[issue1326830] python scratchpad (IDLE)

Terry J. Reedy added the comment: I have a tem.py 'scratch' file which I use often enough that it is nearly always in the recent files list. On the other hand, many beginners do not think to do that, and Guido has said recently to design for beginners. In #19042, I propose that there should

[issue26315] Optimize mod division for ints

Changes by Mark Dickinson : -- nosy: +mark.dickinson ___ Python tracker ___ ___

[issue21152] Idle: timed autosave for shell (and maybe editor) window

Terry J. Reedy added the comment: On idledev, Jeff Silverman requested autosave, I presume for editor windows. If one is doing closely interactive development, hitting F5 to save and run every few minutes, this would not be necessary. But if not, ... I know that autosaves in other programs

[issue19042] Idle: add option to autosave 'Untitled' edit window

Terry J. Reedy added the comment: #1326830 ('python scratchpad', rejected) was about same issue, though less specific about the implementation. A different solution to the multiline statement problem would be to separate Shell into the read-only history (with prompts in a sidebar, so normal 4

[issue26319] Check recData size before unpack in zipfile

New submission from j w: Encountered on version: 2.7.3 Exception message: "error: unpack requires a string argument of length 22" Stack trace: ... elif zipfile.is_zipfile(_file):> File "/usr/lib/python2.7/zipfile.py", line 152, in is_zipfile> result = _check_zipfile(fp)> File

[issue21042] ctypes.util.find_library() should return full pathname instead of filename in linux

Tamás Bence Gedai added the comment: I fixed the ABI matching, it was a stupid mistake, thanks for pointing it out :) I think now it works as expected. I really don't find a place for testing. Maybe a new test file could be added, but I think the testing code for find_library wouldn't be more

[issue25901] make test crash in test_httpservers

Changes by Martin Panter : -- title: make test crash -> make test crash in test_httpservers ___ Python tracker ___

[issue21955] ceval.c: implement fast path for integers with a single digit

Case Van Horsen added the comment: I ran the mpmath test suite with the fastint6 and fastint5_4 patches. fastint6 results without gmpy: 0.25% faster with gmpy: 3% slower fastint5_4 results without gmpy: 1.5% slower with gmpy: 5.5% slower -- ___

[issue25994] File descriptor leaks in os.scandir()

STINNER Victor added the comment: Josh Rosenberg added the comment: > Adding a ResourceWarning even if the generator is run to completion? I'm ok to only emit the warning is the generator is not exhausted. The warning would be emited in the destructor if the generator is not closed. In

[issue26309] socketserver.BaseServer._handle_request_noblock() don't shutdwon request if verify_request is False

Martin Panter added the comment: The patch looks good to me. Are you interested in writing a unit test for it? Having said that, it might be a tricky test to write. -- nosy: +martin.panter stage: -> patch review ___ Python tracker

[issue21955] ceval.c: implement fast path for integers with a single digit

STINNER Victor added the comment: Case Van Horsen added the comment: > I ran the mpmath test suite with the fastint6 and fastint5_4 patches. > > fastint6 results > without gmpy: 0.25% faster > with gmpy: 3% slower > > fastint5_4 results > without gmpy: 1.5% slower > with gmpy: 5.5% slower I'm

[issue26243] zlib.compress level as keyword argument

Roundup Robot added the comment: New changeset dc758f51b8f5 by Martin Panter in branch 'default': Issue 26243: Forgot to update zlib doc strings in Argument Clinic https://hg.python.org/cpython/rev/dc758f51b8f5 -- ___ Python tracker

[issue26243] zlib.compress level as keyword argument

Martin Panter added the comment: I did add a What’s New entry (I hope nobody minds), and updated the doc strings in the zlib module. Thanks for the work on this Aviv. -- resolution: -> fixed stage: patch review -> resolved status: open -> closed

[issue26321] datetime.strptime fails to parse AM/PM correctly

New submission from Andrew Page: ## ## It appears that strptime is ignoring the AM/PM field ## from datetime import datetime d1 = datetime.strptime("1:00 PM", "%H:%M %p") d2 = datetime.strptime("1:00 AM", "%H:%M %p") d1.hour, d2.hour (1, 1) # d1 should be 13 d1 == d2 True # and these should not

[issue25994] File descriptor leaks in os.scandir()

Serhiy Storchaka added the comment: Updated patch adds a resource warning, tests, improves the documentation, and addresses other comments. -- Added file: http://bugs.python.org/file41869/scandir_close_2.patch ___ Python tracker

[issue26289] Optimize floor division for ints

Mark Dickinson added the comment: ... though on second thoughts, it's probably better to spell that div = -1 - (left - 1) / right to avoid compiler warnings about bit operations on signed types. A good compiler should be able to optimise `-1 - x` to `~x` anyway. --

[issue26289] Optimize floor division for ints

Serhiy Storchaka added the comment: Since the patch is changed (and may be changed further if accept Mark's suggestion), benchmarks results should be updated. Is it worth to move the optimization inside l_divmod? Will this speed up or slow down other operations that use l_divmod? --

[issue26282] Add support for partial keyword arguments in extension functions

Serhiy Storchaka added the comment: Proposed patch adds support for positional-only parameters in PyArg_ParseTupleAndKeywords(). It is simple, the most complex part is generating detailed error message for the case of calling with insufficient number of positional arguments. --

[issue26289] Optimize floor division for ints

STINNER Victor added the comment: STINNER Victor: >> This change looks related to the issue #21955. IMHO we should take the same >> decision. I mean, maybe it's better to implement the fast-path only in >> ceval.c? Or maybe in ceval.c and longobject.c? Yury Selivanov: > There is no drastic

[issue26289] Optimize floor division for ints

Changes by Mark Dickinson : -- nosy: +mark.dickinson ___ Python tracker ___ ___

[issue26320] Web documentation for 2.7 has unreadable highlights in Table of Contents

Georg Brandl added the comment: In addition, the object description headers (function names...) are not styled as they should be. Has the doc style been changed recently? -- nosy: +benjamin.peterson, ezio.melotti, georg.brandl ___ Python tracker

[issue26318] `io.open(fd, ...).name` returns numeric fd instead of None

Марк Коренберг added the comment: In particular, `io.open(fd, ...)` can not be used with tarfile, since it use it's "name" property, and think that it is real file name, and not expecte it to be int. -- ___ Python tracker

[issue26317] Build Problem with GCC + Macintosh OS X 10.11 El Capitain

Robert P Fischer added the comment: I'm trying to compile the Objective-C portions of Python (the Python Launcher app) with clang, not GCC. Why does Python insist on using GCC, even when I set the OBJC env variable to clang? That seems to be a problem that the Python project CAN fix. On Tue,

[issue26320] Web documentation for 2.7 has unreadable highlights in Table of Contents

New submission from Rory McCann: Using Firefox 41.0.2 on Ubuntu 14.04, most table of contents in the online docs have unreadable highlighted sections, where modules/methods are named. See https://docs.python.org/2/library/datetime.html for a good example, or the attached screenshot. I think

[issue26317] Build Problem with GCC + Macintosh OS X 10.11 El Capitain

Ronald Oussoren added the comment: The build fails while building the "Python Launcher" application. This is an Objective-C project. If GCC doesn't support the system headers on OSX 10.11 there's nothing the Python project can do about that. -- nosy: +ronaldoussoren

[issue26289] Optimize floor division for ints

Mark Dickinson added the comment: A slightly neater way to compute the result in the case that the signs differ is div = ~((left - 1) / right) That saves the extra multiplication and equality check. -- ___ Python tracker

[issue25994] File descriptor leaks in os.scandir()

STINNER Victor added the comment: > Updated patch adds a resource warning, tests, improves the documentation, and > addresses other comments. Review sent. -- ___ Python tracker

[issue26321] datetime.strptime fails to parse AM/PM correctly

Zachary Ware added the comment: This is a documented peculiarity of the %p format code, see the '%p' entry in the table at [1], note 3 in particular. >>> from datetime import datetime >>> datetime.strptime('1:00 PM', '%H:%M %p').hour 1 >>> datetime.strptime('1:00 PM', '%I:%M %p').hour 13 [1]

[issue26289] Optimize floor division for ints

Yury Selivanov added the comment: Attaching a new patch -- big thanks to Mark and Serhiy. > div = ~((left - 1) / right) The updated code works slightly faster - ~0.285 usec vs ~0.3 usec. -- Added file: http://bugs.python.org/file41870/floor_div_4.patch

[issue26289] Optimize floor division for ints

Yury Selivanov added the comment: > Is it worth to move the optimization inside l_divmod? Will this speed up or > slow down other operations that use l_divmod? Attaching a new patch -- fast_divmod.patch It combines patches for this issue and issue #26315. Individual timeit benchmarks work as

[issue25985] Use sys.version_info instead of sys.version

Serhiy Storchaka added the comment: Update patch uses less cryptic formatting expressions (thanks SilentGhost for suggestion), fixes Makefile.pre.in (thanks Andrew), and fixes yet few files. -- Added file: http://bugs.python.org/file41872/use_version_info_2.patch

[issue26289] Optimize floor division for ints

Mark Dickinson added the comment: Just for the record, there's a less-branchy analog of the floor division code I suggested for modulo. In Python-land, it looks like this: def propermod(a, b): # mimic the C setup assert a != 0 and b != 0 left = abs(a) size_a = -1 if a < 0 else

[issue25848] Tkinter tests failed on Windows buildbots

Zachary Ware added the comment: I added the note to the devguide instead of pep101, and will be removing the 3.4 builders shortly. -- resolution: -> out of date stage: -> resolved status: open -> closed ___ Python tracker

[issue25848] Tkinter tests failed on Windows buildbots

Roundup Robot added the comment: New changeset 5430f4fd4ca7 by Zachary Ware in branch 'default': Issue #25848: Add section on running a build on all custom builders https://hg.python.org/devguide/rev/5430f4fd4ca7 -- nosy: +python-dev ___ Python

[issue26322] Missing docs for typing.Set

Changes by Ezio Melotti : -- nosy: +ezio.melotti stage: -> needs patch type: -> enhancement ___ Python tracker ___

[issue26289] Optimize floor division for ints

Yury Selivanov added the comment: > mod = left % right if size_a == size_b else right - 1 - (left - 1) % right This works, Mark! Again, the difference in performance is very subtle, but the code is more compact. -m timeit -s "x=22331" "x//2;x//-3;x//4;x//5;x//-6;x//7;x//8;x//-99;x//100;"

[issue26289] Optimize floor division for ints

Changes by Yury Selivanov : Added file: http://bugs.python.org/file41876/fast_divmod_3.patch ___ Python tracker ___

[issue1767511] SocketServer.DatagramRequestHandler with empty response

Martin Panter added the comment: Sorry that nobody has responded to your patch in almost a decade. :( Unfortunately in the meantime the test file seems to have been completely rewritten. For the original problem, I think it would be better if we just documented that this is the way it works.

[issue26279] time.strptime does not properly convert out-of-bounds values

Tamás Bence Gedai added the comment: This is connected to http://bugs.python.org/issue23718 I submitted a patch that hopefully solves this problem. -- nosy: +beng94 ___ Python tracker

[issue26204] compiler: ignore constants used as statements (don't emit LOAD_CONST+POP_TOP)

Jim Jewett added the comment: I think the warning was helpful; it just had confusing wording. Instead of: """ >>> def f(): ... False ... :2: SyntaxWarning: ignore constant statement """ perhaps: """ >>> def f(): ... False ... :2: SyntaxWarning: ignoring constant statement """ or even: """

[issue23718] strptime() can produce invalid date with negative year day

Tamás Bence Gedai added the comment: I made a patch, that solves this issue. The problem was that there wasn't any Sunday (%w = 0) on the 0th week in 2015. 2015 started on Thursday, therefore the first Sunday was on 2014.12.28. julian variable is used to set the tm_yday, which was a minus

[issue26322] Missing docs for typing.Set

New submission from Guido van Rossum: The typing docs don't seem to mention Set (which is a bit of an anomaly, since it corresponds to builtins.set, not to collections.abc.Set -- the latter is typing.AbstractSet). Also, AbstractSet is mentioned twice. (And the second occurrence somehow

[issue25994] File descriptor leaks in os.scandir()

Serhiy Storchaka added the comment: Updated patch addresses new Victor's comments. -- Added file: http://bugs.python.org/file41873/scandir_close_3.patch ___ Python tracker

[issue21955] ceval.c: implement fast path for integers with a single digit

Yury Selivanov added the comment: > Maybe we should just close the issue? I'll take a closer look at gmpy later. Please don't close. -- ___ Python tracker

[issue26324] sum() incorrect on negative zeros

New submission from Antoine Pitrou: >>> sum([-0.0,-0.0]) 0.0 -- components: Interpreter Core messages: 259961 nosy: eric.smith, lemburg, mark.dickinson, pitrou, stutzbach priority: low severity: normal status: open title: sum() incorrect on negative zeros type: behavior versions: Python

[issue26324] sum() incorrect on negative zeros

Antoine Pitrou added the comment: "Expect" is a strong word here ;-) The problem is the start value is implicit and is supposed to be a "good default". Of course, solving this consistently may not be trivial. -- ___ Python tracker

[issue23453] Opening a stream with tarfile.open() triggers a TypeError: can't concat bytes to str error

Changes by Martin Panter : -- stage: -> needs patch type: crash -> behavior ___ Python tracker ___

[issue26324] sum() incorrect on negative zeros

Emanuel Barry added the comment: This is consistent with the advertised equivalence of sum(): >>> -0.0 + -0.0 + 0 0.0 This works as you'd expect: >>> sum([-0.0, -0.0], -0.0) -0.0 It has a start value of 0, and I don't think we should special-case this. -- nosy: +ebarry

[issue26323] Add a assert_called() method for mock objects

Changes by Amit Saha : -- type: -> enhancement ___ Python tracker ___ ___

[issue26324] sum() incorrect on negative zeros

Emanuel Barry added the comment: "Not trivial" might be an understatement. You need the start value to begin summing the items, but you may face with a non-repeatable generator with side-effects. Sure, you could special-case some builtins, but I'm not too keen on adding special cases for such

[issue26324] sum() incorrect on negative zeros

Ethan Furman added the comment: If one is using sum on floats, and a list of -0.0's is a possibility, and the sign matters... well, I would hope that one is using -0.0 as the start value. I don't think this is worth "fixing". -- nosy: +ethan.furman

[issue21996] gettarinfo method does not handle files without text string names

Changes by Martin Panter : -- dependencies: +Tarfile using fstat on GZip file object ___ Python tracker ___

[issue26323] Add a assert_called() method for mock objects

New submission from Amit Saha: Would a patch for adding a assert_called() method to mocked objects be welcome for inclusion? We do have a assert_not_called() method, so I think this may be a good idea. Please let me know and I will work on it. -- components: Library (Lib) messages:

[issue25593] _sock_connect_cb can be called twice resulting in InvalidStateError

Alexander Mohr added the comment: update: its unrelated to the number of sessions or SSL, but instead to the number of concurrent aiohttp requests. When set to 500, I get the error, when set to 100 I do not. -- ___ Python tracker

[issue26320] Web documentation for 2.7 has unreadable highlights in Table of Contents

Benjamin Peterson added the comment: I upgraded Sphinx a month or so ago. Could that do it? On Tue, Feb 9, 2016, at 05:30, Georg Brandl wrote: > > Georg Brandl added the comment: > > In addition, the object description headers (function names...) are not > styled as they should be. > > Has

[issue26318] `io.open(fd, ...).name` returns numeric fd instead of None

Martin Panter added the comment: I agree that the business with “name” is a mess. I don’t know what the best solution is. FileIO.name is already documented as being a file descriptor in some cases. Perhaps we should document that the Buffered and Text wrappers inherit the wrapped name, but

[issue25924] investigate if getaddrinfo(3) on OSX is thread-safe

A. Jesse Jiryu Davis added the comment: Martin, here's a third patch, "try 3", which does a runtime version check instead of compile-time. It's a bit complex, compared to "try 2", which did a compile-time check instead. I do NOT think this extra complexity is worth it to support Pythons that

[issue8013] time.asctime segfaults when given a time in the far future

Mandar Gokhale added the comment: [Strictly speaking, this is actually issue #10563, but that was marked superseded by the changes for this issue, hence the comment here.] The 5-digit year still displays an extra newline in Python 2.7.11 (there's extra whitespace on OSX as well, but that

[issue26318] `io.open(fd, ...).name` returns numeric fd instead of None

Марк Коренберг added the comment: in any case, passing INTEGER as name is wrong thing, since (due to duck typing), name should be a string. -- ___ Python tracker

[issue25994] File descriptor leaks in os.scandir()

Martin Panter added the comment: About testing that list(iterator) is empty after the iterator is closed, IMO this is an implementation detail. It would be equally (or more) sensible to raise an exception, like proposed for “async def” coroutines in Issue 25887. I suppose the main purpose of

[issue5824] SocketServer.DatagramRequestHandler Broken under Linux

Martin Panter added the comment: It is not that wfile.getvalue() returns None, it is that self.client_address is None. If the client explicitly binds to a name, then the server gets that name in the client_address, and then everything seems to work properly for me on Linux. On Linux, using

[issue25924] investigate if getaddrinfo(3) on OSX is thread-safe

Ned Deily added the comment: I agree that the approach in "try 2" is fine and the runtime check in "try 3" is overkill. While it is possible to do so, we've never really supported building on an OS X release n for release m, where m < n, without using the m SDK on n (and in particular for

[issue25593] _sock_connect_cb can be called twice resulting in InvalidStateError

Alexander Mohr added the comment: I'm not sure if you guys are still listening on this closed bug but I think I've found another issue ;) I'm using python 3.5.1 + asyncio 3.4.3 with the latest aiobotocore (which uses aiohttp 0.21.0) and had two sessions (two TCPConnectors), one doing a

[issue26136] DeprecationWarning for PEP 479 (generator_stop)

Roundup Robot added the comment: New changeset 73f89182bb4d by Martin Panter in branch 'default': Issue #26136: Upgrade the generator_stop warning to DeprecationWarning https://hg.python.org/cpython/rev/73f89182bb4d -- nosy: +python-dev ___ Python

[issue25924] investigate if getaddrinfo(3) on OSX is thread-safe

Martin Panter added the comment: I am happy to defer to Ned who probably knows a lot more than me about the OS X situation. FTR it was the code added in 2006 by r45660 (Issue 1471925) that motivated my concern. Since that was added by Ronald, it would be good to hear his opinion :)

[issue14307] Make subclassing SocketServer simpler for non-blocking frameworks

Martin Panter added the comment: FTR the python-dev discussion looks like it is . The argument for this change seems to be to support users monkey-patching the socket module, but not monkey-patching the select module. But this seems

[issue25698] The copy_reg module becomes unexpectedly empty in test_cpickle

Benjamin Peterson added the comment: That seems fine except you might as well just put the GetWithError prototype in the header file. -- ___ Python tracker

[issue26304] Fix “allows to ” in documentation

Roundup Robot added the comment: New changeset 818e91105418 by Martin Panter in branch '3.5': Issue #26304: Change "allows to " to "allows ing" or similar https://hg.python.org/cpython/rev/818e91105418 New changeset d01abc2d5197 by Martin Panter in branch 'default': Issue #26304: Merge doc

[issue26304] Fix “allows to ” in documentation

Martin Panter added the comment: Thanks for the reviews, and spotting my typo. I also sneaked in eight more changes (“allowing to”), to asyncio-eventloop.rst, ctypes.rst, whatsnew/3.3.rst, howto/pyporting.rst, Misc/NEWS, and Misc/HISTORY. -- resolution: -> fixed stage: patch review

[issue26204] compiler: ignore constants used as statements (don't emit LOAD_CONST+POP_TOP)

STINNER Victor added the comment: Sorry you are late :-) I started a thread on python-dev and it was decided to let linters handle this warning. -- ___ Python tracker

[issue26323] Add a assert_called() method for mock objects

Florian Bruhin added the comment: I agree this would be useful. "assert themock.called" (or self.assertTrue) works, but always trips me up. -- nosy: +The Compiler ___ Python tracker

[issue26323] Add a assert_called() method for mock objects

Changes by Florian Bruhin : -- nosy: +michael.foord ___ Python tracker ___ ___

[issue26136] DeprecationWarning for PEP 479 (generator_stop)

Martin Panter added the comment: Thanks for your work Anish -- resolution: -> fixed stage: patch review -> resolved status: open -> closed ___ Python tracker

[issue25937] DIfference between utf8 and utf-8 when i define python source code encoding.

Serhiy Storchaka added the comment: I think the correct way is not add "utf8" to special case, but removes "utf-8". Here is a patch. -- components: +Interpreter Core stage: -> patch review type: -> behavior Added file: http://bugs.python.org/file41879/bad_utf8.patch

[issue26304] Fix “allows to ” in documentation

Roundup Robot added the comment: New changeset 4cfc19f667aa by Martin Panter in branch '2.7': Issue #26304: Change "allows to " to "allows ing" or similar https://hg.python.org/cpython/rev/4cfc19f667aa -- ___ Python tracker

[issue26324] sum() incorrect on negative zeros

Mark Dickinson added the comment: For what it's worth, NumPy has exactly the same behaviour: >>> np.array([-0.0, -0.0]).sum() 0.0 ... which is a bit surprising, given that this is a much easier problem to fix when you know the type of everything in the array in advance. The Decimal type has

[issue26324] sum() incorrect on negative zeros

Raymond Hettinger added the comment: > I think it's best left as it is I concur. -- nosy: +rhettinger ___ Python tracker ___

[issue26253] tarfile in stream mode always set zlib compression level to 9

Martin Panter added the comment: Actually it’s not really obvious from the signatures, but in the middle of the tarfile.open() documentation it says “. . . tarfile.open() accepts the keyword argument _compresslevel_”, so it should already be possible. --

[issue25496] tarfile: Default value for compresslevel is not documented

Martin Panter added the comment: For gzip compression, there is also a proposal to change the default: Issue 26253. Anyway, the patch looks good for now. -- nosy: +martin.panter stage: needs patch -> patch review versions: +Python 2.7 -Python 3.4