[issue29029] Faster positional arguments parsing in PyArg_ParseTupleAndKeywords

INADA Naoki added the comment: LGTM again. -- ___ Python tracker ___ ___ Python-bugs-list mailing list

[issue16623] argparse help formatter does not honor non-breaking space

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

[issue29011] No entry Deque in typing.py

Changes by Raymond Hettinger : -- resolution: -> fixed status: open -> closed ___ Python tracker ___

[issue29011] No entry Deque in typing.py

Roundup Robot added the comment: New changeset dfefbf6f6c73 by Raymond Hettinger in branch '3.5': Issue #29011: Fix an important omission by adding Deque to the typing module. https://hg.python.org/cpython/rev/dfefbf6f6c73 -- nosy: +python-dev ___

[issue28638] Creating namedtuple is too slow to be used in common stdlib (e.g. functools)

Raymond Hettinger added the comment: Sorry INADA but I think this is all a foolish and inconsequential optimization that complicates the code in a way that isn't worth it (saving only a 1/2 millisecond in a single import. Also, we don't want the argument clinic code to start invading the pure

[issue29286] Use METH_FASTCALL in str methods

Serhiy Storchaka added the comment: Oh, I have wrote almost the same patch before going to sleep yesteday! ;) But the building crashed (likely due to a bug in _PyStack_UnpackDict()) and it was too late to resolve this. I would prefer to rename "l" to "nargs" in PyArg_UnpackStack_impl and

[issue16623] argparse help formatter does not honor non-breaking space

Xiang Zhang added the comment: Ahh, sorry. I made a mistake. It's not the same as 29290. This is about 2.7. Although #20491 is closed but 2.7 seems not patched. Nosy Serhiy. -- nosy: +serhiy.storchaka resolution: fixed -> stage: resolved -> needs patch status: closed -> open

[issue16623] argparse help formatter does not honor non-breaking space

Xiang Zhang added the comment: #29290 reports the same problem as here. -- nosy: +xiang.zhang resolution: -> fixed stage: -> resolved status: open -> closed superseder: -> argparse breaks long lines on NO-BREAK SPACE ___ Python tracker

[issue29290] argparse breaks long lines on NO-BREAK SPACE

Xiang Zhang added the comment: I think this is a regression when coming to 3.x. In 2.7, r'\s+' is by default in ASCII mode and won't match unicode non-breaking spaces. In 3.x it's by default unicode mode so non-breaking spaces are replaced by spaces. I think we can just use [ \t\n\r\f\v]+.

[issue29290] argparse breaks long lines on NO-BREAK SPACE

Xiang Zhang added the comment: textwrap has been fixed in #20491 but this problem still exists. The reason seems to be that argparse replaces the non-break spaces with spaces: before self.whitespace_matcher.sub 'Disable default font-style: condensed. Also disables "M+\\xa01M" condensed

[issue29290] argparse breaks long lines on NO-BREAK SPACE

Martin Panter added the comment: Maybe a duplicate of Issue 16623 -- nosy: +martin.panter ___ Python tracker ___

[issue29290] argparse breaks long lines on NO-BREAK SPACE

Steven D'Aprano added the comment: Here's a slightly simpler demo, without the (fortunately harmless) typo. -- Added file: http://bugs.python.org/file46311/argparse_nobreak.py ___ Python tracker

[issue29290] argparse breaks long lines on NO-BREAK SPACE

Changes by Steven D'Aprano : Removed file: http://bugs.python.org/file46310/argparse_nobreak.py ___ Python tracker ___

[issue29290] argparse breaks long lines on NO-BREAK SPACE

New submission from Steven D'Aprano: argparse help incorrectly breaks long lines on U+u00A0 NO-BREAK SPACE. The attached script has been run on Python 3.5.3rc1 in a terminal window 80 columns wide, and it produces output:: usage: argparse_nobreak.py [-h] [--no-condensed] optional

[issue29289] Convert OrderedDict methods to Argument Clinic

STINNER Victor added the comment: Oh, I missed Naoki's comment on the review :-/ I got the notification of this review after I pushed the change. I will fix the docstring later. -- ___ Python tracker

[issue29289] Convert OrderedDict methods to Argument Clinic

STINNER Victor added the comment: Raymond: " The argument clinic doesn't cope well with optional arguments without a default value." Right, it's a bug in AC. I will try to fix it later. In the meanwhile, I pushed the first uncontroversal part. --

[issue29289] Convert OrderedDict methods to Argument Clinic

Roundup Robot added the comment: New changeset 9f7d16266928 by Victor Stinner in branch 'default': Convert some OrderedDict methods to Argument Clinic https://hg.python.org/cpython/rev/9f7d16266928 -- nosy: +python-dev ___ Python tracker

[issue29259] Add tp_fastcall to PyTypeObject: support FASTCALL calling convention for all callable objects

STINNER Victor added the comment: Patch version 2: * Add slot_tp_fastcall(): __call__() wrapper using the FASTCALL calling convention * Add Py_TPFLAGS_HAVE_FASTCALL flag * Remove completely the "cached args" optimization from property_descr_get() * Rebase the patch serie on top of the default

[issue29259] Add tp_fastcall to PyTypeObject: support FASTCALL calling convention for all callable objects

STINNER Victor added the comment: > Additionally, there are not enough METH_FASTCALL CFunctions. I pushed many changes tonight which convert many C functions to METH_FASTCALL. > ..., by hand or by AC. IMHO performance is a good motivation to convert code to Argument Clinic ;-) But

[issue29289] Convert OrderedDict methods to Argument Clinic

INADA Naoki added the comment: LGTM, except pop(). -- ___ Python tracker ___ ___ Python-bugs-list mailing

[issue29286] Use METH_FASTCALL in str methods

Roundup Robot added the comment: New changeset 3bf78c286daf by Victor Stinner in branch 'default': Add _PyArg_UnpackStack() function helper https://hg.python.org/cpython/rev/3bf78c286daf New changeset 905e902bd47e by Victor Stinner in branch 'default': Argument Clinic: Use METH_FASTCALL for

[issue29289] Convert OrderedDict methods to Argument Clinic

Raymond Hettinger added the comment: This patch segfaults for me (Mac OS 10.12.2 with the default clang-800.0.42.1). For now, pop() should be omitted. The argument clinic doesn't cope well with optional arguments without a default value. The generated help "pop(self, /, key, default=None)"

[issue29286] Use METH_FASTCALL in str methods

STINNER Victor added the comment: Naoki> This patch makes AC produces more FASTCALL instead of VARARGS. Oh, funny, I wrote the same patch :-) (almost) > When looking AC output, one downside is it produces many consts like: > +static const char * const _keywords[] = {"", "", "", "", "",

[issue29286] Use METH_FASTCALL in str methods

Roundup Robot added the comment: New changeset d07fd6e6d449 by Victor Stinner in branch 'default': Rename _PyArg_ParseStack to _PyArg_ParseStackAndKeywords https://hg.python.org/cpython/rev/d07fd6e6d449 New changeset 01c57ef1b651 by Victor Stinner in branch 'default': Add _PyArg_ParseStack()

[issue29286] Use METH_FASTCALL in str methods

STINNER Victor added the comment: ac_more_fastcalls.patch: Hum, I guess that your change on _PyStack_UnpackDict() is related to a bug in the function prototype. The function is unable to report failure if args is NULL. It changed the API in the change 38ab8572fde2. --

[issue29286] Use METH_FASTCALL in str methods

INADA Naoki added the comment: This patch makes AC produces more FASTCALL instead of VARARGS. When looking AC output, one downside is it produces many consts like: +static const char * const _keywords[] = {"", "", "", "", "", NULL}; -- Added file:

[issue9216] FIPS support for hashlib

Robert Collins added the comment: A few thoughts; usedforsecurity=xxx seems awkward: I wouldn't want, as a user of hashlib, to have to put that in literally every use I make of it. If I understand the situation correctly, the goal is for both linters, and at runtime, identification of the

[issue29029] Faster positional arguments parsing in PyArg_ParseTupleAndKeywords

STINNER Victor added the comment: > Unpatched: Median +- std dev: 2.01 us +- 0.09 us > Patched:Median +- std dev: 1.23 us +- 0.03 us Oh wow, impressive speedup! As usual, good job Serhiy ;-) PyArg_ParseTupleAndKeywords-faster-positional-args-parse-2.patch: LGTM! Can you please push it?

[issue29029] Faster positional arguments parsing in PyArg_ParseTupleAndKeywords

STINNER Victor added the comment: Oh, I missed this issue. Since I made minor cleanup recently, I took the liberty of rebasing your patch. I also pushed your "keywords => kwargs" change, just to make the patch simpler to review. -- nosy: +haypo Added file:

[issue29029] Faster positional arguments parsing in PyArg_ParseTupleAndKeywords

Roundup Robot added the comment: New changeset a36e60c5fee0 by Victor Stinner in branch 'default': Rename keywords to kwargs in getargs.c https://hg.python.org/cpython/rev/a36e60c5fee0 -- nosy: +python-dev ___ Python tracker

[issue29289] Convert OrderedDict methods to Argument Clinic

STINNER Victor added the comment: Converted methods: * fromkeys() (@classmethod) * setdefault() * pop() * popitem() * move_to_end() -- ___ Python tracker

[issue29289] Convert OrderedDict methods to Argument Clinic

New submission from STINNER Victor: Attached patch converts methods using METH_VARARGS|METH_KEYWORDS calling convention to METH_FASTCALL using Argument Clinic. This calling convention is faster, and Argument Clinic provides better docstring. See also issue #29263 "Implement

[issue29286] Use METH_FASTCALL in str methods

INADA Naoki added the comment: Parsing positional arguments for METH_KEYWORDS and METH_FASTCALL can be faster by http://bugs.python.org/issue29029 -- ___ Python tracker

[issue29288] Lookup Error while importing idna from a worker thread

New submission from Ilya Kulakov: See this post: https://github.com/kennethreitz/requests/issues/3578 The current workaround for requests is to have a no-op import somewhere in the code. However, that doesn't really work for us: our python and stdlib are bundled by pyqtdeploy as in Qt

[issue29287] IDLE needs syntax highlighting for f-strings

New submission from Raymond Hettinger: Follow the lead from Vim, MagicPython, and PyCharm. Needs separate colorization to make the expression distinct from the rest of the string. Needs close-brace matching. Would be desirable to have autocompletion as well.

[issue29286] Use METH_FASTCALL in str methods

STINNER Victor added the comment: > What is python-patch? It's Python patched with attached unicode_allow_kw.patch: allow to pass parameters as keywords for Unicode methods. -- keywords: +patch Added file: http://bugs.python.org/file46305/unicode_allow_kw.patch

[issue27659] Prohibit implicit C function declarations

Martin Panter added the comment: I would say it is more important to fit in with the surrounding style than mindlessly follow PEP 7. IMO the indentation in the configure script is a mess, but if we fix it up, it should probably be done separately to adding this extra flag. --

[issue29282] Fused multiply-add: proposal to add math.fma()

Mark Dickinson added the comment: Here's a pure Python reference implementation, with tests. -- Added file: http://bugs.python.org/file46304/fma_reference.py ___ Python tracker

[issue29286] Use METH_FASTCALL in str methods

Serhiy Storchaka added the comment: What is python-patch? -- ___ Python tracker ___ ___ Python-bugs-list

[issue29230] Segfault in sqlite3

Ned Deily added the comment: > I'm still trying to find a good way to get pyenv to recognize and use the > newer library Thanks for confirming that the problem was due to the version of sqlite3 in use. If you haven't already, suggest you open an issue on the pyenv project tracker or ask

[issue29230] Segfault in sqlite3

Daniel Fackrell added the comment: I was using versions of python from pyenv, but they may have been using the system version of sqlite3. I recreated the virtualenv using python 2.7.13 installed with Brew, and it doesn't crash. sqlite3.sqlite_version went from 3.8.10.2 to 3.16.2, which seems

[issue29286] Use METH_FASTCALL in str methods

Yury Selivanov added the comment: > * Allow passing arguments as keywoards: str.replace(old='a', new='b') I think Guido explicitly stated that he doesn't like the idea to always allow keyword arguments for all methods. I.e. `str.find('aaa')` just reads better than `str.find(needle='aaa')`.

[issue29286] Use METH_FASTCALL in str methods

STINNER Victor added the comment: Background: see also the old issue #17170 closed as rejected: "string method lookup is too slow". -- ___ Python tracker

[issue29286] Use METH_FASTCALL in str methods

New submission from STINNER Victor: Changes 27dc9a1c061e and 01b06ca45f64 converted the str (Unicode) methods to Argument Clinic, cool! But str methods taking more than one argument use positional-only arguments. Currently, Argument Clinic doesn't use METH_FASTCALL for these methods, but

[issue29285] Unicode errors occur inside of multi-line comments

Zachary Ware added the comment: Triple-quoted strings are strings, not multi-line comments, though people do abuse them for that purpose sometimes. Changing this behavior would be a huge change for next to no benefit, so I'm closing the issue. If you insist on using a triple-quoted string as

[issue29286] Use METH_FASTCALL in str methods

STINNER Victor added the comment: The issue #29263 "Implement LOAD_METHOD/CALL_METHOD for C functions" should also optimize C methods even more. -- ___ Python tracker

[issue29285] Unicode errors occur inside of multi-line comments

New submission from John Taylor: I am using Python 3.5.2 on OS X 10.11.6. The same error occurs with 3.5.2 on Windows 10. When I run the attached code, I get this error: SyntaxError: (unicode error) 'unicodeescape' codec can't decode bytes in position 23-24: truncated \u escape I think

[issue29284] Include thread_name_prefix in the concurrent.futures.ThreadPoolExecutor example 17.4.2.1

New submission from John Taylor: Please include how to use the thread_name_prefix method argument (new to Python 3.6) in the example: 17.4.2.1. ThreadPoolExecutor Example -- assignee: docs@python components: Documentation messages: 285572 nosy: docs@python, jftuga priority: normal

[issue29259] Add tp_fastcall to PyTypeObject: support FASTCALL calling convention for all callable objects

Roundup Robot added the comment: New changeset 2ccdf2b3819d by Victor Stinner in branch 'default': Optimize _PyCFunction_FastCallKeywords() https://hg.python.org/cpython/rev/2ccdf2b3819d -- nosy: +python-dev ___ Python tracker

[issue29259] Add tp_fastcall to PyTypeObject: support FASTCALL calling convention for all callable objects

INADA Naoki added the comment: Additionally, there are not enough METH_FASTCALL CFunctions. I'm interested in performance difference between METH_VARARGS and METH_FASTCALL. If METH_FASTCALL is significant faster than VARARGS, we can convert some common methods from VARARGS to FASTCALL, by

[issue28911] Clarify the behaviour of assert_called_once_with

Michael Foord added the comment: I like the documentation improvement, thank you. (The purpose of the method is to combine an assert about the arguments the method was called with and an assertion that it was only called once. To change the semantics would be both undesirable and backwards

[issue29259] Add tp_fastcall to PyTypeObject: support FASTCALL calling convention for all callable objects

STINNER Victor added the comment: > Strange, I'm unable to reproduce the result on a different computer: (...) Ah, the benchmark result change after a reboot. I reran the same benchmark (LTO+PGO on speed-python) after a reboot: --- haypo@speed-python$ python3 -m perf compare_to

[issue27659] Prohibit implicit C function declarations

Chi Hsuan Yen added the comment: Thanks for the comment and sorry for the mistake. Here's another updated patch. In PEP7: > Use 4-space indents and no tabs at all. Does that apply to configuration files, too? -- Added file:

[issue29282] Fused multiply-add: proposal to add math.fma()

Mark Dickinson added the comment: > Isn't the behaviour of quiet NaNs kindof implementation-dependent already? Not as far as IEEE 754-2008 is concerned, and not as far as Python's math module is concerned, either: handling of special cases is, as far as I know, both consistent across

[issue29282] Fused multiply-add: proposal to add math.fma()

Antoine Pitrou added the comment: Isn't the behaviour of quiet NaNs kindof implementation-dependent already? -- ___ Python tracker ___

[issue29282] Fused multiply-add: proposal to add math.fma()

Mark Dickinson added the comment: An implementation note: IEEE 754-2008 leaves it to the implementation to decide whether FMA operations like: 0 * inf + nan and inf * 0 + nan (where nan represents a quiet NaN and the inf and 0 can have arbitrary signs) signal the invalid operation

[issue20180] Derby #11: Convert 50 sites to Argument Clinic across 9 files

Roundup Robot added the comment: New changeset 01b06ca45f64 by INADA Naoki in branch 'default': Issue #20180: forgot to update AC output. https://hg.python.org/cpython/rev/01b06ca45f64 -- ___ Python tracker

[issue20180] Derby #11: Convert 50 sites to Argument Clinic across 9 files

INADA Naoki added the comment: Uhhh! I'm sorry. -- ___ Python tracker ___ ___ Python-bugs-list mailing list

[issue20180] Derby #11: Convert 50 sites to Argument Clinic across 9 files

Serhiy Storchaka added the comment: Seems you haven't updated generated AC files. -- ___ Python tracker ___

[issue27400] Datetime NoneType after calling Py_Finalize and Py_Initialize

Denny Weinberg added the comment: Any news here? 3.6.0 is also affected by this bug. -- ___ Python tracker ___

[issue20180] Derby #11: Convert 50 sites to Argument Clinic across 9 files

INADA Naoki added the comment: Thank you, Martin and Serhiy. As http://bugs.python.org/issue20180#msg247987 , remained module to converted is transmogrify.h. But docstrings of methods in transmogrify.h were moved to bytes_methods.c in issue26765. May I close this issue for now? --

[issue20180] Derby #11: Convert 50 sites to Argument Clinic across 9 files

Roundup Robot added the comment: New changeset 27dc9a1c061e by INADA Naoki in branch 'default': Issue #20180: convert unicode methods to AC. https://hg.python.org/cpython/rev/27dc9a1c061e -- nosy: +python-dev ___ Python tracker

[issue29282] Fused multiply-add: proposal to add math.fma()

Mark Dickinson added the comment: > The performance argument unlikely is applicable in this case. Agreed. This is mainly about accuracy, not speed: the FMA operation is a fundamental building block for floating-point arithmetic, is useful in some numerical algorithms, and essential in others

[issue1294959] Problems with /usr/lib64 builds.

jan matejek added the comment: Attached is a patch that I'd like to propose for inclusion. It introduces a new configure option "--with-custom-platlibdir=", which defaults to `basename $libdir`. This is converted to makefile variable "platlibdir", which is used in getpath.c to generate value

[issue29283] duplicate README in site-packages

New submission from jan matejek: Lib/site-packages directory now contains README.txt in addition to README. Both files are identical. One of them should probably go away? -- components: Installation messages: 28 nosy: matejcik priority: normal severity: normal status: open title:

[issue20180] Derby #11: Convert 50 sites to Argument Clinic across 9 files

Serhiy Storchaka added the comment: LGTM again. -- ___ Python tracker ___ ___ Python-bugs-list mailing list

[issue29282] Fused multiply-add: proposal to add math.fma()

Serhiy Storchaka added the comment: The performance argument unlikely is applicable in this case. I suppose that an overhead of function call in Python make two operators faster than one function call. Alternatives to fma() for exact computations are integer arithmetic (if all values can be

[issue29282] Fused multiply-add: proposal to add math.fma()

Antoine Pitrou added the comment: I don't know. If I want to compute a dot product, the first thing I'll do is import Numpy and then use the `@` operator on Numpy arrays. -- ___ Python tracker

[issue29282] Fused multiply-add: proposal to add math.fma()

Juraj Sukop added the comment: I would say because it has wide applicability, especially considering the amount of code it adds. It is similar in spirit to `copysign` or `fsum` which are already there and C99 includes it anyway. For simpler things like dot product or polynomial evaluation

[issue29282] Fused multiply-add: proposal to add math.fma()

Antoine Pitrou added the comment: What's the point of adding this in the math module rather than a more specialized library like Numpy? -- nosy: +pitrou ___ Python tracker

[issue29282] Fused multiply-add: proposal to add math.fma()

STINNER Victor added the comment: Thread on python-ideas: https://mail.python.org/pipermail/python-ideas/2017-January/044300.html -- ___ Python tracker

[issue29282] Fused multiply-add: proposal to add math.fma()

Changes by STINNER Victor : -- nosy: +haypo ___ Python tracker ___ ___

[issue29262] Provide a way to check for *real* typing.Union instances

flying sheep added the comment: Cool! This set of basic initial check will consist of all the is_* functions that were mentioned right? FWIW I also think that this is the way to go, as it’s not obvious if the semantics should be “conforms to this type annotation” or “is a type annotation of

[issue29282] Fused multiply-add: proposal to add math.fma()

Changes by STINNER Victor : -- nosy: +mark.dickinson ___ Python tracker ___ ___

[issue29282] Fused multiply-add: proposal to add math.fma()

Changes by Juraj Sukop : Added file: http://bugs.python.org/file46300/setup.py ___ Python tracker ___

[issue29282] Fused multiply-add: proposal to add math.fma()

Changes by Juraj Sukop : Added file: http://bugs.python.org/file46298/xmathmodule.c ___ Python tracker ___

[issue29282] Fused multiply-add: proposal to add math.fma()

Changes by Juraj Sukop : Added file: http://bugs.python.org/file46299/example.py ___ Python tracker ___

[issue29282] Fused multiply-add: proposal to add math.fma()

New submission from Juraj Sukop: Fused multiply-add (henceforth FMA) is an operation which calculates the product of two numbers and then the sum of the product and a third number with just one floating-point rounding. More concretely: r = x*y + z The value of `r` is the same as if the

[issue20180] Derby #11: Convert 50 sites to Argument Clinic across 9 files

Changes by INADA Naoki : Added file: http://bugs.python.org/file46297/unicodeobject.c.v7.patch ___ Python tracker ___

[issue29274] Change “tests cases” → “test cases”

Michael Foord added the comment: Yep, looks like an improvement to me! -- ___ Python tracker ___ ___

[issue29281] json.loads documentation missing "versionchanged" statement

New submission from Alexey Popravka: json.loads function was changed in Python 3.6 to accept bytes and bytearrays, however documentation is missing `versionchanged` block describing this changes. -- assignee: docs@python components: Documentation messages: 285545 nosy: Alexey Popravka,

[issue26110] Speedup method calls 1.2x

Changes by INADA Naoki : -- status: open -> closed ___ Python tracker ___ ___

[issue26110] Speedup method calls 1.2x

Roundup Robot added the comment: New changeset a6241b2073c6 by INADA Naoki in branch 'default': Issue #26110: Add document for LOAD_METHOD and CALL_METHOD opcode. https://hg.python.org/cpython/rev/a6241b2073c6 -- ___ Python tracker

[issue10513] sqlite3.InterfaceError after commit

Roundup Robot added the comment: New changeset 74a68d86569c by Benjamin Peterson in branch '2.7': revert dd13098a5dc2 (#29006, #10513) https://hg.python.org/cpython/rev/74a68d86569c -- ___ Python tracker

[issue29006] 2.7.13 _sqlite more prone to "database table is locked"

Roundup Robot added the comment: New changeset 74a68d86569c by Benjamin Peterson in branch '2.7': revert dd13098a5dc2 (#29006, #10513) https://hg.python.org/cpython/rev/74a68d86569c -- ___ Python tracker

[issue29057] Compiler failure on Mac OS X - sys/random.h

Larry Hastings added the comment: Releasing 3.5.3 even though technically this is an open release blocker. IIUC the fix is checked in, and fixed the issue for OS X. We don't know whether or not it is also fixed on OpenBSD, because we don't know anybody running OpenBSD, and nobody contacted