priority (#4150)
Victor Stinner [Wed, 8 Nov 2017 14:06:24 +0000 (06:06 -0800)]
bpo-31338: C API intro: add missing versionadded (#4339)
Berker Peksag [Wed, 8 Nov 2017 13:36:58 +0000 (16:36 +0300)]
Fix broken link in Doc/whatsnew/3.7.rst (GH-4335)
Petr Viktorin [Wed, 8 Nov 2017 13:11:16 +0000 (14:11 +0100)]
Docs: Mention that Py_UNREACHABLE was added in 3.7 (#4337)
The macro was added for bpo-31338 in commit
b2e5794870eb4728ddfaafc0f79a40299576434f
Sanyam Khurana [Wed, 8 Nov 2017 10:50:56 +0000 (16:20 +0530)]
bpo-21862: Add -m option to cProfile for profiling modules (#4297)
* bpo-21862: Add -m option to cProfile for profiling modules
Simon Willison [Tue, 7 Nov 2017 21:06:55 +0000 (13:06 -0800)]
Remove unused var from CheckCancelOperation test (GH-4317)
It looks like this was copied from one of the previous tests, which did use it.
Justus Schwabedal [Tue, 7 Nov 2017 20:51:43 +0000 (15:51 -0500)]
Added :const:`mmap.ACCESS_DEFAULT` constant. (#4093)
Suren Nihalani [Tue, 7 Nov 2017 17:35:23 +0000 (09:35 -0800)]
bpo-31620: have asyncio/queues not leak memory when you've exceptions during waiting (#3813)
Barry Warsaw [Tue, 7 Nov 2017 17:05:15 +0000 (09:05 -0800)]
Fix a typo (#4323)
Julien Palard [Tue, 7 Nov 2017 16:46:50 +0000 (17:46 +0100)]
bpo-31793: Doc: Specialize smart-quotes for Japanese (GH-4006)
Antoine Pitrou [Tue, 7 Nov 2017 16:23:29 +0000 (17:23 +0100)]
bpo-31970: Reduce performance overhead of asyncio debug mode. (#4314)
* bpo-31970: Reduce performance overhead of asyncio debug mode.
Jelle Zijlstra [Tue, 7 Nov 2017 16:13:02 +0000 (08:13 -0800)]
bpo-31965: fix doc for multiprocessing.connection.Client and Listener (#4304)
* fix doc for multiprocessing.connection.Client
The authenticate argument does not exist on either Client or Listener:
- https://github.com/python/cpython/blob/master/Lib/multiprocessing/connection.py#L483 (master)
- https://github.com/python/cpython/blob/3.6/Lib/multiprocessing/connection.py#L478 (3.6)
- https://github.com/python/cpython/blob/3.5/Lib/multiprocessing/connection.py#L478 (3.5)
- https://github.com/python/cpython/blob/3.4/Lib/multiprocessing/connection.py#L487 (3.4)
The documentation also claimed that these functions will call `current_process().auth_key`, for which I could find no evidence in the code. I rewrote the documentation to reflect the actual behavior.
Also made some small changes to vary sentence structure.
Antoine Pitrou [Tue, 7 Nov 2017 16:03:28 +0000 (17:03 +0100)]
bpo-31960: Fix asyncio.Future documentation for thread (un)safety. (#4319)
Serhiy Storchaka [Tue, 7 Nov 2017 13:08:09 +0000 (15:08 +0200)]
Fix the sizeof test for dicts with shared keys. (#4311)
By accident the size of the empty dict keys object matched the
size of values array.
Berker Peksag [Tue, 7 Nov 2017 12:58:53 +0000 (15:58 +0300)]
bpo-20486: Implement Database.Close() method in msilib (GH-4141)
Serhiy Storchaka [Tue, 7 Nov 2017 10:46:42 +0000 (12:46 +0200)]
bpo-31626: Mark ends of the reallocated block in debug build. (#4210)
Few bytes at the begin and at the end of the reallocated blocks, as well
as the header and the trailer, now are erased before calling realloc()
in debug build. This will help to detect using or double freeing the
reallocated block.
Zackery Spytz [Tue, 7 Nov 2017 10:03:09 +0000 (03:03 -0700)]
Fix a memory leak in _msi.c (#4127)
Antoine Pitrou [Tue, 7 Nov 2017 09:26:32 +0000 (10:26 +0100)]
bpo-31950: Improve event loop policy doc (#4306)
Marat Sharafutdinov [Tue, 7 Nov 2017 09:06:05 +0000 (12:06 +0300)]
Add asyncio.Handle.cancelled() method (#2388)
Serhiy Storchaka [Tue, 7 Nov 2017 06:55:38 +0000 (08:55 +0200)]
bpo-31415: Improve error handling and caching of the importtime option. (#4138)
Mariatta [Tue, 7 Nov 2017 03:31:53 +0000 (19:31 -0800)]
bpo-28791: Update Windows builds to use SQLite 3.21.0. (GH-4246)
Barry Warsaw [Tue, 7 Nov 2017 02:18:40 +0000 (18:18 -0800)]
Fix a minor typo and hyphenate "multi-threading" (#4237)
Anders Lorentsen [Tue, 7 Nov 2017 00:47:43 +0000 (01:47 +0100)]
bpo-31843: sqlite3.connect() now accepts PathLike objects as database name (#4299)
Oren Milman [Tue, 7 Nov 2017 00:09:49 +0000 (02:09 +0200)]
bpo-31764: Prevent a crash in sqlite3.Cursor.close() in case the Cursor object is uninitialized (#3958)
Oren Milman [Tue, 7 Nov 2017 00:01:47 +0000 (02:01 +0200)]
bpo-31770: Prevent a crash and refleaks when calling sqlite3.Cursor.__init__() more than once (#3968)
Nir Soffer [Mon, 6 Nov 2017 21:16:37 +0000 (23:16 +0200)]
bpo-31945: Configurable blocksize in HTTP(S)Connection (#4279)
blocksize was hardcoded to 8192, preventing efficient upload when using
file-like body. Add blocksize argument to __init__, so users can
configure the blocksize to fit their needs.
I tested this uploading data from /dev/zero to a web server dropping the
received data, to test the overhead of the HTTPConnection.send() with a
file-like object.
Here is an example 10g upload with the default buffer size (8192):
$ time ~/src/cpython/release/python upload-httplib.py 10 https://localhost:8000/
Uploaded 10.00g in 17.53 seconds (584.00m/s)
real 0m17.574s
user 0m8.887s
sys 0m5.971s
Same with 512k blocksize:
$ time ~/src/cpython/release/python upload-httplib.py 10 https://localhost:8000/
Uploaded 10.00g in 6.60 seconds (1551.15m/s)
real 0m6.641s
user 0m3.426s
sys 0m2.162s
In real world usage the difference will be smaller, depending on the
local and remote storage and the network.
See https://github.com/nirs/http-bench for more info.
Steve Dower [Mon, 6 Nov 2017 20:52:09 +0000 (12:52 -0800)]
bpo-31957: Fixes version detection. (#4298)
Berker Peksag [Mon, 6 Nov 2017 16:06:05 +0000 (19:06 +0300)]
Remove outdated with_threads checks in configure.ac (GH-4294)
luzpaz [Sun, 5 Nov 2017 13:37:50 +0000 (07:37 -0600)]
Fix miscellaneous typos (#4275)
Serhiy Storchaka [Sun, 5 Nov 2017 09:27:48 +0000 (11:27 +0200)]
bpo-28994: PyErr_NormalizeException() no longer use C stack for recursion. (#2035)
MemoryError raised when normalizing a RecursionError raised during exception normalization now not always causes a fatal error.
Nick Coghlan [Sun, 5 Nov 2017 04:58:45 +0000 (14:58 +1000)]
bpo-22257: Mention startup refactoring in What's New (GH-4286)
While technically a purely internal change, bpo-31845 was
a fairly significant externally visible bug caused by
these changes (environment variable based configuration
was being ignored due to a change in the relative order
of reading the environment and reading command line settings,
and the test suite was only testing the command line options)
Hence this note to essentially say "If you see odd startup
problems in 3.7 that you've never seen in previous releases,
it's probably our fault, so let us know, and we'll fix it".
Steve Dower [Sat, 4 Nov 2017 23:29:03 +0000 (16:29 -0700)]
bpo-31609: Fixes quotes in PCbuild/clean.bat (#4280)
Steve Dower [Sat, 4 Nov 2017 23:28:47 +0000 (16:28 -0700)]
bpo-31944: Fixes build and Modify button (#4278)
davy wybiral [Sat, 4 Nov 2017 23:00:51 +0000 (18:00 -0500)]
bpo-31923: Fix spelling in sqlite3 docs (GH-4227)
Berker Peksag [Sat, 4 Nov 2017 12:17:56 +0000 (15:17 +0300)]
Add version{changed,added} markers and make minor style changes (GH-4273)
Serhiy Storchaka [Sat, 4 Nov 2017 12:16:35 +0000 (14:16 +0200)]
bpo-28564: Use os.scandir() in shutil.rmtree(). (#4085)
This speeds up it to 20-40%.
Phobosmir [Sat, 4 Nov 2017 10:39:45 +0000 (13:39 +0300)]
bpo-31678: Fix typo in PyDateTime_DELTA_GET_MICROSECONDS (GH-3869)
Antoine Pitrou [Sat, 4 Nov 2017 10:05:49 +0000 (11:05 +0100)]
bpo-21423: Add an initializer argument to {Process,Thread}PoolExecutor (#4241)
* bpo-21423: Add an initializer argument to {Process,Thread}PoolExecutor
* Fix docstring
Chillar Anand [Sat, 4 Nov 2017 08:13:16 +0000 (13:43 +0530)]
bpo-18699: Corrected documentation for window.chgat in curses module (#1430)
Serhiy Storchaka [Sat, 4 Nov 2017 07:37:32 +0000 (09:37 +0200)]
bpo-9678: Fix determining the MAC address in the uuid module. (#4264)
* Using ifconfig on NetBSD and OpenBSD.
* Using arp on Linux, FreeBSD, NetBSD and OpenBSD.
Based on patch by Takayuki Shimizukawa.
Barry Warsaw [Fri, 3 Nov 2017 20:45:46 +0000 (13:45 -0700)]
Fix a grammatical problem and reword for clarity. (#4257)
bpo-31936
Jack O'Connor [Fri, 3 Nov 2017 19:02:41 +0000 (15:02 -0400)]
bpo-31933: fix blake2 multi-byte params on big endian platforms (#4250)
All Blake2 params have to be encoded in little-endian byte order. For
the two multi-byte integer params, leaf_length and node_offset, that
means that assigning a native-endian integer to them appears to work on
little-endian platforms, but gives the wrong result on big-endian. The
current libb2 API doesn't make that very clear, and @sneves is working
on new API functions in the GH issue above. In the meantime, we can work
around the problem by explicitly assigning little-endian values to the
parameter block.
See https://github.com/BLAKE2/libb2/issues/12.
Antoine Pitrou [Fri, 3 Nov 2017 18:58:46 +0000 (19:58 +0100)]
bpo-30057: Fix potential missed signal in signal.signal(). (#4258)
Bug report and patch by Jeroen Demeyer.
Serhiy Storchaka [Fri, 3 Nov 2017 18:29:33 +0000 (20:29 +0200)]
bpo-31924: Fix test_curses on NetBSD 8. (#4228)
Antoine Pitrou [Fri, 3 Nov 2017 13:31:38 +0000 (14:31 +0100)]
bpo-31310: multiprocessing's semaphore tracker should be launched again if crashed (#3247)
* bpo-31310: multiprocessing's semaphore tracker should be launched again if crashed
* Avoid mucking with process state in test.
Add a warning if the semaphore process died, as semaphores may then be leaked.
* Add NEWS entry
Antoine Pitrou [Fri, 3 Nov 2017 12:34:22 +0000 (13:34 +0100)]
bpo-31308: If multiprocessing's forkserver dies, launch it again when necessary (#3246)
* bpo-31308: If multiprocessing's forkserver dies, launch it again when necessary.
* Fix test on Windows
* Add NEWS entry
* Adopt a different approach: ignore SIGINT and SIGTERM, as in semaphore tracker.
* Fix comment
* Make sure the test doesn't muck with process state
* Also test previously-started processes
* Update 2017-08-30-17-59-36.bpo-31308.KbexyC.rst
* Avoid masking SIGTERM in forkserver. It's not necessary and causes a race condition in test_many_processes.
Tal Einat [Fri, 3 Nov 2017 09:09:00 +0000 (11:09 +0200)]
bpo-31926: fix missing *_METHODDEF statements by argument clinic (#4230)
When a single .c file contains several functions and/or methods with
the same name, a safety _METHODDEF #define statement is generated
only for one of them.
This fixes the bug by using the full name of the function to avoid
duplicates rather than just the name.
Barry Warsaw [Thu, 2 Nov 2017 23:13:36 +0000 (16:13 -0700)]
bpo-31415: Support PYTHONPROFILEIMPORTTIME envvar equivalent to -X importtime (#4240)
Support PYTHONPROFILEIMPORTTIME envvar equivalent to -X importtime
Anselm Kruis [Thu, 2 Nov 2017 22:54:57 +0000 (23:54 +0100)]
bpo-31890: define METH_STACKLESS (#4159)
Add METH_STACKLESS to prevent future collisions.
Neil Schemenauer [Thu, 2 Nov 2017 17:46:02 +0000 (10:46 -0700)]
bpo-28643: Record profile-opt build progress with stamp files (#4223)
* bpo-28643: Record profile-opt build progress with stamp files
The profile-opt makefile target is expensive to build. Since the
makefile does not contain complete dependency information for this
target, much extra work can get done if the build is interrupted and
re-started. Even running "make" a second time will result in a huge
amount of redundant work.
As a minimal fix (rather than removing recursive "make" and adding a
proper dependency graph), split the profile-opt target into parts:
- ensure tree is clean (profile-clean-stamp)
- build with profile generation enabled (profile-gen-stamp)
- run task to generate profile information (profile-run-stamp)
- build optimized Python using above information (profile-opt)
We use "stamp" files to record completion of the steps. Running
"make clean" will not remove the profile-run-stamp file.
Other minor changes:
- remove the "build_all_use_profile" target. I don't expect callers
of the makefile to use this target so that should be safe.
- remove execution of "profile-removal" at end of "profile-opt". I
don't see any reason to not to keep the profile information, given
the cost to generate it. Removing the "profile-run-stamp" file
will force re-generation of it.
Christian Heimes [Thu, 2 Nov 2017 16:38:11 +0000 (17:38 +0100)]
Update multissltests: 1.0.2m, 1.1.0g (#4232)
Simon Willison [Thu, 2 Nov 2017 14:34:12 +0000 (07:34 -0700)]
Returning 1 from the set_progress_handler handler cancels query (GH-4120)
Victor Stinner [Thu, 2 Nov 2017 14:28:27 +0000 (07:28 -0700)]
bpo-31784: Implement PEP 564: add time.time_ns() (#3989)
Add new time functions:
* time.clock_gettime_ns()
* time.clock_settime_ns()
* time.monotonic_ns()
* time.perf_counter_ns()
* time.process_time_ns()
* time.time_ns()
Add new _PyTime functions:
* _PyTime_FromTimespec()
* _PyTime_FromNanosecondsObject()
* _PyTime_FromTimeval()
Other changes:
* Add also os.times() tests to test_os.
* pytime_fromtimeval() and pytime_fromtimeval() now return
_PyTime_MAX or _PyTime_MIN on overflow, rather than undefined
behaviour
* _PyTime_FromNanoseconds() parameter type changes from long long to
_PyTime_t
Vincent Michel [Thu, 2 Nov 2017 12:47:04 +0000 (13:47 +0100)]
bpo-31307: Make ConfigParser.read() accept bytes objects (GH-3420)
Victor Stinner [Thu, 2 Nov 2017 11:19:19 +0000 (04:19 -0700)]
bpo-31917: Add 3 new clock identifiers (#4207)
Add new clock identfiers:
* time.CLOCK_BOOTTIME
* time.CLOCK_PROF
* time.CLOCK_UPTIME
stratakis [Thu, 2 Nov 2017 10:32:54 +0000 (11:32 +0100)]
bpo-23699: Use a macro to reduce boilerplate code in rich comparison functions (GH-793)
Serhiy Storchaka [Wed, 1 Nov 2017 18:48:49 +0000 (20:48 +0200)]
bpo-27666: Fixed stack corruption in curses.box() and curses.ungetmouse(). (#4220)
Serhiy Storchaka [Wed, 1 Nov 2017 14:03:40 +0000 (16:03 +0200)]
bpo-15037: Add a workaround for getkey() in curses for ncurses 5.7 and earlier. (#3826)
Skip a test for unget_wch()/get_wch() on OpenBSD since they are broken
in ncurses 5.7.
jimmylai [Wed, 1 Nov 2017 13:54:45 +0000 (06:54 -0700)]
[asyncio] bpo-30423: add regression test for orphan future causes "RuntimeError: Event loop stopped before Future completed." (#3295)
* call remove_done_callback in finally section
* [asyncio] bpo-30423 bug: add regression test for orphan future causes "RuntimeError: Event loop stopped before Future completed."
Serhiy Storchaka [Wed, 1 Nov 2017 12:34:20 +0000 (14:34 +0200)]
bpo-31919: Fix building the curses module on OpenIndiana. (#4211)
Lars Viklund [Wed, 1 Nov 2017 12:23:09 +0000 (13:23 +0100)]
Remove nested comments in blake2 (#4173)
Replace occurence of nested comments in blake2 reference implementation
with preprocessor directive for disabling unused code.
`blake2s-load-xop.h` is conditionally pulled in only on chips with XOP
support, among others the AMD Bulldozer. The malformed comments in the
source file breaks the build of `hashlib`'s `_blake2` on GCC 6.3.0.
Official reference code on github uses `#if` so this change should be
uncontroversial.
Masayuki Yamamoto [Wed, 1 Nov 2017 12:05:26 +0000 (21:05 +0900)]
bpo-25720: Fix the method for checking pad state of curses WINDOW (#4164)
Modify the code to use ncurses is_pad() instead of checking WINDOW
_flags field. If your platform does not provide the is_pad(), the
existing way that checks the field will be enabled.
Note: This change does not drop support for platforms where do not
have both WINDOW _flags field and is_pad().
Victor Stinner [Wed, 1 Nov 2017 02:48:14 +0000 (19:48 -0700)]
Fix test_socket.test_create_connection() (#4206)
bpo-31910: test_create_connection() now catchs also EADDRNOTAVAIL to
fix the test on Travis CI.
Victor Stinner [Tue, 31 Oct 2017 19:18:10 +0000 (12:18 -0700)]
bpo-18835: Cleanup pymalloc (#4200)
Cleanup pymalloc:
* Rename _PyObject_Alloc() to pymalloc_alloc()
* Rename _PyObject_FreeImpl() to pymalloc_free()
* Rename _PyObject_Realloc() to pymalloc_realloc()
* pymalloc_alloc() and pymalloc_realloc() don't fallback on the raw
allocator anymore, it now must be done by the caller
* Add "success" and "failed" labels to pymalloc_alloc() and
pymalloc_free()
* pymalloc_alloc() and pymalloc_free() don't update
num_allocated_blocks anymore: it should be done in the caller
* _PyObject_Calloc() is now responsible to fill the memory block
allocated by pymalloc with zeros
* Simplify pymalloc_alloc() prototype
* _PyObject_Realloc() now calls _PyObject_Malloc() rather than
calling directly pymalloc_alloc()
_PyMem_DebugRawAlloc() and _PyMem_DebugRawRealloc():
* document the layout of a memory block
* don't increase the serial number if the allocation failed
* check for integer overflow before computing the total size
* add a 'data' variable to make the code easiler to follow
test_setallocators() of _testcapimodule.c now test also the context.
Victor Stinner [Tue, 31 Oct 2017 16:37:25 +0000 (09:37 -0700)]
bpo-20064: Document PyObject_Malloc() (#4199)
Document the following functions:
* PyObject_Malloc()
* PyObject_Calloc()
* PyObject_Realloc()
* PyObject_Free()
Fix also PyMem_RawFree() documentation.
Serhiy Storchaka [Tue, 31 Oct 2017 16:18:21 +0000 (18:18 +0200)]
bpo-31893: Fix errors in
b9052a0f91d2e83bbc27267247a5920c82b242a3. (#4196)
* Fix a compilation error on FreeBSD.
* Fix the data attribute size on Mac OS X.
Victor Stinner [Tue, 31 Oct 2017 15:41:10 +0000 (08:41 -0700)]
pythoninfo: add Py_DEBUG (#4198)
Victor Stinner [Tue, 31 Oct 2017 15:40:59 +0000 (08:40 -0700)]
test_capi.test__testcapi() becomes more verbose (#4197)
Write the name of each subtest on a new line to help debugging when
a test does crash Python.
Serhiy Storchaka [Tue, 31 Oct 2017 12:05:53 +0000 (14:05 +0200)]
bpo-31897: Convert unexpected errors when read bogus binary plists into InvalidFileException. (#4171)
Serhiy Storchaka [Tue, 31 Oct 2017 12:05:03 +0000 (14:05 +0200)]
bpo-31626: Fixed a bug in debug memory allocator. (#3844)
Removed a code that incorrectly detected in-place resizing in realloc()
and wrote to freed memory.
Serhiy Storchaka [Tue, 31 Oct 2017 11:59:55 +0000 (13:59 +0200)]
bpo-31893: Fixed select.kqueue(). (#4166)
* Fixed the layout of the kqueue_event structure on OpenBSD and NetBSD.
* Fixed the comparison of the kqueue_event objects.
Serhiy Storchaka [Tue, 31 Oct 2017 11:56:44 +0000 (13:56 +0200)]
bpo-31891: Fix building the curses module on NetBSD. (#4165)