[issue30064] BaseSelectorEventLoop.sock_{recv, sendall}() don't remove their callbacks when canceled

Fantix King added the comment: Following the discussion - looks like `add_reader()` will cancel the previous `handle` on the same `fd` if found, so I think it is reasonable to solve this cancellation-race issue by checking if the `handle` is cancelled as shown in the attached patch (tests

[issue40582] Inconsistent exceptions caused by typing + tuple subclasses

Dennis Sweeney added the comment: I think the behavior is consistent between tuple and an empty subclass: >>> from typing import List >>> class T(tuple): pass == Empty tuple/T == >>> List[()] Traceback (most recent call last): ... TypeError: Too few

[issue40334] PEP 617: new PEG-based parser

Guido van Rossum added the comment: New changeset ac7a92cc0a821699df48bc2e30a02c25d6338f78 by Pablo Galindo in branch 'master': bpo-40334: Avoid collisions between parser variables and grammar variables (GH-19987)

[issue32824] Docs: Using Python on a Macintosh has bad info per Apple site

Glenn Travis added the comment: So, how do we wake this sleeping concern up? -- nosy: +TotallyLost ___ Python tracker ___ ___

[issue40578] Deprecate numeric item access for platform.uname()

Steven D'Aprano added the comment: You have given no justification for removing item access for the fields except "some users are still relying on ... access by item position and length". Normally the fact that some people are doing this would be reason to reject this

[issue40571] Make lru_cache(maxsize=None) more discoverable

Change by Raymond Hettinger : -- keywords: +patch pull_requests: +19330 stage: -> patch review pull_request: https://github.com/python/cpython/pull/20019 ___ Python tracker

[issue40397] Refactor typing._GenericAlias

Guido van Rossum added the comment: I think I have a shorter repro, not involving unions: from typing import * T = TypeVar("T") S = TypeVar("S") U = TypeVar("U") class A(Generic[T]): ... class B(Generic[T]): ... class C(Generic[T, S]): ... print(C[A[U], B[int]][str]) Fails in the same place,

[issue38323] asyncio: MultiLoopWatcher has a race condition (test_asyncio: test_close_kill_running() hangs on AMD64 RHEL7 Refleaks 3.x)

Chris Jerdonek added the comment: > 4. use call_later() to terminate the future after 5 seconds You should read that as "3.5" (I inserted it later). -- ___ Python tracker

[issue40334] PEP 617: new PEG-based parser

Guido van Rossum added the comment: FWIW I propose that we add another Misc/NEWS for the same issue summarizing what happened between alpha6 and beta1. (A lot!) -- ___ Python tracker

[issue40581] scipy's differential_evolution bug

Christian Heimes added the comment: SciPy and Sage are 3rd party extensions to CPython and not maintained by us. Please report the issue with SciPi, https://www.scipy.org/scipylib/bug-report.html -- nosy: +christian.heimes resolution: -> third party stage: -> resolved status: open

[issue26817] Docs for StringIO should link to io.BytesIO

Change by Zachary Ware : -- resolution: -> out of date stage: -> resolved status: open -> closed ___ Python tracker ___ ___

[issue40569] Add optional weights to random.sample()

Raymond Hettinger added the comment: Negative weights are undefined for choices() as well. Just like bisect() and merge() don't verify their inputs are sorted. Usually, full scan precondition checks are expensive in pure python and aren't worth penalizing correct code. As Henk-Jaap

[issue40581] scipy's differential_evolution bug

New submission from Mike : I have a random bug with scipy's optimize . I use it in the context of SageMath (Python 3.7.3). I checked 3 algorithms : shgo, dual_annealing and full_optimize. All don't work well (at all !). I optimise with a 3 parameters functions with given bounds. The

[issue38323] asyncio: MultiLoopWatcher has a race condition (test_asyncio: test_close_kill_running() hangs on AMD64 RHEL7 Refleaks 3.x)

Kyle Stanley added the comment: > Would it make sense to deprecate it and stick with ThreadedChildWatcher? I had proposed deprecating it in another bpo issue, but it was brought up that MutliLoopWatcher had only been added recently in 3.8. So, it might be a bit too soon to deprecate it

[issue32824] Docs: Using Python on a Macintosh has bad info per Apple site

Rémi Lapeyre added the comment: Hi Gleen, the best way forward is to propose an improvement to the current documentation. It will get reviewed and merge if appropriate. The source for this part is at https://github.com/python/cpython/blob/master/Doc/using/mac.rst and you can find the

[issue40576] Align docs for list.sort, sorted, and on the website

Raymond Hettinger added the comment: To me, the PR isn't an improvement. It makes the docs much more verbose without really adding information. We already link to the sorting-howto which already discusses the pragmatics that arise in actual applications. Also, there is no special benefit

[issue40561] Provide docstrings for public-facing webbrowser functions

Brad Solomon added the comment: To no surprise, not a lot of activity with the module over the last few years as it is fairly cut-and-dry. $ git shortlog -sn --since '5 years ago' Lib/webbrowser.py 3 Serhiy Storchaka 1 David Steele 1 Guido van Rossum 1 Michael Haas

[issue40334] PEP 617: new PEG-based parser

Change by Pablo Galindo Salgado : -- pull_requests: +19331 pull_request: https://github.com/python/cpython/pull/20020 ___ Python tracker ___

[issue40580] Macintosh Documentation Still Bad

Glenn Travis added the comment: Pull - Request? -- ___ Python tracker ___ ___ Python-bugs-list mailing list Unsubscribe:

[issue32824] Docs: Using Python on a Macintosh has bad info per Apple site

Glenn Travis added the comment: Thank you Remi > On May 9, 2020, at 4:15 PM, Rémi Lapeyre wrote: > > > Rémi Lapeyre added the comment: > > Hi Gleen, the best way forward is to propose an improvement to the current > documentation. It will get reviewed and merge if appropriate. > > The

[issue40569] Add optional weights to random.sample()

David Heiberg added the comment: Just playing around with this and I think one thing to consider is how to handle negative weights. Should they even be allowed? One interesting behaviour I encountered with the current draft is the following: >>> r.sample(['katniss', 'prim', 'gale',

[issue40553] Python 3.8.2 Mac freezing/not responding when saving new programs

Terry J. Reedy added the comment: I asked about other people's experiences on python-list and got 2 responses so far. --- On macOS The default Save/Save as dialogs are short, only displaying a few major folders along with Favorites and Recents. That dialog doesn’t display folder contents.

[issue40582] Inconsistent exceptions caused by typing + tuple subclasses

New submission from Ruairidh MacLeod : When incorrectly defining a function with a typed List[T] argument where T is a tuple instance, a TypeError is correctly raised: t = (1,) def f(a: List[t]): ... # => TypeError: Parameters to generic types must be types. Got 1. When t is an instance of

[issue40561] Provide docstrings for public-facing webbrowser functions

Terry J. Reedy added the comment: I just got my development machine back, so I will handle this. I first determined that code and doc are same in all 3 versions (except for audit event bit, which I am willing to leave out). -- versions: +Python 3.7

[issue38323] asyncio: MultiLoopWatcher has a race condition (test_asyncio: test_close_kill_running() hangs on AMD64 RHEL7 Refleaks 3.x)

Nathaniel Smith added the comment: How do you know that your reproducer is showing the same bug, or anything related to signals? IIUC subprocess waiting by default doesn't involve a signal handler. On Sat, May 9, 2020, 14:40 Chris Jerdonek wrote: > > Chris Jerdonek added the comment: > > >

[issue40569] Add optional weights to random.sample()

Henk-Jaap Wagenaar added the comment: @rhettinger: I like this proposal. @dheiberg: to me, it seems negative weights are explicitly not supported, from: https://docs.python.org/3/library/random.html#random.choices "Weights are assumed to be non-negative." Unless I am missing something. In

[issue38323] asyncio: MultiLoopWatcher has a race condition (test_asyncio: test_close_kill_running() hangs on AMD64 RHEL7 Refleaks 3.x)

Chris Jerdonek added the comment: > this seems like an awful lot of energy to spend on some code that's not even used by default. True, but it seems likely to me that this signals :) a deeper, more general issue about CPython / signals (which is why I spent time on it). For example, my

[issue26817] Docs for StringIO should link to io.BytesIO

Zackery Spytz added the comment: Python 2 is EOL. -- nosy: +ZackerySpytz ___ Python tracker ___ ___ Python-bugs-list mailing list

[issue40561] Provide docstrings for public-facing webbrowser functions

Terry J. Reedy added the comment: Current open doc: https://docs.python.org/3.9/library/webbrowser.html#webbrowser.open -- ___ Python tracker ___

[issue32824] Docs: Using Python on a Macintosh has bad info per Apple site

Glenn Travis added the comment: Thank you, but how in the world does one know where to look or find that document at github. I tried to search earlier and got hundreds of listings for python doc. Sent from my iPhone > On May 9, 2020, at 16:36, Glenn Travis wrote: > > Thank you Remi >

[issue38323] asyncio: MultiLoopWatcher has a race condition (test_asyncio: test_close_kill_running() hangs on AMD64 RHEL7 Refleaks 3.x)

Chris Jerdonek added the comment: I came up with the script by (1) running the test locally and seeing the same hang, (2) moving the test function to its own script separate from the unit tests and seeing the same hang, and (3) successively stripping away code while continuing to check for

[issue40581] scipy's differential_evolution bug

Mike added the comment: ok thank you Christian. I will do. > Le 9 mai 2020 à 23:06, Christian Heimes a écrit : > > > Christian Heimes added the comment: > > SciPy and Sage are 3rd party extensions to CPython and not maintained by us. > Please report the issue with SciPi, >

[issue40578] Deprecate numeric item access for platform.uname()

New submission from Jason R. Coombs : In issue40570, I learned that some users are still relying on legacy tuple-like behaviors of `platform.uname()`, namely the access by item position and length: ``` platform.uname()[0] len(platform.uname()) ``` I propose to deprecate these behaviors and

[issue40570] len(platform.uname()) has changed in Python 3.9

Jason R. Coombs added the comment: Thanks Marc-Andre for the suggestion, but I don't think that approach is viable here. The whole point of issue35967 was to defer the execution of the `.processor` behavior until it was requested. The intention is not to extend the tuple, but to shorten it

[issue37573] asyncio: freeze when using MultiLoopChildWatcher on Solaris

Chris Jerdonek added the comment: This looks like a duplicate of #38323: https://bugs.python.org/issue38323 -- nosy: +chris.jerdonek ___ Python tracker ___

[issue40576] Align docs for list.sort, sorted, and on the website

Sam Lijin added the comment: I do clarify that they're not the same thing and that `sorted` accepts any iterable: https://github.com/python/cpython/pull/20017/files#diff-30d76a3dc0c885f86917b7d307ccf279 -- ___ Python tracker

[issue40570] len(platform.uname()) has changed in Python 3.9

Jason R. Coombs added the comment: New changeset 2c3d508c5fabe40dac848fb9ae558069f0576879 by Jason R. Coombs in branch 'master': bpo-40570: Improve compatibility of uname_result with late-bound .platform (#20015)

[issue40570] len(platform.uname()) has changed in Python 3.9

Jason R. Coombs added the comment: I've gone ahead and merged PR 20015 to fix the issue, but I'm happy to revisit if a better approach is proposed. -- keywords: +3.9regression -patch resolution: -> fixed stage: patch review -> resolved status: open -> closed

[issue40576] Align docs for list.sort, sorted, and on the website

Eric V. Smith added the comment: sorted() does not just delegate to list.sort(): it sorts any iterable. >>> sorted({0:1,2:3,-1:4}) [-1, 0, 2] I think your change conflates the two by pointing to list.sort() from sorted(). -- nosy: +eric.smith ___

[issue40578] Deprecate numeric item access for platform.uname()

Jason R. Coombs added the comment: In https://github.com/jaraco/cpython/tree/bc73729eb9, I started drafting a proposed implementation to address this concern, but ran into a snag - the tests immediately reveal that `tuple(platform.uname())` invokes `__len__`, triggering the deprecation

[issue38323] asyncio: MultiLoopWatcher has a race condition (test_asyncio: test_close_kill_running() hangs on AMD64 RHEL7 Refleaks 3.x)

Chris Jerdonek added the comment: I'm adding Nathaniel in case he can recognize this as one of the signal handler cases he's already come across. -- nosy: +njs ___ Python tracker

[issue40576] Align docs for list.sort, sorted, and on the website

Change by Sam Lijin : -- keywords: +patch pull_requests: +19327 stage: -> patch review pull_request: https://github.com/python/cpython/pull/20017 ___ Python tracker ___

[issue40579] Second argument to iterator.next not described

New submission from Andrew Black : The documentation for the built-in function next (which calls the __next__ method on an iterator) discusses its behavior when the iterator is exhausted. It talks about the StopIteration exception. However, it does not say anything about calling next with

[issue40576] Align docs for list.sort, sorted, and on the website

Eric V. Smith added the comment: I still think it's a mistake to point to list.sort() from sorted(). If anything I think it should be the other way around. We'll see what other people think. -- ___ Python tracker

[issue40570] len(platform.uname()) has changed in Python 3.9

Jason R. Coombs added the comment: > you added a late binding optimization for the whole uname return tuple to save the effort of ... shell access. It wasn't to save the effort and it wasn't an optimization. There was a fundamental race condition where it became impossible to implement a

[issue40397] Refactor typing._GenericAlias

Guido van Rossum added the comment: @Serhiy can you look at this? A possible fix might be: diff --git a/Lib/typing.py b/Lib/typing.py index 681ab6d21e..adcef1e82b 100644 --- a/Lib/typing.py +++ b/Lib/typing.py @@ -701,7 +701,8 @@ class _GenericAlias(_BaseGenericAlias, _root=True):

[issue40578] Deprecate numeric item access for platform.uname()

Jason R. Coombs added the comment: In issue40570, Marc-Andre writes: > I don't think that deprecating standard tuple access is an option for the uname() return value, since it's documented to be a tuple. My thinking here is that as part of the deprecation, the documentation would be updated

[issue40580] Macintosh Documentation Still Bad

New submission from Glenn Travis : This was reported two years ago, and still is not fixed https://bugs.python.org/issue32824#msg312028 and the documentation is even older. It starts off referencing "Mac OS X 10.8" and Apple Documents that are archived. Then the section referencing using

[issue40580] Macintosh Documentation Still Bad

Rémi Lapeyre added the comment: Hi Gleen, this looks like there already exists an issue for this problem, please use it instead of opening a new one. Can you please open a new Pull-Request for solving this issue? -- nosy: +remi.lapeyre ___ Python

[issue40580] Macintosh Documentation Still Bad

Change by Benjamin Peterson : -- resolution: -> duplicate stage: -> resolved status: open -> closed superseder: -> Docs: Using Python on a Macintosh has bad info per Apple site ___ Python tracker

[issue40580] Macintosh Documentation Still Bad

Glenn Travis added the comment: Thank you for your reply. I just wanted to renew or recall attention to the older one. It seems to have been dormant for too long. A C program is like a fast dance on a newly waxed dance floor by people carrying razors. -W.R. And now for something

[issue40570] len(platform.uname()) has changed in Python 3.9

Marc-Andre Lemburg added the comment: Ok, let me add some more context: When I wrote the uname interface I was aware that calling the API will take some resources. That's why I added the cache. IMO, that was enough as optimization. Now, you added a late binding optimization for the whole

[issue40536] Addition of a "list of available time zones" function to zoneinfo

Paul Ganssle added the comment: I have an initial implementation against the reference implementation here: https://github.com/pganssle/zoneinfo/pull/60 Once GH-19909 is merged, I will turn that into a PR against CPython. For the first pass I went with: 1. free-standing function 2.

[issue38323] asyncio: MultiLoopWatcher has a race condition (test_asyncio: test_close_kill_running() hangs on AMD64 RHEL7 Refleaks 3.x)

Nathaniel Smith added the comment: I don't have any particular insight into the bug, but I do have an observation: this seems like an awful lot of energy to spend on some code that's not even used by default. Would it make sense to deprecate it and stick with ThreadedChildWatcher? On Sat, May

[issue40561] Provide docstrings for public-facing webbrowser functions

Terry J. Reedy added the comment: Except is isn't, as he has been inactive for years. Git log may or may not show more recent commits. My dev machine is being repaired but should be back soon. -- ___ Python tracker

[issue40579] Second argument to iterator.next not described

Raymond Hettinger added the comment: The next() function has a different signature than the __next__() method. The former takes one or two arguments. That latter only takes one. >>> it = iter('abcde') >>> next(it) 'a' >>> next(it, 'default') 'b' >>> it.__next__()

[issue40561] Provide docstrings for public-facing webbrowser functions

Change by Brad Solomon : -- nosy: +georg.brandl ___ Python tracker ___ ___ Python-bugs-list mailing list Unsubscribe:

[issue40561] Provide docstrings for public-facing webbrowser functions

Brad Solomon added the comment: The module source notes "Maintained by Georg Brandl." -- ___ Python tracker ___ ___

[issue40575] _PyDict_GetItemIdWithError() should call _PyDict_GetItem_KnownHash()

Change by Stefan Behnel : -- keywords: +patch pull_requests: +19329 stage: needs patch -> patch review pull_request: https://github.com/python/cpython/pull/20018 ___ Python tracker

[issue40579] Second argument to iterator.next not described

Rémi Lapeyre added the comment: It's documented here: https://docs.python.org/3/library/functions.html#next -- nosy: +remi.lapeyre ___ Python tracker ___

[issue40575] _PyDict_GetItemIdWithError() should call _PyDict_GetItem_KnownHash()

Change by Stefan Behnel : -- components: +Interpreter Core keywords: +easy (C) ___ Python tracker ___ ___ Python-bugs-list mailing

[issue38323] asyncio: MultiLoopWatcher has a race condition (test_asyncio: test_close_kill_running() hangs on AMD64 RHEL7 Refleaks 3.x)

Chris Jerdonek added the comment: I'm attaching a slightly simpler version of the script. -- Added file: https://bugs.python.org/file49145/test-kill2.py ___ Python tracker

[issue40577] Spyder doesn't launch from Anaconda on MacBook

Christian Heimes added the comment: This is a bug in a 3rd party module, not in CPython. Please report the issue with Anaconda and Spyder. -- nosy: +christian.heimes resolution: -> third party stage: -> resolved status: open -> closed ___ Python

[issue40575] _PyDict_GetItemIdWithError() should call _PyDict_GetItem_KnownHash()

Change by Inada Naoki : -- nosy: +inada.naoki ___ Python tracker ___ ___ Python-bugs-list mailing list Unsubscribe:

[issue40570] len(platform.uname()) has changed in Python 3.9

Marc-Andre Lemburg added the comment: Hi Jason, to achieve better backwards compatibility, it's probably better to use the approach taken for CodeInfo in the codecs.py module: class CodecInfo(tuple): """Codec details when looking up the codec registry""" def __new__(cls, encode,

[issue40566] Apply PEP 573 to abc module

miss-islington added the comment: New changeset 77c614624b6bf2145bef69830d0f499d8b55ec0c by Dong-hee Na in branch 'master': bpo-40566: Apply PEP 573 to abc module (GH-20005) https://github.com/python/cpython/commit/77c614624b6bf2145bef69830d0f499d8b55ec0c -- nosy: +miss-islington

[issue40551] PRs should be rebased on top of master before running the build/tests

Filipe Laíns added the comment: > * Travis-CI, at least, does test against "merge commit", not HEAD of PR > branch. Where? https://github.com/python/cpython/blob/master/.travis.yml > * As you said already, master branch grow after PR is created anyway. > This issue proposes " this only

[issue40575] _PyDict_GetItemIdWithError() should call _PyDict_GetItem_KnownHash()

New submission from Stefan Behnel : _PyDict_GetItemIdWithError() looks up interned strings, which always have their hash value initialised. It can call _PyDict_GetItem_KnownHash() directly. -- messages: 368506 nosy: scoder priority: low severity: normal stage: needs patch status: open

[issue40577] Spyder doesn't launch from Anaconda on MacBook

New submission from Abhinav Vikram : I just installed Anaconda to start coding in Python. However, after installing, the moment I launch Spyder I get the error "Python quit unexpectedly" error. I have tried running "conda update --all" and even reinstalling anaconda but no help. Just for

[issue40566] Apply PEP 573 to abc module

Dong-hee Na added the comment: I am now closing the issue. Thanks for the review -- nosy: +christian.heimes resolution: -> fixed stage: patch review -> resolved status: open -> closed ___ Python tracker

[issue40551] PRs should be rebased on top of master before running the build/tests

Inada Naoki added the comment: On Sat, May 9, 2020 at 6:08 PM Filipe Laíns wrote: > > > * Travis-CI, at least, does test against "merge commit", not HEAD of PR > > branch. > > Where? https://github.com/python/cpython/blob/master/.travis.yml https://docs.travis-ci.com/user/pull-requests/

[issue40551] PRs should be rebased on top of master before running the build/tests

Filipe Laíns added the comment: > "Rather than build the commits that have been pushed to the branch the pull request is from, we build the merge between the source branch and the upstream branch." Okay, great. But this still only happens for Linux. And still has the same problem, the build

[issue38323] asyncio: MultiLoopWatcher has a race condition (test_asyncio: test_close_kill_running() hangs on AMD64 RHEL7 Refleaks 3.x)

Chris Jerdonek added the comment: I'm attaching a stand-alone script that can reproduce the issue. It doesn't use unittest or even MultiLoopChildWatcher. It starts an event loop and then repeatedly calls loop.subprocess_exec() with 0.2 seconds in between until the "hang" happens (which

[issue40426] Unable to use lowercase hexadecimal digits for percent encoding

Serhiy Storchaka added the comment: Concur with Christian. Heh, I was going to propose the same. -- resolution: -> rejected stage: patch review -> resolved status: open -> closed ___ Python tracker

[issue40576] Align docs for list.sort, sorted, and on the website

New submission from Sam Lijin : It's mildly confusing to see `pydoc sorted` and `pydoc list.sort` to look so different, esp. since sorted just delegates to list.sort. By extension, https://docs.python.org/3/library/functions.html#sorted should also be aligned with this, but I'm not super

[issue40551] PRs should be rebased on top of master before running the build/tests

Filipe Laíns added the comment: > re-trigger the CI, to make everything works correctly with master. *re-trigger the CI, to make *sure* everything works correctly with master. -- ___ Python tracker