[issue15695] Correct __sizeof__ support for StgDict

Serhiy Storchaka added the comment: Now changes are simpler since _PyDict_SizeOf() already was implemented. -- priority: normal -> low ___ Python tracker

[issue29568] undefined parsing behavior with the old style string formatting

Changes by Serhiy Storchaka : -- pull_requests: +422 ___ Python tracker ___ ___

[issue25455] Some repr implementations don't check for self-referential structures

Changes by Serhiy Storchaka : -- pull_requests: +423 ___ Python tracker ___ ___

[issue29735] Optimize functools.partial() for positional arguments

Changes by STINNER Victor : -- pull_requests: +425 ___ Python tracker ___ ___

[issue15695] Correct __sizeof__ support for StgDict

Changes by Serhiy Storchaka : -- pull_requests: +418 ___ Python tracker ___ ___

[issue29714] can't interpolate byte string with \x00 before replacement identifier

Xiang Zhang added the comment: I think no. String is not affected now and its code uses related macros so I don't think it could suffer possible regression. -- ___ Python tracker

[issue29733] concurrent.futures as_completed raise TimeoutError wrong

New submission from jiangwanwei: when I use as_completed function to wait my futures, if I sleep more than timeout seconds in each iteration , I found that futures has been set result, but raise TimeoutError. as my test example code shows: from concurrent import futures from multiprocessing

[issue29410] Moving to SipHash-1-3

INADA Naoki added the comment: microbench result: https://gist.github.com/methane/33c7b01c45ce23b67246f5ddaff9c9e7 Not significant ~ very little difference for small data. For 4KB: Median +- std dev: [python.default] 3.26 us +- 0.03 us -> [python.siphash13] 2.03 us +- 0.02 us: 1.60x faster

[issue28230] tarfile does not support pathlib

Changes by Serhiy Storchaka : -- pull_requests: +421 ___ Python tracker ___ ___

[issue28230] tarfile does not support pathlib

Serhiy Storchaka added the comment: Updated documentation. Note that path-like objects are supported only for external names. Internal names are not OS paths. -- ___ Python tracker

[issue29700] readline memory corruption when sys.stdin fd >= FD_SETSIZE for select()

Marien Zwart added the comment: On Python 3, crash_readline_fdset.py does not crash for me, because its input() contains a check documented as: /* We should only use (GNU) readline if Python's sys.stdin and sys.stdout are the same as C's stdin and stdout, because we

[issue8256] input() doesn't catch _PyUnicode_AsString() exception; io.StringIO().encoding is None

Changes by Serhiy Storchaka : -- pull_requests: +427 ___ Python tracker ___ ___

[issue25455] Some repr implementations don't check for self-referential structures

Changes by Serhiy Storchaka : -- versions: +Python 3.7 -Python 3.4 ___ Python tracker ___

[issue29735] Optimize functools.partial() for positional arguments

New submission from STINNER Victor: The pull request makes functools.partial() faster for positional arguments. It avoids the creation of a tuple for positional arguments. It allocates a small buffer for up to 5 parameters. But it seems like even if the small buffer is not used, it's still

[issue28243] Performance regression in functools.partial()

STINNER Victor added the comment: > Oh, functools.partial.__call__() doesn't use fastcall yet. This issue reminded me that I didn't finish to optimize partial_call(): see issue #29735 for a minor optimization. -- ___ Python tracker

[issue29655] Certain errors during IMPORT_STAR can leak a reference

Changes by Serhiy Storchaka : -- pull_requests: +419 ___ Python tracker ___ ___

[issue28231] zipfile does not support pathlib

Changes by Serhiy Storchaka : -- pull_requests: +420 ___ Python tracker ___ ___

[issue28243] Performance regression in functools.partial()

STINNER Victor added the comment: Oh wait, there was a major regression in my perf module :-( The --compare-to option was completely broken in the development branch (but it works for timeit --compare-to in the latest release). It's now fixed! So please ignore results of my previous comment.

[issue26121] Use C99 functions in math if available

Serhiy Storchaka added the comment: I don't know how to run custom build on buildbots with new workflow. -- versions: +Python 3.7 -Python 3.6 ___ Python tracker

[issue29700] readline memory corruption when sys.stdin fd >= FD_SETSIZE for select()

Changes by Xiang Zhang : -- nosy: +xiang.zhang ___ Python tracker ___ ___ Python-bugs-list

[issue29565] Still broken ctypes calling convention on MSVC / 64-bit Windows (large structs)

Serhiy Storchaka added the comment: New test causes a compiler warning: /home/serhiy/py/cpython/Modules/_ctypes/_ctypes_test.c: In function ‘_testfunc_large_struct_update_value’: /home/serhiy/py/cpython/Modules/_ctypes/_ctypes_test.c:53:42: warning: parameter ‘in’ set but not used

[issue29734] nt._getfinalpathname handle leak

New submission from Eryk Sun: The implementation of nt._getfinalpathname leaks a File handle if calling GetFinalPathNameByHandle fails. The latter function is practically guaranteed to fail when resolving the path for a non-file-system device. It also fails when VOLUME_NAME_DOS is requested

[issue26121] Use C99 functions in math if available

Changes by Serhiy Storchaka : -- pull_requests: +424 ___ Python tracker ___ ___

[issue9051] Improve pickle format for timezone aware datetime instances

Changes by Serhiy Storchaka : -- status: open -> pending ___ Python tracker ___

[issue8256] input() doesn't catch _PyUnicode_AsString() exception; io.StringIO().encoding is None

Changes by Serhiy Storchaka : -- versions: +Python 3.7 -Python 3.4 ___ Python tracker ___

[issue29735] Optimize functools.partial() for positional arguments

Changes by STINNER Victor : -- nosy: +ncoghlan, rhettinger ___ Python tracker ___

[issue29735] Optimize functools.partial() for positional arguments

STINNER Victor added the comment: functools.partial() is commonly used in the the asyncio module. The asyncio doc suggests to use it, because of deliberate limitations of the asyncio API. -- nosy: +inada.naoki, serhiy.storchaka, yselivanov ___

[issue15343] "pydoc -w " writes out page with empty "Package Contents" section

Wolfgang Maier added the comment: Sorry, for generating noise on this very old issue, but was there a specific reason to duplicate the code of ImpImporter.find_module in changeset 9101eab6178c instead of factoring it out? -- nosy: +wolma ___ Python

[issue29695] Weird keyword parameter names in builtins

Serhiy Storchaka added the comment: If David agrees with this. -- ___ Python tracker ___ ___ Python-bugs-list

[issue27577] Make implementation and doc of tuple and list more compliant

Changes by Xiang Zhang : -- status: open -> closed ___ Python tracker ___ ___

[issue29736] Optimize builtin types constructor

New submission from STINNER Victor: Attached PR replaces PyArg_ParseTupleAndKeywords() with _PyArg_ParseTupleAndKeywordsFast() to optimize the constructor of the builtin types: * bool: bool_new() * bytes: bytes_new() * complex: complex_new() * float: float_new() * int: long_new() * list:

[issue29736] Optimize builtin types constructor

Changes by STINNER Victor : -- pull_requests: +429 ___ Python tracker ___ ___

[issue27577] Make implementation and doc of tuple and list more compliant

Xiang Zhang added the comment: This issue doesn't make any sense once #29695 is applied. So close. -- resolution: -> fixed stage: -> resolved ___ Python tracker

[issue29736] Optimize builtin types constructor

Serhiy Storchaka added the comment: The half of constructors no longer use PyArg_ParseTupleAndKeywords(). I would wait until constructors be converted to Argument Clinic. There are ready patches for some of these builtin types, patches for other are in proggress. -- nosy:

[issue29695] Weird keyword parameter names in builtins

STINNER Victor added the comment: Unhappy buildbot. http://buildbot.python.org/all/builders/AMD64%20Debian%20root%203.x/builds/456/steps/test/logs/stdio == ERROR: test_keyword_arguments

[issue29695] Weird keyword parameter names in builtins

Changes by Serhiy Storchaka : -- pull_requests: +430 ___ Python tracker ___ ___

[issue29695] Weird keyword parameter names in builtins

Serhiy Storchaka added the comment: Thanks all! -- resolution: -> fixed stage: patch review -> resolved status: open -> closed ___ Python tracker ___

[issue29695] Weird keyword parameter names in builtins

STINNER Victor added the comment: > If you think that it would be better to remove without deprecation, the > following patch does this. @Serhiy: Can you please create a PR for it? -- ___ Python tracker

[issue29714] can't interpolate byte string with \x00 before replacement identifier

STINNER Victor added the comment: Xiang Zhang added the comment: > I think no. String is not affected now and its code uses related macros so I > don't think it could suffer possible regression. Regressions are not something "expected", shit happens, all the time. Unexpected corner cases are

[issue29695] Weird keyword parameter names in builtins

Changes by Serhiy Storchaka : -- pull_requests: +428 ___ Python tracker ___ ___

[issue29695] Weird keyword parameter names in builtins

R. David Murray added the comment: If Raymond is on the side of skipping the deprecation than I'm good with it. Like I said, this is a marginal case. -- ___ Python tracker

[issue29735] Optimize functools.partial() for positional arguments

Serhiy Storchaka added the comment: What about C stack consumption? Is not this increase it? Since nested partial()`s are collapsed, you need to interlace them with other wrapper for testing. def decorator(f): def wrapper(*args): return f(*args) return wrapper def f(*args):

[issue29733] concurrent.futures as_completed raise TimeoutError wrong

Josh Rosenberg added the comment: The docs ( https://docs.python.org/3/library/concurrent.futures.html#concurrent.futures.as_completed ) do seem to indicate it shouldn't do so as long as results were available before the timeout expired: "The returned iterator raises a

[issue29571] test_re is failing when local is set for `en_IN`

Nick Coghlan added the comment: Thanks for the explanation - given that, I agree that simply reverting the attempted test-based fix and instead relying on the issue 20087 updates is the way to go. -- ___ Python tracker

[issue29545] Python behavioral difference between Linux and AIX

Changes by Anna Henningsen : -- nosy: +addaleax ___ Python tracker ___ ___

[issue29735] Optimize functools.partial() for positional arguments

Serhiy Storchaka added the comment: If the underlying function doesn't support fast call, and either args or pto->args are empty, partial_call() makes two unneeded copyings. Arguments are copied from a tuple to the raw array and from the array to new tuple. This is what the current code does,

[issue13566] Increase pickle compatibility

Josh Rosenberg added the comment: Right, but Antoine's objection is that suddenly strs pickled in Py3 can end up as strs in Py2, rather than unicode. If the library enforces a Py3-like type separation on Py2 (text arguments are unicode only, binary data is str only), then you have the problem

[issue29676] verbose output of test_cprofile

Changes by INADA Naoki : -- pull_requests: +431 ___ Python tracker ___ ___

[issue29723] 3.6.1rc1 adds the current directory to sys.path when running a subdirectory's __main__.py; previous versions did not

Nick Coghlan added the comment: I think Eryk's diagnosis is correct: this is a straight up bug in the original patch, where it's missing the check to only use the new logic when in isolated mode (as it's compensating for the fact that there is no extra sys.path[0] entry inserted in that

[issue29676] verbose output of test_cprofile

INADA Naoki added the comment: OK, now I found what caused this difference. lsprof expect `func` is PyCFunction object. But it can be PyMethodDescrObject when LOAD_METHOD is used to call C method. In Modules/_lsprof.c (line 459): case PyTrace_C_CALL: if ProfilerObject

[issue29708] support reproducible Python builds

Changes by Brett Cannon : -- nosy: +brett.cannon ___ Python tracker ___ ___

[issue29691] Some tests fail in coverage Travis check

Brett Cannon added the comment: The test_traceback changes should be easy to tweak to use a regex or startwith() check so the tests pass again. For the test_xml_etree failures I'm not sure how best to fix that other than taking it out. Maybe if sys.gettrace() returns something then skip the

[issue15954] No error checking after using of the wcsxfrm()

Changes by Mark Lawrence : -- nosy: -BreamoreBoy ___ Python tracker ___ ___

[issue15954] No error checking after using of the wcsxfrm()

Changes by Serhiy Storchaka : -- assignee: -> serhiy.storchaka resolution: -> fixed stage: patch review -> resolved status: open -> closed versions: +Python 3.7 -Python 3.4, Python 3.5 ___ Python tracker

[issue29716] Python 3 Module doc still sounds like __init__.py is required

Brett Cannon added the comment: I've gone ahead and closed this. Thanks for taking the time to check in on this, James! -- nosy: +brett.cannon resolution: -> not a bug stage: -> resolved status: open -> closed ___ Python tracker

[issue29691] Some tests fail in coverage Travis check

Changes by Brett Cannon : -- assignee: -> brett.cannon ___ Python tracker ___ ___

[issue29691] Some tests fail in coverage Travis check

Serhiy Storchaka added the comment: Is this a duplicate of issue29048? -- nosy: +serhiy.storchaka ___ Python tracker ___

[issue29691] Some tests fail in coverage Travis check

Brett Cannon added the comment: BTW, just because I assigned this to me doesn't mean others can't write a PR to solve this. :) I will review any PR that fixes this issue, I'm just going to make sure it gets solved no matter what if no one else creates a PR. --

[issue29737] Optimize concatenating empty tuples

Changes by Serhiy Storchaka : -- pull_requests: +432 ___ Python tracker ___ ___

[issue29738] Fix memory leak in SSLSocket.getpeercert()

New submission from Olivier Vielpeau: The code snippet in #25569 reproduces the memory leak with Python 3.6.0 and 2.7.13. The current memory leak is a regression that was introduced in #26470. Going to attach a PR on github that fixes the issue shortly. -- assignee: christian.heimes

[issue29691] Some tests fail in coverage Travis check

Serhiy Storchaka added the comment: The list of failing tests in issue29048 contains test_traceback and test_xml_etree. Thus this issue may be a part of issue29048. Some tests may be fixed since reporting issue29048. -- ___ Python tracker

[issue29691] Some tests fail in coverage Travis check

Brett Cannon added the comment: I don't think it's a duplicate because we had a passing coverage run when we initially made the migration so something got tweaked to lead to the failures. -- ___ Python tracker

[issue29737] Optimize concatenating empty tuples

New submission from Serhiy Storchaka: Since tuples are immutable, concatenating with empty tuple can be optimized by returning an opposite argument. Microbenchmarks (the difference is larger for larger tuples): $ ./python -m perf timeit --duplicate=100 -s 'a = (1, 2)' 'a + ()' Unpatched:

[issue29691] Some tests fail in coverage Travis check

Jelle Zijlstra added the comment: I'll look into creating a PR when I have some time. It would also be useful to tweak the Travis/coverage configuration so that it fails loudly if one of the tests doesn't pass in the coverage check. -- ___ Python

[issue29738] Fix memory leak in SSLSocket.getpeercert()

Changes by Olivier Vielpeau : -- pull_requests: +433 ___ Python tracker ___

[issue29700] readline memory corruption when sys.stdin fd >= FD_SETSIZE for select()

Martin Panter added the comment: “Input9)” is probably a typo for “input()”. In Python 2, sys.stdin etc are by default wrappers around ’s “stdin” etc, and can easily be wrappers around other FILE objects, so the PyOS_Readline API and Python’s “readline” module pass these objects directly

[issue29736] Optimize builtin types constructor

STINNER Victor added the comment: > I would wait until constructors be converted to Argument Clinic. There are > ready patches for some of these builtin types, patches for other are in > proggress. Do you have links for these changes? It's painful to search for Argument Clinic changes, the

[issue29740] Visual C++ CRT security update from 14 June 2011

New submission from Markus: In 14 June 2011 Microsoft released Visual C++ 2008 runtime MFC Security Update https://www.microsoft.com/en-us/download/details.aspx?id=26368 The Security Update also updates the CRT runtime (used by Python 2.7) Without the security update, Python 2.7.13 uses

[issue29741] BytesIO methods don't accept integer types, while StringIO counterparts do

New submission from Oren Milman: current state import io class IntLike(): def __init__(self, num): self._num = num def __index__(self): return self._num __int__ = __index__ io.StringIO('blah blah').read(IntLike(2)) io.StringIO('blah

[issue29736] Optimize builtin types constructor

Serhiy Storchaka added the comment: float, list: issue20185. -- ___ Python tracker ___ ___ Python-bugs-list

[issue29739] zipfile raises wrong exception for some incorrect passwords

New submission from Jack Cushman: This bug arises when attempting to unzip a password-protected zipfile using the wrong password. Usually when zipfile extraction is attempted with an incorrect password, zipfile raise `RuntimeError("Bad password for file")`. But for a small subset of

[issue29743] Closing transport during handshake process leaks open socket

New submission from Nikolay Kim: https://github.com/python/asyncio/issues/487 https://github.com/KeepSafe/aiohttp/issues/1679 -- messages: 289143 nosy: fafhrd91 priority: normal pull_requests: 436 severity: normal status: open title: Closing transport during handshake process leaks open

[issue29581] __init_subclass__ causes TypeError when used with standard library metaclasses (such as ABCMeta)

Changes by Roundup Robot : -- pull_requests: +437 ___ Python tracker ___

[issue29741] BytesIO methods don't accept integer types, while StringIO counterparts do

Martin Panter added the comment: What is the use case? Unless changing the behaviour would be useful, I think the simplest solution would be to document that the methods should only be given instances of “int”, so that it is clear that other kinds of numbers are unsupported. -- nosy:

[issue29676] C method is not profiled by lsprof

Changes by INADA Naoki : -- components: +Interpreter Core -Tests title: verbose output of test_cprofile -> C method is not profiled by lsprof ___ Python tracker

[issue9051] Improve pickle format for timezone aware datetime instances

Changes by Alexander Belopolsky : -- resolution: -> rejected stage: patch review -> resolved status: pending -> closed ___ Python tracker

[issue29723] 3.6.1rc1 adds the current directory to sys.path when running a subdirectory's __main__.py; previous versions did not

Ned Deily added the comment: Thanks for the analysis, Eryk and Nick. Then this sounds like a release blocker problem that should be fixed for 3.6.1. Steve, can you take a look, please? -- nosy: +steve.dower priority: normal -> release blocker stage: -> needs patch type: ->

[issue29742] asyncio get_extra_info() throws exception

New submission from Nikolay Kim: https://github.com/python/asyncio/issues/494 -- messages: 289138 nosy: fafhrd91 priority: normal pull_requests: 435 severity: normal status: open title: asyncio get_extra_info() throws exception versions: Python 3.5, Python 3.6, Python 3.7

[issue29742] asyncio get_extra_info() throws exception

Nikolay Kim added the comment: exception on get_extra_info() on closed ssl transport ``` Feb 18 13:18:09 btc electrumx_server.py[1732]: ERROR:ElectrumX:[15328] Traceback (most recent call last): Feb 18 13:18:09 btc electrumx_server.py[1732]: File

[issue29742] asyncio get_extra_info() throws exception

R. David Murray added the comment: Thanks for the report and PR, but your fix is not obviously correct. In general exceptions are the way in Python that errors are reported, and asking for extra_info on a closed stream is an error. The exception raised is not clear, so perhaps the asyncio

[issue29742] asyncio get_extra_info() throws exception

Nikolay Kim added the comment: get_extra_info() returns optional transport information, I think it is ok to return None for closed transport. https://github.com/python/cpython/blob/master/Lib/asyncio/transports.py#L18 I propose this feature initially, during early tulip development but now I

[issue29742] asyncio get_extra_info() throws exception

Yury Selivanov added the comment: > Thanks for the report and PR, but your fix is not obviously correct. Not for this particular asyncio method though, see asyncio PEP [1]. But there's an actual bug in the PR, instead of returning `None` we should return `default`. [1]

[issue27035] Cannot set exit code in atexit callback

Glyph Lefkowitz added the comment: I just bumped into this myself. If this really is only fixable in a major release, there ought to at least be a minor release for the *documentation* to update it to be correct. -- nosy: +glyph ___ Python tracker

[issue29742] asyncio get_extra_info() throws exception

Changes by Nikolay Kim : -- pull_requests: +438 ___ Python tracker ___ ___

[issue29742] asyncio get_extra_info() throws exception

Changes by Guido van Rossum : -- nosy: -gvanrossum ___ Python tracker ___ ___

[issue29744] memmap behavior changed

New submission from Andrea Giovannucci: The previous version 2.7.12 was returning a memmap file when slicing with a list of integers, now it returns an array. This affects the behaviour of several functions in my package. Is that an aware choice or a side product of some other change?

[issue29745] asyncio: Make pause/resume_reading idepotent and no-op for closed transports

New submission from Nikolay Kim: https://github.com/python/asyncio/issues/488 -- messages: 289147 nosy: fafhrd91 priority: normal pull_requests: 439 severity: normal status: open title: asyncio: Make pause/resume_reading idepotent and no-op for closed transports versions: Python 3.5,

[issue29741] BytesIO methods don't accept integer types, while StringIO counterparts do

Oren Milman added the comment: I don't have a use case for that. (I noticed this behavior by chance, while working on some other issue.) However, IIUC, commit 4fa88fa0ba35e25ad9be66ebbdaba9aca553dc8b, by Benjamin Peterson, Antoine Pitrou and Amaury Forgeot d'Arc, includes patching the

[issue29680] gdb/libpython.py does not work with gdb 7.2

Changes by Mariatta Wijaya : -- pull_requests: +442 ___ Python tracker ___ ___

[issue29737] Optimize concatenating empty tuples

Raymond Hettinger added the comment: Is it at all common to concatenate empty tuples? This seems like optimizing something that rarely occurs. Also, the timings can be misleading if in real code these changes cause new branch mispredictions here which can slow the common case. See

[issue29176] /tmp does not exist on Android and is used by curses.window.putwin()

Changes by Mariatta Wijaya : -- pull_requests: +441 ___ Python tracker ___ ___

[issue29739] zipfile raises wrong exception for some incorrect passwords

Changes by Xiang Zhang : -- nosy: +serhiy.storchaka ___ Python tracker ___ ___

[issue29744] memmap behavior changed

Changes by Andrea Giovannucci : -- resolution: -> not a bug ___ Python tracker ___

[issue28728] test_host_resolution in test_socket fails

Changes by Xiang Zhang : -- pull_requests: +440 ___ Python tracker ___ ___ Python-bugs-list

[issue28728] test_host_resolution in test_socket fails

Changes by Xiang Zhang : -- resolution: -> fixed stage: -> resolved status: open -> closed ___ Python tracker ___

[issue29744] memmap behavior changed

Changes by Xiang Zhang : -- stage: -> resolved status: open -> closed ___ Python tracker ___

[issue25996] Add support of file descriptor in os.scandir()

Eryk Sun added the comment: > There is no similar function taking a directory handle In 3.5+ the CRT has O_OBTAIN_DIR (0x2000) for opening a directory, i.e. to call CreateFile with backup semantics. A directory can be read via GetFileInformationByHandleEx [1] using the information classes

[issue29737] Optimize concatenating empty tuples

Serhiy Storchaka added the comment: I agree with your comments Raymond, but let me to expose my reasons. An idea of optimizing empty tuple concatenating is inspired by partial(). It concatenates two tuples of arguments and it is common when one of tuple is empty. Current C implementation

[issue29741] BytesIO methods don't accept integer types, while StringIO counterparts do

Serhiy Storchaka added the comment: Are there tests for accepting non-integers in other classes? If no then don't bother about tests. I suppose all this came from the implementation of the 'n' format unit in PyArg_Parse* functions. -- ___ Python

[issue29676] C method is not profiled by lsprof

Changes by INADA Naoki : -- resolution: -> fixed stage: -> resolved status: open -> closed ___ Python tracker ___

  1   2   >