[issue43565] PyUnicode_KIND macro does not has specified return type

Petr Viktorin added the comment: New changeset 47fd4726a2ce8599cc397ddeae40f70eb471e868 by Ammar Askar in branch 'main': bpo-43565: Document PyUnicode_KIND's return type as an unsigned int (GH-25724) https://github.com/python/cpython/commit/47fd4726a2ce8599cc397ddeae40f70eb471e868

[issue42035] [C API] PyType_GetSlot cannot get tp_name

Petr Viktorin added the comment: New changeset a390ebea17a96d1c93fc5f75b1e19916090a4561 by Hai Shi in branch 'main': bpo-42035: Add a PyType_GetName() to get type's short name. (GH-23903) https://github.com/python/cpython/commit/a390ebea17a96d1c93fc5f75b1e19916090a4561

[issue42035] [C API] PyType_GetSlot cannot get tp_name

Petr Viktorin added the comment: > Can we create PyType_GetDataSlot function to return the data pointer? No, see the borrowed pointer issues above. > I will create another PR after PR-23903 merged. Great, thank you! -- ___ Python tracker

[issue44688] [sqlite3] Remove ASCII limitation from sqlite3.Connection.create_collation()

Petr Viktorin added the comment: New changeset 5269c091458c5ea76eb625e4fabc9980b6309266 by Erlend Egeberg Aasland in branch 'main': bpo-44688: Remove ASCII limitation from `sqlite3` collation names (GH-27395) https://github.com/python/cpython/commit/5269c091458c5ea76eb625e4fabc9980b6309266

[issue44769] socketserver.shutdown should stop serve_forever() immediately

Petr Viktorin added the comment: This isn't a high priority for me, but I have a WIP branch in https://github.com/encukou/cpython/tree/http-server-poll Feel free to continue it if you get to it sooner than I do. -- ___ Python tracker <ht

[issue44769] socketserver.shutdown should stop serve_forever() immediately

New submission from Petr Viktorin : Currently, socketserver.serve_forever() sets a variable and serve_forever() polls for it, with a configurable interval. A comment in the code already says: # XXX: Consider using another file descriptor or connecting to the # socket to wake this up instead

[issue43377] _PyErr_Display should be available in the CPython-specific API

Petr Viktorin added the comment: PyErr_Display (without underscore) is already exposed. This issue is about adding a variant with an additional output file argument. The function formats an exception "just as Python would". The exact output will naturally be different betwee

[issue29298] argparse fails with required subparsers, un-named dest, and empty argv

Change by Petr Viktorin : -- resolution: -> fixed stage: patch review -> resolved status: open -> closed ___ Python tracker <https://bugs.python.or

[issue43235] Tools/scripts/stable_abi.py should also check PC/python3dll.c (Windows stable ABI)

Petr Viktorin added the comment: Since PEP 652, PC/python3dll.c is generated from the stable ABI definition. Checking (i.e. running the tool without --generate) ensures it is up-to-date. -- resolution: -> fixed stage: patch review -> resolved status: open -&g

[issue43377] _PyErr_Display should be available in the CPython-specific API

Petr Viktorin added the comment: If it should be exposed, it should be renamed to a public (non-underscored) name, like PyErr_DisplayToFile. It should also ideally get documentation and tests (along with PyErr_Display, which is sadly missing those as well). (Also, note

[issue42064] Convert sqlite3 to multi-phase initialisation (PEP 489)

Petr Viktorin added the comment: New changeset 890e22957d427ee994b85d62dfe4d5a7cbd34ec5 by Erlend Egeberg Aasland in branch 'main': bpo-42064: Migrate to `sqlite3_create_collation_v2` (GH-27156) https://github.com/python/cpython/commit/890e22957d427ee994b85d62dfe4d5a7cbd34ec5

[issue42035] [C API] PyType_GetSlot cannot get tp_name

Petr Viktorin added the comment: Sorry for the delay; getting 652 into Python 3.10 took up most of my time. So, in the current proposal: - `PyType_GetName(t)` is equivalent to `PyObject_GetAttrString(t, "__name__")` - for the qualified name you can use `PyObject_GetAt

[issue42747] Remove Py_TPFLAGS_HAVE_VERSION_TAG flag?

Petr Viktorin added the comment: I usually wait until buildbots are green before closing the bpo, but I don't think there's anything else. -- resolution: -> fixed stage: patch review -> resolved status: open -> closed ___ Python tracke

[issue41756] Do not always use exceptions to return result from coroutine

Petr Viktorin added the comment: that is, bpo-44727 -- ___ Python tracker <https://bugs.python.org/issue41756> ___ ___ Python-bugs-list mailing list Unsub

[issue41756] Do not always use exceptions to return result from coroutine

Petr Viktorin added the comment: Hello! This change added an enum to the stable ABI: typedef enum { PYGEN_RETURN = 0, PYGEN_ERROR = -1, PYGEN_NEXT = 1, } PySendResult; Adding new values to enums might break the stable ABI in some (admittedly rare) cases; so

[issue44727] Stable ABI should avoid `enum`

Petr Viktorin added the comment: As far as I can see, the current enums in the stable ABI are: PySendResult from object.h, return value of PyObject_Send: typedef enum { PYGEN_RETURN = 0, PYGEN_ERROR = -1, PYGEN_NEXT = 1, } PySendResult; (This is unlikely

[issue44727] Stable ABI should avoid `enum`

Petr Viktorin added the comment: Devguide PR: https://github.com/python/devguide/pull/730 -- ___ Python tracker <https://bugs.python.org/issue44727> ___ ___

[issue44727] Stable ABI should avoid `enum`

New submission from Petr Viktorin : Adding a new enumerator to a C enum can change the size of the type, which would break the ABI. This is not often a problem in practice, but the rules around when it is a problem and when it isn't are complicated enough that I believe enum should

[issue42747] Remove Py_TPFLAGS_HAVE_VERSION_TAG flag?

Petr Viktorin added the comment: New changeset 632e8a69593efb12ec58d90e624ddf249a7a1b65 by Miss Islington (bot) in branch '3.10': bpo-42747: Remove Py_TPFLAGS_HAVE_AM_SEND and make Py_TPFLAGS_HAVE_VERSION_TAG no-op (GH-27260) (GH-27306) https://github.com/python/cpython/commit

[issue29298] argparse fails with required subparsers, un-named dest, and empty argv

Petr Viktorin added the comment: New changeset 097801844c99ea3916bebe1cc761257ea7083d34 by Miss Islington (bot) in branch '3.9': bpo-29298: Fix crash with required subparsers without dest (GH-3680) (GH-27304) https://github.com/python/cpython/commit/097801844c99ea3916bebe1cc761257ea7083d34

[issue29298] argparse fails with required subparsers, un-named dest, and empty argv

Petr Viktorin added the comment: New changeset c589992e09d0db7cb47d21d5948929e599fdbb94 by Miss Islington (bot) in branch '3.10': bpo-29298: Fix crash with required subparsers without dest (GH-3680) (GH-27303) https://github.com/python/cpython/commit/c589992e09d0db7cb47d21d5948929e599fdbb94

[issue44588] Possible double Py_XDECREF in cpython typeobject.c

Change by Petr Viktorin : -- keywords: +patch pull_requests: +25808 stage: -> patch review pull_request: https://github.com/python/cpython/pull/27260 ___ Python tracker <https://bugs.python.org/issu

[issue44588] Possible double Py_XDECREF in cpython typeobject.c

Petr Viktorin added the comment: Please use Py_TPFLAGS_DEFAULT when creating objects. Do you have a reason to not do it? The flag Py_TPFLAGS_HAVE_VERSION_TAG is unneeded and Python should not check for it. There's bpo-42747 open for that already; I sent a PR for it. -- nosy

[issue42747] Remove Py_TPFLAGS_HAVE_VERSION_TAG flag?

Change by Petr Viktorin : -- keywords: +patch pull_requests: +25805 stage: -> patch review pull_request: https://github.com/python/cpython/pull/27260 ___ Python tracker <https://bugs.python.org/issu

[issue42747] Remove Py_TPFLAGS_HAVE_VERSION_TAG flag?

Petr Viktorin added the comment: The bit cannot be repurposed, since older extensions using the stable ABI might set it. It doesn't make much sense to remove the Py_TPFLAGS_HAVE_VERSION_TAG or Py_TPFLAGS_HAVE_FINALIZE defines; I'd let them stay to document that the bits are reserved

[issue42085] Add dedicated slot for sending values

Petr Viktorin added the comment: Py_TPFLAGS_HAVE_AM_SEND is unnecessary and I'd like to remove it. It is not possible for type objects to have a different layout than the interpreter: - extensions using the regular ABI must be recompiled for each feature version of Python - extensions using

[issue44663] Possible bug in datetime utc

Change by Petr Viktorin : -- components: +Library (Lib) -C API ___ Python tracker <https://bugs.python.org/issue44663> ___ ___ Python-bugs-list mailing list Unsub

[issue43334] venv does not install libpython

Petr Viktorin added the comment: A venv does *not* create a replica of a python installation directory. Files shared with the main Python installation are not copied nor linked. This goes for the standard library, as well as the libraries and headers. (*Executables* are an exception; they're

[issue43629] fix _PyRun_SimpleFileObject create __main__ module and cache. Call this function multiple times, the attributes stored in the module dict will affect eachother.

Petr Viktorin added the comment: The public function that calls _PyRun_SimpleFileObject is PyRun_SimpleStringFlags, and the behavior is as specified in its documentation: "Executes the Python source code from command in the __main__ module according to the flags argument. If __main__

[issue42064] Convert sqlite3 to multi-phase initialisation (PEP 489)

Petr Viktorin added the comment: New changeset 4c0deb25ac899fbe4da626ce3cb21f204cdd3aa9 by Erlend Egeberg Aasland in branch 'main': bpo-42064: Finalise establishing sqlite3 global state (GH-27155) https://github.com/python/cpython/commit/4c0deb25ac899fbe4da626ce3cb21f204cdd3aa9

[issue25653] ctypes+callbacks+fork+selinux = crash

Petr Viktorin added the comment: Here's a simpler reproducer. -- nosy: +petr.viktorin Added file: https://bugs.python.org/file50134/y.py ___ Python tracker <https://bugs.python.org/issue25

[issue40939] Remove the old parser

Petr Viktorin added the comment: New changeset dc10264eb880ed63fcf42c17057f3f5d879a0a0c by Miss Islington (bot) in branch '3.10': bpo-40939: Remove documentation for `PyParser_*` & add porting notes (GH-26855) (GH-26898) https://github.com/python/cpython/co

[issue40939] Remove the old parser

Petr Viktorin added the comment: New changeset 29987f72650b7cccee4df216c8297e8484a44e6a by Petr Viktorin in branch 'main': bpo-40939: Remove documentation for `PyParser_*` & add porting notes (GH-26855) https://github.com/python/cpython/commit/29987f72650b7cccee4df216c8297e8484a4

[issue44498] move deprecated asynchat, asyncore and smtpd from the stdlib to test.support

Petr Viktorin added the comment: Please follow the backwards compatibility policy (PEP 387): - Have a discussion - Raise deprecation warnings - Wait at least two minor Python versions where warnings are raised - Consider any feedback - Then remove stuff (Or ask the SC for an exception

[issue40939] Remove the old parser

Petr Viktorin added the comment: - The removed functions are still listed in the documentation; this is confusing. - `struct _node` is not very usable in the C API, but what I saw in mod_wsgi was giving gave PyParser_* output to PyNode_Compile. I tried to write porting notes for this usage

[issue40939] Remove the old parser

Change by Petr Viktorin : -- nosy: +petr.viktorin nosy_count: 11.0 -> 12.0 pull_requests: +25434 pull_request: https://github.com/python/cpython/pull/26855 ___ Python tracker <https://bugs.python.org/issu

[issue39767] create multiprocessing.SharedMemory by pointing to existing memoryview

Change by Petr Viktorin : -- components: +Library (Lib) -C API ___ Python tracker <https://bugs.python.org/issue39767> ___ ___ Python-bugs-list mailing list Unsub

[issue39620] PyObject_GetAttrString and tp_getattr do not agree

Change by Petr Viktorin : -- nosy: +petr.viktorin ___ Python tracker <https://bugs.python.org/issue39620> ___ ___ Python-bugs-list mailing list Unsubscribe:

[issue39355] The Python library will not compile with a C++2020 compiler because the code uses the reserved “module” keyword

Change by Petr Viktorin : -- nosy: +petr.viktorin ___ Python tracker <https://bugs.python.org/issue39355> ___ ___ Python-bugs-list mailing list Unsubscribe:

[issue39123] PyThread_xxx() not available when using limited API

Petr Viktorin added the comment: In Python 3.10, the functions: PyThread_allocate_lock, PyThread_exit_thread, PyThread_free_lock, PyThread_get_stacksize, PyThread_get_thread_ident, PyThread_get_thread_native_id, PyThread_init_thread, PyThread_release_lock, PyThread_set_stacksize

[issue39078] __function.__defaults__ breaks for __init__ of dataclasses with default factory

Change by Petr Viktorin : -- components: +Library (Lib) -C API ___ Python tracker <https://bugs.python.org/issue39078> ___ ___ Python-bugs-list mailing list Unsub

[issue38829] Make the function flush_io accessible in the C-API

Petr Viktorin added the comment: To follow up on the StackOverflow discussion: A call to PyErr_Print should invoke sys.excepthook. Unless sys.excepthook was changed, this should print the message *and* flush standard output. If it doesn't, I recommend investigating why. If that does

[issue44441] Malformed PyImport_Inittab after re-initialization

Change by Petr Viktorin : -- nosy: +vstinner ___ Python tracker <https://bugs.python.org/issue1> ___ ___ Python-bugs-list mailing list Unsubscribe:

[issue43795] Implement PEP 652 -- Maintaining the Stable ABI

Petr Viktorin added the comment: New changeset 7cad9cb51bdae2144cbab330f13a607ba3471742 by Petr Viktorin in branch 'main': bpo-43795: Don't list private names in the limited API (GH-26740) https://github.com/python/cpython/commit/7cad9cb51bdae2144cbab330f13a607ba3471742

[issue43795] Implement PEP 652 -- Maintaining the Stable ABI

Change by Petr Viktorin : -- pull_requests: +25326 pull_request: https://github.com/python/cpython/pull/26740 ___ Python tracker <https://bugs.python.org/issue43

[issue42064] Convert sqlite3 to multi-phase initialisation (PEP 489)

Petr Viktorin added the comment: New changeset 10a5c806d4dec6c342dcc9888fbe4fa1fa9b7a1f by Erlend Egeberg Aasland in branch 'main': bpo-42064: Move sqlite3 types to global state (GH-26537) https://github.com/python/cpython/commit/10a5c806d4dec6c342dcc9888fbe4fa1fa9b7a1f -- nosy

[issue44351] distutils.sysconfig.parse_makefile() regression in Python 3.10

Petr Viktorin added the comment: New changeset 2f2ea96c4429b81f491aa1cdc4219ef2fd6d37fb by Miss Islington (bot) in branch '3.10': bpo-44351: Restore back parse_makefile in distutils.sysconfig (GH-26637) (GH-26673) https://github.com/python/cpython/commit

[issue44392] Py_GenericAlias is not documented

Petr Viktorin added the comment: > Can we remove this from the stable API during beta? It was added (to PC/python3.def) in 3.9, so removing in 3.10 beta wouldn't be good. It can be deprecated; should it? -- ___ Python tracker <

[issue44351] distutils.sysconfig.parse_makefile() regression in Python 3.10

Petr Viktorin added the comment: New changeset fc98266ff627ba0f56f8ae241245b66bc983baa3 by Lumír 'Frenzy' Balhar in branch 'main': bpo-44351: Restore back parse_makefile in distutils.sysconfig (GH-26637) https://github.com/python/cpython/commit/fc98266ff627ba0f56f8ae241245b66bc983baa3

[issue44351] distutils.sysconfig.parse_makefile() regression in Python 3.10

Petr Viktorin added the comment: IMO, the functionality should only be preserved until distutils is removed. So: - distutils.sysconfig.parse_makefile should use TextFile as before, so projects that use it aren't broken *yet* - nothing else should call distutils.sysconfig.parse_makefile, so

[issue43795] Implement PEP 652 -- Maintaining the Stable ABI

Petr Viktorin added the comment: New changeset 75185561a9a3b6dede3ad87bd83bab66847bd425 by Miss Islington (bot) in branch '3.10': bpo-43795: Note Stable ABI PEP in What's New (GH-26479) (GH-26603) https://github.com/python/cpython/commit/75185561a9a3b6dede3ad87bd83bab66847bd425

[issue43795] Implement PEP 652 -- Maintaining the Stable ABI

Petr Viktorin added the comment: New changeset 257e400a19b34c7da6e2aa500d80b54e4c4dbf6f by Petr Viktorin in branch 'main': bpo-43795: Note Stable ABI PEP in What's New (GH-26479) https://github.com/python/cpython/commit/257e400a19b34c7da6e2aa500d80b54e4c4dbf6f

[issue44285] Coverity scan: Modules/getpath.c. "calculate_open_pyenv" allocates memory that is stored into "env_file".

Petr Viktorin added the comment: New changeset bdb56902a3bfe12b10f85a941d5dd0eae739f1a8 by stratakis in branch 'main': bpo-44285: getpath.c: Assert that env_file is NULL during an error check (GH-26486) https://github.com/python/cpython/commit/bdb56902a3bfe12b10f85a941d5dd0eae739f1a8

[issue44285] Coverity scan: Modules/getpath.c. "calculate_open_pyenv" allocates memory that is stored into "env_file".

Petr Viktorin added the comment: +1, for adding the assertion. It's not trivial to see that env_file must be NULL here, even for me (a human). -- nosy: +petr.viktorin ___ Python tracker <https://bugs.python.org/issue44

[issue29086] Document C API that is not part of the limited API

Petr Viktorin added the comment: PEP 652 is now in, and with it a list of everything that *is* in the stable ABI: https://docs.python.org/3.10/c-api/stable.html#contents-of-limited-api I'm closing the issue. -- resolution: -> fixed stage: patch review -> resolved status

[issue23903] Generate PC/python3.def by scraping headers

Petr Viktorin added the comment: PEP 652 (bpo-43795) is now in; I'm closing this issue. A cross-platform C parser/checker would still be an improvement, but the constraints (no external dependencies in CPython) and benefits (not many compared to the goal of strict separation of internal

[issue43795] Implement PEP 652 -- Maintaining the Stable ABI

Change by Petr Viktorin : -- pull_requests: +25074 pull_request: https://github.com/python/cpython/pull/26479 ___ Python tracker <https://bugs.python.org/issue43

[issue43795] Implement PEP 652 -- Maintaining the Stable ABI

Change by Petr Viktorin : -- pull_requests: +24946 pull_request: https://github.com/python/cpython/pull/26354 ___ Python tracker <https://bugs.python.org/issue43

[issue44230] lookup extensions with the stable ABI under a platform specific name

Petr Viktorin added the comment: How will these filenames be generated? Won't build tools like setuptools need to be changed as well? -- ___ Python tracker <https://bugs.python.org/issue44

[issue41282] Deprecate and remove distutils

Petr Viktorin added the comment: Thank you for doing the hard part, Ned! -- ___ Python tracker <https://bugs.python.org/issue41282> ___ ___ Python-bugs-list m

[issue41282] Deprecate and remove distutils

Change by Petr Viktorin : -- pull_requests: +24921 pull_request: https://github.com/python/cpython/pull/26329 ___ Python tracker <https://bugs.python.org/issue41

[issue41282] Deprecate and remove distutils

Petr Viktorin added the comment: When building Python, we need two distinct "include" directories: - source .h files - install target for .h files Note that this doesn't matter except when building Python from source. Historically: - source .h files were in the sysconfig sc

[issue44099] [C API] Introduce a new slot in PyModuleDef to hold the classes

Petr Viktorin added the comment: First off, note that you can use PyModule_AddType rather than PyModule_AddObject to avoid repeating the name. Adding this *correctly* will be somewhat involved: slots take function pointers, not data pointers which are incompatible according to standard C

[issue43795] Implement PEP 652 -- Maintaining the Stable ABI

Change by Petr Viktorin : -- pull_requests: +24856 pull_request: https://github.com/python/cpython/pull/26241 ___ Python tracker <https://bugs.python.org/issue43

[issue42591] Method Py_FrozenMain missing in libpython3.9

Change by Petr Viktorin : -- pull_requests: +24857 pull_request: https://github.com/python/cpython/pull/26241 ___ Python tracker <https://bugs.python.org/issue42

[issue44131] [C API] Add tests on Py_FrozenMain()

Change by Petr Viktorin : -- pull_requests: +24858 pull_request: https://github.com/python/cpython/pull/26241 ___ Python tracker <https://bugs.python.org/issue44

[issue42591] Method Py_FrozenMain missing in libpython3.9

Petr Viktorin added the comment: Is this function actually usable in Windows? ISTM that you need to define three more functions, PyWinFreeze_ExeInit, PyWinFreeze_ExeTerm and PyInitFrozenExtensions. Was adding this undocumented function to the Windows stable ABI deliberate? I see

[issue44131] [C API] Add tests on Py_FrozenMain()

Petr Viktorin added the comment: The tests are now skipped on Windows. It might be better to use a feature check: hasattr(ctypes.pythonapi, 'Py_FrozenMain') IMO, Py_FrozenMain is quite specific: it's OK if it's exported, but also OK if it's missing. (Unless you're compiling a special build

[issue41111] [C API] Convert a few stdlib extensions to the limited C API (PEP 384)

Petr Viktorin added the comment: METH_VARARGS and METH_KEYWORDS are part of the stable ABI. -- ___ Python tracker <https://bugs.python.org/issue4> ___ ___

[issue43795] Implement PEP 652 -- Maintaining the Stable ABI

Change by Petr Viktorin : -- pull_requests: +24742 pull_request: https://github.com/python/cpython/pull/26101 ___ Python tracker <https://bugs.python.org/issue43

[issue44116] The _csv module can't be garbage-collected after _csv.register_dialect is called

Petr Viktorin added the comment: Changes to _csv.Error should not be necessary, there everything is handled by the superclass. -- stage: patch review -> ___ Python tracker <https://bugs.python.org/issu

[issue44116] The _csv module can't be garbage-collected after _csv.register_dialect is called

Petr Viktorin added the comment: The urllib.request one was caused by _hashlib, see GH-26072. -- stage: patch review -> ___ Python tracker <https://bugs.python.org/issu

[issue40645] Use OpenSSL's HMAC API

Change by Petr Viktorin : -- nosy: +petr.viktorin nosy_count: 5.0 -> 6.0 pull_requests: +24712 pull_request: https://github.com/python/cpython/pull/26072 ___ Python tracker <https://bugs.python.org/issu

[issue44116] The _csv module can't be garbage-collected after _csv.register_dialect is called

Petr Viktorin added the comment: *facepalm* Yes, that's it. If you have the time now, could you send he PR? -- ___ Python tracker <https://bugs.python.org/issue44

[issue44116] The _csv module can't be garbage-collected after _csv.register_dialect is called

Change by Petr Viktorin : -- keywords: +patch pull_requests: +24709 stage: -> patch review pull_request: https://github.com/python/cpython/pull/26068 ___ Python tracker <https://bugs.python.org/issu

[issue44116] The _csv module can't be garbage-collected after _csv.register_dialect is called

Petr Viktorin added the comment: Hm, a similar thing apparently happens with urllib.request. -- ___ Python tracker <https://bugs.python.org/issue44116> ___ ___

[issue43795] Implement PEP 652 -- Maintaining the Stable ABI

Petr Viktorin added the comment: I opened https://bugs.python.org/issue44116 for the leak when trying to unload _csv. Looks like investigating that will take a while, so I'll send a small PR to unblock the buildbots. -- ___ Python tracker <ht

[issue44116] The _csv module can't be garbage-collected after _csv.register_dialect is called

New submission from Petr Viktorin : After `_csv.register_dialect` is called, the csv module is alive even after it's removed from sys.modules. It should be garbage-collected. (It's not that big a deal: unloading _csv isn't something users should do. But it might be hiding a deeper issue

[issue43795] Implement PEP 652 -- Maintaining the Stable ABI

Petr Viktorin added the comment: Thanks for the note! This is due to `csv`. Adding a script in Tools/scripts that imports csv will also make the refleak check fail. I'll investigate further. -- ___ Python tracker <https://bugs.python.

[issue40222] "Zero cost" exception handling

Petr Viktorin added the comment: Then, according to PEP 387, "The steering council may grant exceptions to this policy." I think API breaks like this do need coordination at the project level. -- ___ Python tracker <https://bu

[issue40222] "Zero cost" exception handling

Petr Viktorin added the comment: PyCode_NewWithPosOnlyArgs is not part of the stable ABI. It is OK to break its ABI in a minor version (i.e. 3.11). The PyAPI_FUNC makes it part of the public *API*. It needs to be source- compatible; the number of arguments can't change. Could yo u add a new

[issue43760] The DISPATCH() macro is not as efficient as it could be.

Petr Viktorin added the comment: PEP 0497 is rejected; the active one is PEP 387, which says "backwards incompatibility" means preexisting code ceases to comparatively function after a change. So, this does look like a backwards-incompatible change. Unfortunately, not all of

[issue23903] Generate PC/python3.def by scraping headers

Petr Viktorin added the comment: The symbols exported by python3.dll are now generated from Misc/stable_abi.txt as per PEP 652. See the devguide for more details: https://devguide.python.org/c-api/ This is slightly more work than generating the list fully automatically, but the extra work

[issue43795] Implement PEP 652 -- Maintaining the Stable ABI

Change by Petr Viktorin : -- pull_requests: +24588 pull_request: https://github.com/python/cpython/pull/25920 ___ Python tracker <https://bugs.python.org/issue43

[issue42083] PyStructSequence_NewType broken in 3.8

Petr Viktorin added the comment: Changing PyType_FromSpec* to accept NULL has an issue: extensions built and tested with 3.9.5 would not work with the earlier 3.9s. I'll send a PR to fix just PyStructSequence_NewType. -- ___ Python tracker

[issue42083] PyStructSequence_NewType broken in 3.8

Change by Petr Viktorin : -- nosy: +petr.viktorin nosy_count: 5.0 -> 6.0 pull_requests: +24568 pull_request: https://github.com/python/cpython/pull/25896 ___ Python tracker <https://bugs.python.org/issu

[issue43976] Allow Python distributors to add custom site install schemes

Petr Viktorin added the comment: Sorry for not getting to this sooner, but 5 days is really tight for such a change. With -S/-I, It would be great if sys.path only included packages installed as part of the OS, and not those installed by `sudo pip`. (Or `pip --user`, but that's covered

[issue28254] Add C API for gc.enable, gc.disable, and gc.isenabled

Petr Viktorin added the comment: New changeset 14fc2bdfab857718429029e53ceffca456178827 by Petr Viktorin in branch 'master': bpo-28254: Add PyGC_ functions to the stable ABI manifest (GH-25720) https://github.com/python/cpython/commit/14fc2bdfab857718429029e53ceffca456178827

[issue28254] Add C API for gc.enable, gc.disable, and gc.isenabled

Change by Petr Viktorin : -- nosy: +petr.viktorin nosy_count: 5.0 -> 6.0 pull_requests: +24411 pull_request: https://github.com/python/cpython/pull/25720 ___ Python tracker <https://bugs.python.org/issu

[issue43795] Implement PEP 652 -- Maintaining the Stable ABI

Petr Viktorin added the comment: New changeset f6ee4dad589c0953283dacb577a2d808fda7aae9 by Petr Viktorin in branch 'master': bpo-43795: Generate python3dll.c and doc data from manifest (PEP 652) (GH-25315) https://github.com/python/cpython/commit/f6ee4dad589c0953283dacb577a2d808fda7aae9

[issue43795] Implement PEP 652 -- Maintaining the Stable ABI

Petr Viktorin added the comment: I hope the PR fixes that. I plan to merge tomorrow if there ar no objections to it. On April 28, 2021 5:15:19 PM GMT+02:00, STINNER Victor wrote: > >STINNER Victor added the comment: > >Right now, running "make regen-limited-abi" adds

[issue43795] Implement PEP 652 -- Maintaining the Stable ABI

Change by Petr Viktorin : -- pull_requests: +24357 pull_request: https://github.com/python/cpython/pull/25668 ___ Python tracker <https://bugs.python.org/issue43

[issue43933] Regression in python3.10 with traceback frame having lineno of -1

Petr Viktorin added the comment: Please don't lose the big picture here. Previously, the reported line number was *somewhere close* to the currently executing code. That's typically all I need from a traceback: the fix for a bug often needs to go a few lines above/below where it's reported

[issue43868] Remove PyOS_ReadlineFunctionPointer from the stable ABI list

Petr Viktorin added the comment: New changeset 91b69b77cf5f78de6d35dea23098df34b6fd9e53 by Petr Viktorin in branch 'master': bpo-43868: Remove PyOS_ReadlineFunctionPointer from the stable ABI list (GH-25442) https://github.com/python/cpython/commit/91b69b77cf5f78de6d35dea23098df34b6fd9e53

[issue43795] Implement PEP 652 -- Maintaining the Stable ABI

Petr Viktorin added the comment: New changeset 9d6a2d0ee7e55402656e1dec46400591b62db331 by Petr Viktorin in branch 'master': bpo-43795: PEP-652: Clean up the stable ABI/limited API (GH-25482) https://github.com/python/cpython/commit/9d6a2d0ee7e55402656e1dec46400591b62db331

[issue43795] Implement PEP 652 -- Maintaining the Stable ABI

Petr Viktorin added the comment: New changeset e7cc64e297001cc79b9afab80f71d9e6d1267cb7 by Petr Viktorin in branch 'master': bpo-43795: PEP-652: Simplify headers for easier static analysis (GH-25483) https://github.com/python/cpython/commit/e7cc64e297001cc79b9afab80f71d9e6d1267cb7

[issue41282] Deprecate and remove distutils

Petr Viktorin added the comment: New changeset 90d02e5e63e2cb8f66a2c0dd2ea8d7d4f45f4ebf by Lumír 'Frenzy' Balhar in branch 'master': bpo-41282: (PEP 632) Deprecate distutils.sysconfig (partial implementation of the PEP) (GH-23142) https://github.com/python/cpython/commit

[issue37578] Change Glob: Allow Recursion for Hidden Files

Petr Viktorin added the comment: I wonder if it would be worth it to add a new option to include hidden files (except . and ..) For the record, setuptools' fork of glob does this unconditionally: https://github.com/pypa/setuptools/blob/main/setuptools/glob.py -- nosy: +petr.viktorin

[issue43795] Implement PEP 652 -- Maintaining the Stable ABI

Change by Petr Viktorin : -- pull_requests: +24208 pull_request: https://github.com/python/cpython/pull/25483 ___ Python tracker <https://bugs.python.org/issue43

[issue43795] Implement PEP 652 -- Maintaining the Stable ABI

Petr Viktorin added the comment: > Should we mention PEP 652 in Include/README.rst, now that the PEP is accepted? No, we should link to the documentation (when it's written). The PEP is a design document; it'll become outdated. -- ___ Pyt

<    1   2   3   4   5   6   7   8   9   10   >