[issue29183] Unintuitive error handling in wsgiref when a crash happens in write() or close()

Martin Panter added the comment: There are actually two “close” methods in the WSGI package: ServerHandler’s implementation extends the BaseHandler implementation. Making the “close” methods do nothing if called a second time would avoid the error about “self.status” being None, but won’t

[issue35243] readline timeout too long for async gfx use

New submission from pmpp : unlike https://developer.mozilla.org/en-US/docs/Web/API/window/requestAnimationFrame which is fires about each 1/60 second : the callback for gui via PyOS_InputHook is as long as 0.1 second.

[issue28709] PyStructSequence_NewType is broken; makes GC type without setting Py_TPFLAGS_HEAPTYPE

Petr Viktorin added the comment: Should be fixed in PR9665 (bpo-34784), thanks to Eddie Elizondo. -- nosy: +petr.viktorin ___ Python tracker ___

[issue31354] Fixing a bug related to LTO only build

Charalampos Stratakis added the comment: I'd say yes. In general python may have been compiled with -flto, but it's still a bit buggy from the compilers' side. It doesn't work well always or at all depending on the toolchain, and even if python was compiled with -flto, that doesn't mean

[issue35241] [3.7] test_embed fails on MacOS buildbots

Change by STINNER Victor : -- title: test_embed fails on MacOS buildbots -> [3.7] test_embed fails on MacOS buildbots ___ Python tracker ___

[issue35241] [3.7] test_embed fails on MacOS buildbots

STINNER Victor added the comment: macOS buildbots are back to green. Thanks Pablo for creating the issue! I forgot the subtle differences on Py_HasFileSystemDefaultEncoding on Python 3.7... Python 3.8 has a more "portable" behavior now. -- resolution: -> fixed stage: needs patch

[issue35244] Allow to setup Clang as default compiler for modules build

Stéphane Wirtel added the comment: I confirm the issue, the main issue we have to support the CLang compiler into distutils. -- nosy: +matrixise ___ Python tracker ___

[issue35244] Allow to setup Clang as default compiler for modules build

Change by Stéphane Wirtel : -- versions: +Python 3.8 ___ Python tracker ___ ___ Python-bugs-list mailing list Unsubscribe:

[issue35233] _PyMainInterpreterConfig_Copy() doesn't copy install_signal_handlers

STINNER Victor added the comment: New changeset bc09ee8bc9c0ce4873cdaab6ca524a3ee3a36be1 by Victor Stinner in branch '3.7': bpo-35233: Fix test_embed.InitConfigTests on macOS (GH-10539) https://github.com/python/cpython/commit/bc09ee8bc9c0ce4873cdaab6ca524a3ee3a36be1 --

[issue35143] `from __future__ import annotations` has no effect inside `ast.parse`

Łukasz Langa added the comment: I think this is a bug, actually. It's going to be an incompatibility in `ast.parse` so we'll probably want to only enable it in Python 3.8. This should have stringified annotations: >>> p = ast.parse(""" ... from __future__ import annotations ... a: 1 ... """)

[issue33196] SEGV in mp.synchronize.Lock.__repr__ in spawn'ed proc if ctx mismatched

Arcadiy Ivanov added the comment: @gus.goulart you have reproduced it. The screenshot showing `-11` means the process dumped core. Because it's the child that dumps core, it's masked by abrt. Observe: $ python3 --version Python 3.7.1 $ python3 ~/Downloads/test_lock_sigsegv.py Parent r_q:

[issue35242] multiprocessing.Queue in an inconsistent state and a traceback silently suppressed if put an unpickable object and process's target function is finished

Change by Sergei Zobov : -- components: +Library (Lib) ___ Python tracker ___ ___ Python-bugs-list mailing list Unsubscribe:

[issue35230] Remove _Py_REF_DEBUG_COMMA

INADA Naoki added the comment: New changeset a757649fd6535c5e65481eb1077e30687421b92b by INADA Naoki in branch 'master': bpo-35230: dict: Remove some macros (GH-10513) https://github.com/python/cpython/commit/a757649fd6535c5e65481eb1077e30687421b92b --

[issue35230] Remove _Py_REF_DEBUG_COMMA

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

[issue35230] Remove _Py_REF_DEBUG_COMMA

STINNER Victor added the comment: Thanks. I almost started my project to convert macros to static inline functions just to remove ugly things like: #define _Py_REF_DEBUG_COMMA , (Well, there are other reasons to move to proper functions.) --

[issue33196] SEGV in mp.synchronize.Lock.__repr__ in spawn'ed proc if ctx mismatched

STINNER Victor added the comment: > At a minimum I think there should be a check trying to reduce arguments via > incompatible context's process to prevent a SEGV. I'm not sure that I understand the bug. The reproducer script pass a multiprocessing.Queue to a child process and then the

[issue33196] SEGV in mp.synchronize.Lock.__repr__ in spawn'ed proc if ctx mismatched

Change by Arcadiy Ivanov : Added file: https://bugs.python.org/file47933/coredump ___ Python tracker ___ ___ Python-bugs-list mailing list

[issue35241] [3.7] test_embed fails on MacOS buildbots

STINNER Victor added the comment: I merged https://github.com/python/cpython/pull/10539 -- ___ Python tracker ___ ___

[issue35244] Allow to setup Clang as default compiler for modules build

New submission from Jaime Torres : Clang is not in the list of allowed compilers when executed: python setup.py build --help-compiler --compiler=bcpp Borland C++ Compiler --compiler=cygwin Cygwin port of GNU C Compiler for Win32 --compiler=mingw32 Mingw32 port of GNU C Compiler

[issue33196] SEGV in mp.synchronize.Lock.__repr__ in spawn'ed proc if ctx mismatched

Arcadiy Ivanov added the comment: @taleinat The above has been reproduced on Fedora 29. -- ___ Python tracker ___ ___

[issue28709] PyStructSequence_NewType is broken; makes GC type without setting Py_TPFLAGS_HEAPTYPE

Change by Petr Viktorin : -- resolution: -> fixed stage: -> resolved status: open -> closed ___ Python tracker ___ ___

[issue35233] _PyMainInterpreterConfig_Copy() doesn't copy install_signal_handlers

Change by STINNER Victor : -- pull_requests: +9790 ___ Python tracker ___ ___ Python-bugs-list mailing list Unsubscribe:

[issue35253] Linker warning LNK4281

New submission from Jorge Ramos : When building python 3.6.7 x64 from source I get multiple warnings: LINK : LNK4281 - "undesirable base address 0x1D11 for x64 image; set base address above 4GB for best ASLR

[issue35251] FTPHandler.ftp_open documentation error

New submission from Lysandros Nikolaou : On https://docs.python.org/3/library/urllib.request.html?highlight=request#ftphandler-objects there is the sentence "The login is always done with empty username and password.", but in the (not so rare) case I am not missing something here, the code

[issue35252] test_functools dead code after FIXME

New submission from Lysandros Nikolaou : In test_functools.TestSingleDispatch.test_invalid_registrations (https://github.com/python/cpython/blob/4c596d54aa6a55e9d2a3db78891e656ebbfb63c8/Lib/test/test_functools.py#L2299) there is a FIXME with an immediate return afterwards that says that the

[issue35252] test_functools dead code after FIXME

Change by Lysandros Nikolaou : -- keywords: +patch pull_requests: +9797 stage: -> patch review ___ Python tracker ___ ___

[issue31354] Fixing a bug related to LTO only build

Change by Gregory P. Smith : -- versions: +Python 3.6, Python 3.8 ___ Python tracker ___ ___ Python-bugs-list mailing list

[issue31354] Fixing a bug related to LTO only build

Change by Gregory P. Smith : -- priority: normal -> release blocker ___ Python tracker ___ ___ Python-bugs-list mailing list

[issue35251] FTPHandler.ftp_open documentation error

Change by R. David Murray : -- components: +Library (Lib) -email nosy: -barry, r.david.murray ___ Python tracker ___ ___

[issue34784] Heap-allocated StructSequences

Pablo Galindo Salgado added the comment: New changeset bfb855bef6b428d639693caaf4e4c84cbb8a2f51 by Pablo Galindo in branch 'master': bpo-34784: Implement correct cleanup in PyStructSequence new implementation (GH-10536)

[issue35252] test_functools dead code after FIXME

Lysandros Nikolaou added the comment: It actually seems like the test after the FIXME is wrong and should be changed or deleted altogether. Can someone give a pointer or two on what would be the best choice here? -- ___ Python tracker

[issue35239] _PySys_EndInit() doesn't copy main interpreter configuration

STINNER Victor added the comment: Python has 3 kind of configurations: * global configuration variables like Py_VerboseFlag * core configuration: _PyCoreConfig * main interpreter configuration: _PyMainInterpreterConfig I tried to keep them consistency. Yesterday, I rewrote

[issue35246] asyncio.create_subprocess_exec doesn't accept pathlib.Path like subprocess does

New submission from lilydjwg : When I pass pathlib.Path to asyncio.create_subprocess_exec I get: TypeError: program arguments must be a bytes or text string, not PosixPath It's not so good when subprocess accepts this kind of arguments without issues. So can you add support for this?

[issue33196] SEGV in mp.synchronize.Lock.__repr__ in spawn'ed proc if ctx mismatched

Arcadiy Ivanov added the comment: @vstinner > I'm not sure that I understand the bug. The bug is, if a user makes an error and passes a Queue from context 'fork' to a child that is spawned using 'spawn', the passed Queue is, for obvious reasons, broken. The 'print("Child r_q: %r, %r, %r"

[issue35233] _PyMainInterpreterConfig_Copy() doesn't copy install_signal_handlers

Change by STINNER Victor : -- pull_requests: +9791 ___ Python tracker ___ ___ Python-bugs-list mailing list Unsubscribe:

[issue33196] SEGV in mp.synchronize.Lock.__repr__ in spawn'ed proc if ctx mismatched

Gus Goulart added the comment: @vstinner, on Debian 9 I can see the problem as well but wasn't able to debug with the level of details you did. Could you please share the process you followed? What I found was: ./python -X dev test_lock_sigsegv.py Parent r_q: , , Fatal Python error:

[issue35245] list comprehension for flattened or nested list differ too much

Steven D'Aprano added the comment: > l2 = [str(leaf) for leaf in tree for tree in forest] Expanded to nested loops, that becomes: l2 = [] for leaf in tree: for tree in forest: l2.append(str(leaf)) which of course gives a NameError, because you are trying to iterate over a tree

[issue33725] Python crashes on macOS after fork with no exec

Barry A. Warsaw added the comment: I have a reliable way to call *something* in the pthread_atfork prepare handler, but I honestly don't know what to call to prevent the crash. In the Ruby thread, it seemed to say that you could just dlopen

[issue35254] Process finished with exit code -1073741795 (0xC000001D)

Eryk Sun added the comment: 0xC01D is STATUS_ILLEGAL_INSTRUCTION. This is probably due to a third-party bug that's corrupting the return address on the stack, so I'm closing this issue for now. It can be reopened if the developers of the face_recognition module determine that Python

[issue22412] Towards an asyncio-enabled command line

Change by pmpp : -- nosy: +pmpp ___ Python tracker ___ ___ Python-bugs-list mailing list Unsubscribe:

[issue35254] Process finished with exit code -1073741795 (0xC000001D)

New submission from mengdexiaolian : when i run the following script with face_recognition module, there is a unnormal exception occurs: Process finished with exit code -1073741795 (0xC01D) import face_recognition encodings = [] filename1 = "catherine.jpg" image1 =

[issue35226] mock.call equality surprisingly broken

Change by Roundup Robot : -- keywords: +patch pull_requests: +9798 stage: -> patch review ___ Python tracker ___ ___

[issue28009] core logic of uuid.getnode() is broken for AIX - all versions

Michael Felt added the comment: Historically, I started this issue because I saw that none of the calls made in uuid.py were working for AIX. I also assumed that they ALL worked already, at least somewhere. a) one cause is the difference between AIX and (all) others was the letter chosen

[issue28009] core logic of uuid.getnode() is broken for netstat

Change by Michael Felt : -- title: core logic of uuid.getnode() is broken for AIX - all versions -> core logic of uuid.getnode() is broken for netstat ___ Python tracker ___

[issue35226] mock.call equality surprisingly broken

Chris Withers added the comment: Assuming the PR is merged, what do I need to do for 3.7 and 3.6 backports? There's no doubt a doc for this somewhere, so please just point me at it :-) -- ___ Python tracker

[issue32613] Use PEP 397 py launcher in windows faq

Julien Palard added the comment: New changeset 64313478bcbd0a708c3ce5d4d14f977da56e4be9 by Julien Palard in branch 'master': bpo-32613: Update window FAQ (GH-5552) https://github.com/python/cpython/commit/64313478bcbd0a708c3ce5d4d14f977da56e4be9 --

[issue33196] multiprocessing: serialization must ensure that contexts are compatible (the same)

Change by STINNER Victor : -- title: SEGV in mp.synchronize.Lock.__repr__ in spawn'ed proc if ctx mismatched -> multiprocessing: serialization must ensure that contexts are compatible (the same) ___ Python tracker

[issue33196] multiprocessing: serialization must ensure that contexts are compatible (the same)

STINNER Victor added the comment: > The bug is, if a user makes an error and passes a Queue from context 'fork' > to a child that is spawned using 'spawn', the passed Queue is, for obvious > reasons, broken. Ok. I rewrote the issue title. --

[issue32429] Outdated Modules/Setup warning is invisible

Julien Palard added the comment: Fixed in 961d54c5c1916c09883ebcf7191babc969e5a5cf. -- resolution: -> fixed stage: patch review -> resolved status: open -> closed ___ Python tracker

[issue32613] Use PEP 397 py launcher in windows faq

Change by Julien Palard : -- pull_requests: +9792 ___ Python tracker ___ ___ Python-bugs-list mailing list Unsubscribe:

[issue32613] Use PEP 397 py launcher in windows faq

miss-islington added the comment: New changeset 9053d2f2e09f9583473fb3c1960cba9bda8f3be1 by Miss Islington (bot) (Julien Palard) in branch '3.7': [3.7] bpo-32613: Update window FAQ (GH-5552). (GH-10544) https://github.com/python/cpython/commit/9053d2f2e09f9583473fb3c1960cba9bda8f3be1

[issue33196] multiprocessing: serialization must ensure that contexts are compatible (the same)

Change by Davin Potts : -- nosy: +davin ___ Python tracker ___ ___ Python-bugs-list mailing list Unsubscribe:

[issue35248] RawArray causes FileNotFoundError at cleanup

Change by Mathieu Lamarre : -- components: Library (Lib) nosy: Mathieu Lamarre priority: normal severity: normal status: open title: RawArray causes FileNotFoundError at cleanup type: behavior versions: Python 3.6 ___ Python tracker

[issue35233] _PyMainInterpreterConfig_Copy() doesn't copy install_signal_handlers

STINNER Victor added the comment: New changeset 01de89cb59107d4f889aa503a1c0350dae4aebaf by Victor Stinner in branch 'master': bpo-35233: InitConfigTests tests more config vars (GH-10541) https://github.com/python/cpython/commit/01de89cb59107d4f889aa503a1c0350dae4aebaf --

[issue35242] multiprocessing.Queue in an inconsistent state and a traceback silently suppressed if put an unpickable object and process's target function is finished

Change by Davin Potts : -- nosy: +davin ___ Python tracker ___ ___ Python-bugs-list mailing list Unsubscribe:

[issue35247] test.test_socket.RDSTest.testPeek hangs indefinitely

New submission from markmcclain : test.test_socket.RDSTest.testPeek() can hang indefinitely. ThreadableTest attempts to ensure ordering of operations, but allows both the client and server to proceed without guaranteeing that the data is sent prior to the blocking attempt to peek at the

[issue35247] test.test_socket.RDSTest.testPeek hangs indefinitely

STINNER Victor added the comment: FYI I removed RDSTest.testCongestion() in bpo-34587. > test.test_socket.RDSTest.testPeek() can hang indefinitely. Any advice to explain how to reproduce the issue? Can you try to put a sleep() somewhere to make the race condition more likely? --

[issue35247] test.test_socket.RDSTest.testPeek hangs indefinitely

Change by STINNER Victor : -- nosy: +pablogsal ___ Python tracker ___ ___ Python-bugs-list mailing list Unsubscribe:

[issue32613] Use PEP 397 py launcher in windows faq

Change by Julien Palard : -- pull_requests: +9793 ___ Python tracker ___ ___ Python-bugs-list mailing list Unsubscribe:

[issue32613] Use PEP 397 py launcher in windows faq

Change by Julien Palard : -- resolution: -> fixed stage: patch review -> resolved status: open -> closed ___ Python tracker ___

[issue32613] Use PEP 397 py launcher in windows faq

miss-islington added the comment: New changeset 137da0c5300c0a196084ab9efa61d116304de224 by Miss Islington (bot) (Julien Palard) in branch '3.6': [3.6] bpo-32613: Update window FAQ (GH-5552). (GH-10545) https://github.com/python/cpython/commit/137da0c5300c0a196084ab9efa61d116304de224

[issue35233] _PyMainInterpreterConfig_Copy() doesn't copy install_signal_handlers

Change by STINNER Victor : -- pull_requests: +9794 ___ Python tracker ___ ___ Python-bugs-list mailing list Unsubscribe:

[issue35245] list comprehension for flattened or nested list differ too much

New submission from Yechoh : Suppose we have: forest = [[1,2],[3,4]] and we want: l1 = [['1','2'],['3','4']] we could write: l1 = [[str(leaf) for leaf in tree] for tree in forest] Now if we want: l2 = ['1','2','3','4'] What I expect to need to write is: l2 = [str(leaf) for leaf in tree for

[issue35249] Docs Makefile always rebuilds entire doc

New submission from Jules Lasne : When working on the Documentation or when translating it, I often have to rebuild the doc locally to see the changes I made and see how it looks. However, with the current configuration, It takes (on my computer at least) more than 4 minutes to build. After

[issue35249] Docs Makefile always rebuilds entire doc

Change by Roundup Robot : -- keywords: +patch pull_requests: +9795 stage: -> patch review ___ Python tracker ___ ___

[issue35249] Docs Makefile always rebuilds entire doc

Julien Palard added the comment: A `make html` builds the doc in 11s on my machine on the 2nd run, I think this is the build you're searching? -- nosy: +mdk ___ Python tracker

[issue35250] Minor parameter documentation mismatch for turtle

New submission from Shane : I noticed a slight mismatch in the parameter documentation for one of turtle's functions. onclick() accepts the parameters (fun, btn, add), but the documentation describes the parameters (fun, num, add). A minor issue, to be sure, but I wanted to point it out

[issue33725] Python crashes on macOS after fork with no exec

Barry A. Warsaw added the comment: On Nov 14, 2018, at 10:11, Davin Potts wrote: > > > Davin Potts added the comment: > > Barry's effort as well as comments in other links seem to all suggest that > OBJC_DISABLE_INITIALIZE_FORK_SAFETY is not comprehensive in its ability to > make other

[issue31818] [macOS] _scproxy.get_proxies() crash -- get_proxies() is not fork-safe?

Change by Barry A. Warsaw : -- versions: +Python 3.6, Python 3.7, Python 3.8 ___ Python tracker ___ ___ Python-bugs-list mailing

[issue31818] [macOS] _scproxy.get_proxies() crash -- get_proxies() is not fork-safe?

Change by Barry A. Warsaw : -- nosy: +barry ___ Python tracker ___ ___ Python-bugs-list mailing list Unsubscribe:

[issue35249] Docs Makefile always rebuilds entire doc

Jules Lasne added the comment: Fixed in https://github.com/python/python-docs-fr/pull/426 -- stage: patch review -> resolved status: open -> closed ___ Python tracker ___

[issue35249] Docs Makefile always rebuilds entire doc

Change by Julien Palard : -- resolution: -> works for me stage: resolved -> patch review ___ Python tracker ___ ___

[issue35249] Docs Makefile always rebuilds entire doc

Change by Julien Palard : -- stage: patch review -> resolved ___ Python tracker ___ ___ Python-bugs-list mailing list Unsubscribe:

[issue33725] Python crashes on macOS after fork with no exec

Davin Potts added the comment: Given the original post mentioned 2.7.15, I wonder if it is feasible to fork near the beginning of execution, then maintain and pass around a multiprocessing.Pool to be used when needed instead of dynamically forking? Working with legacy code is almost always

[issue35219] macOS 10.14 Mojave crashes in multiprocessing

Davin Potts added the comment: Resolution is marked dupe but status is still open. Are we closing this one or is there a more specific remedy for this situation (as opposed to what issue33725 discusses) that would be helpful to document? -- nosy: +davin

[issue35214] Get the test suite passing with clang Memory Sanitizer enabled

Gregory P. Smith added the comment: New changeset b6f4472dc4190e2fd668490d86aeefd2ab0df935 by Gregory P. Smith in branch '2.7': [2.7] bpo-35214: Fix OOB memory access in unicode escape parser (GH-10506) (GH-10538)

[issue33725] Python crashes on macOS after fork with no exec

Barry A. Warsaw added the comment: FWIW, I suspect that setting the environment variable only helps if it's done before the process starts. You cannot set it before the fork and have it affect the child. -- ___ Python tracker

[issue35233] _PyMainInterpreterConfig_Copy() doesn't copy install_signal_handlers

STINNER Victor added the comment: New changeset 9ee1d42f019ac827f73479ce241e95733d050e67 by Victor Stinner in branch '3.7': bpo-35233: InitConfigTests tests more config vars (GH-10541) (GH-10546) https://github.com/python/cpython/commit/9ee1d42f019ac827f73479ce241e95733d050e67 --

[issue33725] Python crashes on macOS after fork with no exec

Antoine Pitrou added the comment: Legacy code is easy to migrate as long as it uses Python 3. Just call mp.set_start_method('forkserver') at the top of your code and you're done. Some use cases may fail (if sharing non-picklable types), but they're probably not very common. --

[issue31354] Fixing a bug related to LTO only build

Ned Deily added the comment: This does sound like a regression bug that should be fixed in all branches. Another issue is that, if it is propagated to third-party extension module builds through distutils, there is no guarantee that the extension module is being built with the same

[issue35248] RawArray causes FileNotFoundError at cleanup

Mathieu Lamarre added the comment: To repro, the temp directory returned by tempfile must be a network mount to reproduce this error. I have this on server with minimal disk space in /home and TMPDIR points to CIFS mounted volume. So maybe can only be reproduced if shutils operations on the

[issue33725] Python crashes on macOS after fork with no exec

Ned Deily added the comment: _scproxy has been known to be problematic for some time, see for instance Issue31818. That issue also gives a simple workaround: setting urllib's "no_proxy" environment variable to "*" will prevent the calls to the System Configuration framework. --

[issue35248] RawArray causes FileNotFoundError at cleanup

New submission from Mathieu Lamarre : Running: from multiprocessing.sharedctypes import RawArray from ctypes import c_uint32 if __name__ == '__main__': shared_array = RawArray(c_uint32, 1500) Causes: Traceback (most recent call last): File

[issue33725] Python crashes on macOS after fork with no exec

Davin Potts added the comment: Barry's effort as well as comments in other links seem to all suggest that OBJC_DISABLE_INITIALIZE_FORK_SAFETY is not comprehensive in its ability to make other threads "safe" before forking. "Objective-C classes defined by the OS frameworks remain

[issue35219] macOS 10.14 Mojave crashes in multiprocessing

Barry A. Warsaw added the comment: Closing in favor of issue33725 -- stage: -> resolved status: open -> closed ___ Python tracker ___