[issue31507] email.utils.parseaddr has no docstring

Change by Mariatta Wijaya : -- resolution: -> fixed stage: patch review -> resolved status: open -> closed ___ Python tracker ___ __

[issue31507] email.utils.parseaddr has no docstring

Mariatta Wijaya added the comment: New changeset 93c0885dc84381cbbb970402b1a21bf690ee312c by Mariatta (Rohit Balasubramanian) in branch '3.6': bpo-31507 Add docstring to parseaddr function in email.utils.parseaddr (GH-3647) (GH-3733) https://github.com/python/cpython/commit/93c0885dc84381cbbb

[issue31340] Use VS 2017 compiler for build

Ned Deily added the comment: If you think we should do a 3.6.4 to address this, we can do that. I'll leave it up to your judgement. -- ___ Python tracker ___

[issue31340] Use VS 2017 compiler for build

Steve Dower added the comment: > How about giving priority to VS 2015 That would be the fix, but it's not worth releasing an immediate 3.6.4 IMHO and by the time 3.6.4 comes about it could be more pain to change back than to leave it. That's the balance I'm thinking about for a while. Includ

[issue31712] subprocess with stderr=subprocess.STDOUT hang

Gregory P. Smith added the comment: Please try to reproduce this with Python 3.6. On 2.7, I highly recommend you stop using the Python 2.7 subprocess module and install the subprocess32 backport available on PyPI. It is *much* more reliable. -- ___

[issue31712] subprocess with stderr=subprocess.STDOUT hang

Martin Panter added the comment: Presumuing your file descriptor 3 is the read end of the pipe to the child’s output, then there is probably a process somewhere that could still write to the write end. Normally “check_output” waits until it has read all possible output from the pipe(s). This

[issue31340] Use VS 2017 compiler for build

Christoph Gohlke added the comment: I build most of my binaries after calling the correct vcvarsall.bat so I did not notice until today, when I was trying to build outside that environment. Also, I misread your comment (https://bugs.python.org/issue31340#msg301538) earlier and did not worry

[issue31340] Use VS 2017 compiler for build

Steve Dower added the comment: Build artifacts (static libraries, in this case) are not compatible, but the built binaries are. This may be a reasonable argument for not updating distutils's support in minor versions (and I'm totally happy to just stop updating distutils period - see the rec

[issue31340] Use VS 2017 compiler for build

Christoph Gohlke added the comment: I have Visual Studio 2015 and Visual Studio 2017 Community editions with latest patches installed. Building a minimal C extension on Python 3.6.3 with distutils now uses MSVC 14.11.25503. That is unexpected. When building complex extensions that link to st

[issue31165] list_slice() does crash if the list is mutated indirectly by PyList_New()

Serhiy Storchaka added the comment: Or the bug is so hard for reproducing, that nobody had enough information for reporting. -- ___ Python tracker ___

[issue31718] some methods of uninitialized io.IncrementalNewlineDecoder objects raise SystemError

Serhiy Storchaka added the comment: There is other issues with IncrementalNewlineDecoder.__init__ -- it leaks references when called repeatedly. The simplest solution of both issues will be moving the initialization to the new method. But this class looks designed for subclassing, and this ca

[issue31692] [2.7] Test `test_huntrleaks()` of test_regrtest fails in debug build with COUNT_ALLOCS

STINNER Victor added the comment: Serhiy: > Do you going to backport also -X showrefcount? I see -X as a Python3 only thing. If I have to choose, I would prefer to add a new environment variable, since it's more in the "Python2 style". Iryna: > That is how it is bypassed in Fedora Py2 builds

[issue31692] [2.7] Test `test_huntrleaks()` of test_regrtest fails in debug build with COUNT_ALLOCS

Serhiy Storchaka added the comment: I left this on to Benjamin, the RM of Python 2.7. -- ___ Python tracker ___ ___ Python-bugs-list

[issue31165] list_slice() does crash if the list is mutated indirectly by PyList_New()

STINNER Victor added the comment: > This is different case than mutating a dict while iterate it. In this case > the failure is caused by GC, and it is always hard to handle such issues. In > case of a dict you can just copy it before iterating. But what to do with > RuntimeError from slicing

[issue31165] list_slice() does crash if the list is mutated indirectly by PyList_New()

Serhiy Storchaka added the comment: This is different case than mutating a dict while iterate it. In this case the failure is caused by GC, and it is always hard to handle such issues. In case of a dict you can just copy it before iterating. But what to do with RuntimeError from slicing a lis

[issue31719] [2.7] test_regrtest.test_crashed() fails on s390x

Change by STINNER Victor : -- nosy: +cstratak, ishcherb ___ Python tracker ___ ___ Python-bugs-list mailing list Unsubscribe: https:

[issue31692] [2.7] Test `test_huntrleaks()` of test_regrtest fails in debug build with COUNT_ALLOCS

Iryna Shcherbina added the comment: > I don't know if we can easily implement "-X showalloccount", since > Python 2.7 doesn't have sys._xoptions. Maybe we could use a new > environment variable instead: PYTHONSHOWALLOCCOUNT=1? That is how it is bypassed in Fedora Py2 builds currently. I am att

[issue31719] [2.7] test_regrtest.test_crashed() fails on s390x

Change by STINNER Victor : -- keywords: +patch pull_requests: +3885 stage: -> patch review ___ Python tracker ___ ___ Python-bugs-li

[issue31719] [2.7] test_regrtest.test_crashed() fails on s390x

New submission from STINNER Victor : On the s390x architecture, the test_regrtest.test_crashed() fails because Python doesn't crash. test.support._crash_python() calls ctypes.string_at(0) to crash Python. In debug mode, ctypes.string_at(0) fails with an assertion error and the process is kill

[issue31718] some methods of uninitialized io.IncrementalNewlineDecoder objects raise SystemError

Oren Milman added the comment: Yes, although i don't know if there are usecases for that. -- ___ Python tracker ___ ___ Python-bugs-

[issue31692] [2.7] Test `test_huntrleaks()` of test_regrtest fails in debug build with COUNT_ALLOCS

STINNER Victor added the comment: My colleague Iryna Shcherbina came to me with this issue. I'm not sure that we really need to support COUNT_ALLOCS. But strangely, it seems simpler to fix bugs rather than guessing if users like this debug mode or not :-) There are different ways to fix tests

[issue31692] [2.7] Test `test_huntrleaks()` of test_regrtest fails in debug build with COUNT_ALLOCS

Serhiy Storchaka added the comment: Do you going to backport also -X showrefcount? -- ___ Python tracker ___ ___ Python-bugs-list ma

[issue31692] [2.7] Test `test_huntrleaks()` of test_regrtest fails in debug build with COUNT_ALLOCS

Change by Serhiy Storchaka : -- nosy: +benjamin.peterson ___ Python tracker ___ ___ Python-bugs-list mailing list Unsubscribe: https

[issue31165] null pointer deref and segfault in list_slice (listobject.c:455)

STINNER Victor added the comment: "Maybe we should prevent collection of garbage with circular references (that has __del__() or weakref callbacks) from PyObject_GC_New()?" That would be a major change in the garbage collector. I would prefer to not touch the GC, any change can introduce a co

[issue31165] list_slice() does crash if the list is mutated indirectly by PyList_New()

Change by STINNER Victor : -- title: null pointer deref and segfault in list_slice (listobject.c:455) -> list_slice() does crash if the list is mutated indirectly by PyList_New() ___ Python tracker __

[issue31165] null pointer deref and segfault in list_slice (listobject.c:455)

STINNER Victor added the comment: > Oh, you are right Oren. Seems this is the only solution. There are other solutions. I wrote PR 3911 which checks if the list size changed after PyList_New(). If it's the case, a RuntimeError exception is raised. We implemented similar checks in the dict ty

[issue31165] null pointer deref and segfault in list_slice (listobject.c:455)

Change by STINNER Victor : -- keywords: +patch pull_requests: +3884 stage: -> patch review ___ Python tracker ___ ___ Python-bugs-li

[issue31692] [2.7] Test `test_huntrleaks()` of test_regrtest fails in debug build with COUNT_ALLOCS

STINNER Victor added the comment: I wrote PR 3910 to modify COUNT_ALLOCS: alllcations statistics are now written into stderr, rather than stdout. The PR contains also changes to fix tests with COUNT_ALLOCS. I'm not sure about this part of the PR. I would prefer to not dump statistics by defau

[issue19527] Test failures with COUNT_ALLOCS

Change by STINNER Victor : -- pull_requests: +3883 ___ Python tracker ___ ___ Python-bugs-list mailing list Unsubscribe: https://mai

[issue31692] [2.7] Test `test_huntrleaks()` of test_regrtest fails in debug build with COUNT_ALLOCS

Change by STINNER Victor : -- keywords: +patch pull_requests: +3882 stage: -> patch review ___ Python tracker ___ ___ Python-bugs-li

[issue31718] some methods of uninitialized io.IncrementalNewlineDecoder objects raise SystemError

Serhiy Storchaka added the comment: Is IncrementalNewlineDecoder subclassable? -- nosy: +serhiy.storchaka ___ Python tracker ___ ___

[issue31680] Expose curses library name and version on Python level

Serhiy Storchaka added the comment: No need to backport this to 3.6. The test in 3.6 can be skipped on OpenBSD, independently from the ncurses version. -- ___ Python tracker _

[issue31655] SimpleNamespace accepts non-string keyword names

Change by Serhiy Storchaka : -- type: -> behavior versions: +Python 3.6 ___ Python tracker ___ ___ Python-bugs-list mailing list Uns

[issue31655] SimpleNamespace accepts non-string keyword names

Change by Serhiy Storchaka : -- keywords: +patch pull_requests: +3881 stage: -> patch review ___ Python tracker ___ ___ Python-bugs-

[issue31680] Expose curses library name and version on Python level

Terry J. Reedy added the comment: pythoninfo seems like the right place. It is publicly readable. We just reserved the right to change it in any release as needed. (This actually is useful to other users also as long as they know.) #15037 was originally filed for 3.4 and is currently marke

[issue31672] string.Template should use re.ASCII flag

Serhiy Storchaka added the comment: Another solution (works in 3.6 too): set idpattern to r'(?-i:[_a-zA-Z][_a-zA-Z0-9]*)'. This looks pretty weird, setting the re.IGNORECASE flag and then unsetting it. But it works, and don't break the user code that changes idpattern without changing flags.

[issue31655] SimpleNamespace accepts non-string keyword names

Terry J. Reedy added the comment: After reading the doc entry for SimpleNamespace, I see running 'SimpleNamespace(**{0:0})' as a bug because doing so results in an object than contradicts the doc. 1. "A simple object subclass that provides attribute access to its namespace, as well as a mean

[issue31718] some methods of uninitialized io.IncrementalNewlineDecoder objects raise SystemError

New submission from Oren Milman : Given an uninitialized IncrementalNewlineDecoder: uninitialized = io.IncrementalNewlineDecoder.__new__(io.IncrementalNewlineDecoder) each of the following calls would raise a SystemError ('null argument to internal routine'): uninitialized.getstate() uninitiali

[issue31558] gc.freeze() - an API to mark objects as uncollectable

Serhiy Storchaka added the comment: What about msg302790? -- nosy: +serhiy.storchaka ___ Python tracker ___ ___ Python-bugs-list mai

[issue31591] Closing socket raises AttributeError: 'collections.deque' object has no attribute '_decref_socketios'

reidfaiv added the comment: I will withdraw this bug report. I am unable to isolate that issue, hence I can not confirm if this is purely Python crash or caused by some extension. It looks memory corruption to me as segfault moves around and produces different stack traces - the network code

[issue31713] python3 python-config script generates invalid includes

Change by matthewlweber : -- keywords: +patch Added file: https://bugs.python.org/file47195/0029-python-config.sh-don-t-reassign-prefix.patch ___ Python tracker ___ ___

[issue31700] one-argument version for Generator.typing

Ivan Levkivskyi added the comment: You can use Iterator type, for example this works in mypy (didn't test in other type checkers): def f() -> Iterator[int]: yield 42 In case you want annotate something specifically as Generator[int, None, None] (for example to use its .close() method

[issue31717] Socket documentation threading misstep?

Phillip added the comment: I could definitely understand that. After all, if it's slightly askew (or strikes some as such) it forces critical thinking, which is good. I didn't think calling run() was indicative of the three likely pathways to handle the client socket in the following paragraph.

[issue31666] Pandas_datareader Error Message - ModuleNotFoundError: No module named 'pandas_datareader' with module in folder

New submission from Éric Araujo : Hello! Your bug report gives very little information for us to help you. Can you give details such as: your environement / setup, your code, expected result and full error message? https://devguide.python.org/tracker/#reporting-an-issue -- nosy: +me

[issue31558] gc.freeze() - an API to mark objects as uncollectable

Łukasz Langa added the comment: Alright Python people, I don't see anybody being against the idea on the thread. Can we get a review of the linked PR? I don't think it would be good form for me to accept it. -- ___ Python tracker

[issue31717] Socket documentation threading misstep?

R. David Murray added the comment: Or maybe instead of client_handler/run, it should be something like handle_client_asynchronously(clientsocket). -- ___ Python tracker ___ __

[issue31717] Socket documentation threading misstep?

R. David Murray added the comment: Based on the paragraph following the example, I don't think client_thread is a threading.Thread, and 'run' is meant to be a generic representation of a possible API. Since Threads do have a 'run' method, this is certainly potentially confusing. Maybe we sh

[issue31714] Improve re documentation

Brian Ward added the comment: In re.rst, the instances of "Correcsponds the" should be "Corresponds to the" and "Doesn't have correcsponding inline flag" should be "No corresponding inline flag." -- nosy: +Brian Ward ___ Python tracker

[issue31716] os.path.isdir returns true for dots

Eryk Sun added the comment: This is standard Windows API behavior for the final path component. A single dot component means the current directory. Two dots means the parent directory. More than two dots and/or trailing spaces, gets reduced to a single dot, meaning the current directory. For

[issue31717] Socket documentation threading misstep?

New submission from Phillip : Very small, but, https://docs.python.org/2/howto/sockets.html https://docs.python.org/3/howto/sockets.html have : while True: # accept connections from outside (clientsocket, address) = serversocket.accept() # now do something with the clientsocket

[issue31165] null pointer deref and segfault in list_slice (listobject.c:455)

Serhiy Storchaka added the comment: Oh, you are right Oren. Seems this is the only solution. -- nosy: +haypo, lemburg, pitrou, serhiy.storchaka, tim.peters, twouters ___ Python tracker ___

[issue31716] os.path.isdir returns true for dots

Change by Mor Haviv : -- resolution: -> not a bug ___ Python tracker ___ ___ Python-bugs-list mailing list Unsubscribe: https://mai

[issue31716] os.path.isdir returns true for dots

Change by Mor Haviv : -- stage: -> resolved status: open -> closed ___ Python tracker ___ ___ Python-bugs-list mailing list Unsubscr

[issue31714] Improve re documentation

Serhiy Storchaka added the comment: It is explained in the documentation: """ (In the rest of this section, we'll write RE's in ``this special style``, usually without quotes, and strings to be matched ``'in single quotes'``.) """ -- ___ Python trac

[issue31692] [2.7] Test `test_huntrleaks()` of test_regrtest fails in debug build with COUNT_ALLOCS

Serhiy Storchaka added the comment: Commit a793037d803abf098d172f686e2b95d27863c54d is not related to this issue. It skips the tests that were failed due to other side effect of COUNT_ALLOCS -- it makes type immortal. This is not a problem in 2.7 since all types are immortal in 2.7. But ther

[issue31715] Add mimetype for extension .mjs

Change by Bradley Meck : -- keywords: +patch pull_requests: +3879 stage: -> patch review ___ Python tracker ___ ___ Python-bugs-list

[issue31716] os.path.isdir returns true for dots

New submission from Mor Haviv : I uploaded this as a question on Stack Overflow and I suspect it might be a bug. Here is the link for the Stack Overflow question: https://stackoverflow.com/questions/46608731/python-os-path-isdir-returns-true-for-dots/46608842#46608842 The problem itself (copie

[issue31714] Improve re documentation

Ezio Melotti added the comment: ISTM that ``x`` is used when x is a regex or regex metachar, whereas ``'x'`` is used when 'x' is a string. I find this distinction reasonable. -- ___ Python tracker _

[issue31714] Improve re documentation

Henk-Jaap Wagenaar added the comment: I was looking at your changes and got myself in a muddle. What is you rational for when you use ``[character or string]`` versus ``'[character or string]``? You seem to be creating consistency there, but I cannot quite see the rules you are aiming for! -

[issue31714] Improve re documentation

Change by Serhiy Storchaka : -- nosy: +gregory.p.smith ___ Python tracker ___ ___ Python-bugs-list mailing list Unsubscribe: https:/

[issue24896] It is undocumented that re.UNICODE and re.LOCALE affect re.IGNORECASE

Serhiy Storchaka added the comment: I tried to correct the documentation in issue31714. -- ___ Python tracker ___ ___ Python-bugs-li

[issue31715] Add mimetype for extension .mjs

New submission from Bradley Meck : I propose to add a mapping of file extension .mjs to mime type "text/javascript". The "text/javascript" MIME is registered in https://www.iana.org/assignments/media-types, was moved to *should* as the MIME in HTML ( https://github.com/whatwg/html/pull/3096),

[issue31714] Improve re documentation

Change by Serhiy Storchaka : -- keywords: +patch pull_requests: +3878 ___ Python tracker ___ ___ Python-bugs-list mailing list Unsubs

[issue31714] Improve re documentation

New submission from Serhiy Storchaka : The proposed PR improves the documentation of re module and Regular Expression HOWTO. * Clarify the effect of the LOCALE flag. * Remove outdated statements. * Add an example of escaping a replacement string. * Add or correct some references. * Improve

[issue31692] [2.7] Test `test_huntrleaks()` of test_regrtest fails in debug build with COUNT_ALLOCS

STINNER Victor added the comment: test_regrtest is not the single test failing when COUNT_ALLOCS is defined. Not least than 12 tests are failing. In Python 3, many tests were fixed by skipping them if COUNT_ALLOCS is defined: commit a793037d803abf098d172f686e2b95d27863c54d of bpo-19527. Sinc

[issue31693] Document Py_GETENV

Change by Barry A. Warsaw : -- resolution: -> fixed stage: patch review -> resolved status: open -> closed ___ Python tracker ___ __

[issue7167] Smarter FTP passive mode

Albert-Jan Nijburg added the comment: I understand the standpoint that the server is configured incorrectly, and I see why this might not be the best solution to the problem. But not everyone owns the ftp server they're connecting to. And the `EPSV` command often doesn't include the ipaddres

[issue31713] python3 python-config script generates invalid includes

New submission from matthewlweber : Related to https://bugs.python.org/issue22907 If building in a path that starts with /usr, the includedir=$(echo "@includedir@" | sed "s#^$prefix_build#$prefix_real#") assignment in the python-config.sh ends up having the path it processes ran through a path

[issue31712] subprocess with stderr=subprocess.STDOUT hang

l4mer added the comment: BTW, when you will check point 3 you will notice that we already read id output, so command execute correctly: "uid=0(root" ... And problem is after that. Steps I reproduce this issue: - standard user have keys and in ~/.ssh/config Host * ControlMaster auto Cont

[issue31712] subprocess with stderr=subprocess.STDOUT hang

l4mer added the comment: yes, works perfectly when omit stderr=subprocess.STDOUT no matter if mux exists. Also from bash: ssh user@locahost id always works. Problem exists only when stderr=subprocess.STDOUT and no MUX yet. BTW, first connection create a MUX - but this seems to be an unimportant

[issue31712] subprocess with stderr=subprocess.STDOUT hang

R. David Murray added the comment: Given the title, are you saying that it works if you omit the stderr=subprocess.STDOUT? At the beginning you say you are using an authorized key and mux so you don't need to enter a password, and at the end you say it works fine if the MUX is created. So t

[issue31712] subprocess with stderr=subprocess.STDOUT hang

New submission from l4mer : I am using ssh mux + authorized key. So can exectute commands without passwod. 1. disconnect mux by: ssh user@localhost -O exit 2. run simple script import subprocess if __name__ == '__main__': cmd = ["ssh", "user@localhost", "id"] try: buf = subproce

[issue25658] PyThread assumes pthread_key_t is an integer, which is against POSIX

Nick Coghlan added the comment: This has been merged now: https://github.com/python/cpython/commit/731e18901484c75b60167a06a0ba0719a6d4827d Thank you for the PEP & implementation! :) -- resolution: -> fixed stage: patch review -> resolved status: open -> closed _

[issue31165] null pointer deref and segfault in list_slice (listobject.c:455)

Oren Milman added the comment: Oh, and calls to PyObject_GC_NewVar() might also cause similar issues. -- ___ Python tracker ___ ___

[issue31092] multiprocessing.Manager() race condition

Oren Milman added the comment: Davin and Antoine, i added you to the nosy list because you are listed as multiprocessing experts :) -- nosy: +davin, pitrou ___ Python tracker _

[issue31165] null pointer deref and segfault in list_slice (listobject.c:455)

Oren Milman added the comment: Here is some similar code that crashes for the same reasons: # create a circular reference with a malicious __del__(). class A: def __del__(*args): del list1[0] circ_ref_obj = A() circ_ref_obj._self = circ_ref_obj list1 = [None] list2 = [] del circ_re

[issue31711] ssl.SSLSocket.send(b"") fails

New submission from Jörn Heissler : Traceback (most recent call last): File "client.py", line 10, in conn.send(b'') File "/usr/lib/python3.6/ssl.py", line 941, in send return self._sslobj.write(data) File "/usr/lib/python3.6/ssl.py", line 642, in write return self._sslobj.write

[issue31596] expose pthread_getcpuclockid in time module

STINNER Victor added the comment: Thank you Benjamin, buildots are back to green :-) I prefer the new unit test which only tests pthread_getcpuclockid() clock and not make any assumption on the link between this clock and CLOCK_THREAD_CPUTIME_ID. -- _

[issue31707] Irrational fractions

Mark Dickinson added the comment: I think this is way beyond the scope of the current fractions module. I'd suggest putting something on PyPI as a proof of concept. Given the negative response from other core developers, I'm going to close here. -- resolution: -> rejected stage: ->

[issue31707] Irrational fractions

Henk-Jaap Wagenaar added the comment: I would like to provide some colour to this discussion. In a former life I have coded these during my studies. Ben is talking about implementing the Field of Fractions of an Integral Domain. See https://en.wikipedia.org/wiki/Field_of_fractions The way Fr

[issue31706] urlencode should accept generator as values for mappings when doseq=True

Serhiy Storchaka added the comment: Extending urlencode() to accept iterables instead of just sequences of values makes sense to me. There is no principal reason of requiring sequences. But you need to update the documentation, docstrings and comments. While we are here, we can generalize url