[issue40863] bytes.decode changes/destroys line endings on windows

New submission from Matthias Naegler : ``` # 0x0D, 13 = /r # 0x0A, 10 = /n print('test "\\r\\n"') print('-') b = bytes([0x41, 0x0D, 0x0A]) print("bytes: %s" % b) print("string: %s" % b.decode('utf8'), end='') # expected string: "A\r\n" # ressult string: "A\r\r\n" print('test

[issue40864] spec_set/autospec/spec seems to not be reading attributes defined in class body

Evan Fagerberg added the comment: Sorry one small note, the error in the example happens on ```python logger.propagate = False ``` and not ```python assert logger.propagate is False ``` -- ___ Python tracker

[issue40826] PyOS_InterruptOccurred() now requires to hold the GIL: PyOS_Readline() crash

STINNER Victor added the comment: I am confused. Code which uses the closed file descriptor 0 uses "_Py_BEGIN_SUPPRESS_IPH": _Py_BEGIN_SUPPRESS_IPH hStdIn = (HANDLE)_get_osfhandle(fileno(sys_stdin)); hStdErr = (HANDLE)_get_osfhandle(fileno(stderr));

[issue40863] bytes.decode changes/destroys line endings on windows

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

[issue40686] Compiler warnings in _zoneinfo.c on Windows build in 64-bit

Change by Paul Ganssle : -- pull_requests: +19849 pull_request: https://github.com/python/cpython/pull/20624 ___ Python tracker ___

[issue40275] test.support has way too many imports

Change by hai shi : -- pull_requests: +19850 pull_request: https://github.com/python/cpython/pull/20625 ___ Python tracker ___ ___

[issue40836] logging.fatal() and logging.Logger.fatal() should raise a DeprecationWarning

Rémi Lapeyre added the comment: Thanks for the feedback, I updated the PR accordingly. -- ___ Python tracker ___ ___

[issue17258] multiprocessing.connection challenge implicitly uses MD5

miss-islington added the comment: New changeset 196810a98ab07714df3d329c325575ac36e61318 by Miss Islington (bot) in branch '3.9': bpo-17258: Add requires_hashdigest to multiprocessing tests (GH-20412) https://github.com/python/cpython/commit/196810a98ab07714df3d329c325575ac36e61318

[issue17258] multiprocessing.connection challenge implicitly uses MD5

Change by miss-islington : -- pull_requests: +19851 pull_request: https://github.com/python/cpython/pull/20626 ___ Python tracker ___

[issue40857] tempfile.TemporaryDirectory() context manager can fail to propagate exceptions generated within its context

Rémi Lapeyre added the comment: This was fixed in e9b51c0ad81da1da11ae65840ac8b50a8521373c so it's fixed in Python 3.8 and 3.9. Maybe it should have been backported to 3.7. Python 3.6 only receives security fixes now. See also bpo-26660 and bpo-35144. -- components: +Library (Lib)

[issue40864] spec_set/autospec/spec seems to not be reading attributes defined in class body

New submission from Evan Fagerberg : Hello, I really like that this library allows for really strict mocking however one thing I have noticed is that it seems like using spec on a mock does not properly read the class body for attributes like some of the documentation claims. For example

[issue40863] bytes.decode changes/destroys line endings on windows

Steven D'Aprano added the comment: You don't need `b = bytes([0x41, 0x0D, 0x0A])`, this will work just as well: b = b'\x41\x0D\x0A' and this is even better: b = b'A\r\n' > It seems like bytes.decode always replaces "\n" with "\r\n". What makes you say that? You are passing the

[issue40867] Remove unused include in Module/_randommodule.c

Change by Erlend Egeberg Aasland : -- keywords: +patch pull_requests: +19855 stage: -> patch review pull_request: https://github.com/python/cpython/pull/20635 ___ Python tracker

[issue40521] [subinterpreters] Make free lists and unicode caches per-interpreter

STINNER Victor added the comment: New changeset 69ac6e58fd98de339c013fe64cd1cf763e4f9bca by Victor Stinner in branch 'master': bpo-40521: Make tuple free list per-interpreter (GH-20247) https://github.com/python/cpython/commit/69ac6e58fd98de339c013fe64cd1cf763e4f9bca --

[issue40807] Codeop: Show warnings once during _maybe_compile

Terry J. Reedy added the comment: The purpose of code and codeop is to imitate interactive python in python. Calling compile() thrice is an implementation detail. Exposing that is a bug. On #37824, Serhiy reported 3 'IDLE' bugs for 3.7 to 3.9: tripled DeprecationWarning (now joined by

[issue40865] Remove unused insint() macro in SHA512 module

Erlend Egeberg Aasland added the comment: Indeed. I'll update the PR and the title of this issue. -- ___ Python tracker ___ ___

[issue40865] Remove unused insint() macro from the hash modules

Change by Erlend Egeberg Aasland : -- title: Remove unused insint() macro in SHA512 module -> Remove unused insint() macro from the hash modules ___ Python tracker ___

[issue40826] PyOS_InterruptOccurred() now requires to hold the GIL: PyOS_Readline() crash

Eryk Sun added the comment: > The purpose of _Py_BEGIN_SUPPRESS_IPH is to suppress such popup, no? That macro suppresses the CRT's invalid parameter handler, which by default terminates abnormally with a fastfail (status code 0xC409), even in a release build. In a debug build, we

[issue40807] Codeop: Show warnings once during _maybe_compile

Terry J. Reedy added the comment: I am not sure if 3 compiles are always needed. At one time, compile required a final '\n' to not raise, but that is no longer true. And as near as I can tell, 'code without final newline\n' and 'code without final newline\n\n' always compile the same.

[issue40865] Remove unused insint() macro from the hash modules

Christian Heimes added the comment: Thanks for the cleanup! The 3.9 backport will land any minute. -- resolution: -> fixed stage: patch review -> resolved status: open -> closed ___ Python tracker

[issue40865] Remove unused insint() macro from the hash modules

Change by miss-islington : -- pull_requests: +19853 pull_request: https://github.com/python/cpython/pull/20633 ___ Python tracker ___

[issue40865] Remove unused insint() macro from the hash modules

miss-islington added the comment: New changeset 6ed578f6dbffdec94f62cc2e36d626fc195678d7 by Erlend Egeberg Aasland in branch 'master': bpo-40865: Remove unused insint() macro from hash modules (GH-20627) https://github.com/python/cpython/commit/6ed578f6dbffdec94f62cc2e36d626fc195678d7

[issue40866] Use PyModule_AddType() in posix module initialisation

New submission from Erlend Egeberg Aasland : Use PyModule_AddType() iso. PyModule_AddObject() in posix module initialisation, and use PyTypeObject iso. PyObject for type objects. -- components: Library (Lib) messages: 370731 nosy: erlendaasland priority: normal severity: normal

[issue40863] bytes.decode changes/destroys line endings on windows

Matthias Naegler added the comment: I forgot something important. Using open with 'ab' works. >>> ...above code... with open("./test_binary.txt", "ab") as myfile: ... myfile.write(s.encode('utf-8')) -- ___ Python tracker

[issue40865] Remove unused insint() macro from the hash modules

miss-islington added the comment: New changeset e4e5ec18e2f79f9bba439b531ebf89ac148deda0 by Miss Islington (bot) in branch '3.9': bpo-40865: Remove unused insint() macro from hash modules (GH-20627) https://github.com/python/cpython/commit/e4e5ec18e2f79f9bba439b531ebf89ac148deda0

[issue40866] Use PyModule_AddType() in posix module initialisation

Change by Erlend Egeberg Aasland : -- keywords: +patch pull_requests: +19854 stage: -> patch review pull_request: https://github.com/python/cpython/pull/20634 ___ Python tracker

[issue40863] bytes.decode changes/destroys line endings on windows

Matthias Naegler added the comment: Thanks Steven for your fast response. > The best way to see what your string actually contains is the print the repr: You are right. bytes.decode is correct. Im not a python expert, so thanks for the note about "repr". With repr(...) everything looks fine.

[issue40865] Remove unused insint() macro in SHA512 module

New submission from Erlend Egeberg Aasland : The insint() macro in line 741 is unused and can be removed. -- components: Library (Lib) messages: 370723 nosy: erlendaasland priority: normal severity: normal status: open title: Remove unused insint() macro in SHA512 module versions:

[issue40865] Remove unused insint() macro in SHA512 module

Change by Erlend Egeberg Aasland : -- keywords: +patch pull_requests: +19852 stage: -> patch review pull_request: https://github.com/python/cpython/pull/20627 ___ Python tracker

[issue39573] [C API] Make PyObject an opaque structure in the limited C API

STINNER Victor added the comment: New changeset dc24b8a2ac32114313bae519db3ccc21fe45c982 by Victor Stinner in branch 'master': bpo-39573: Porting to Python 3.10: Py_SET_SIZE() macro (GH-20610) https://github.com/python/cpython/commit/dc24b8a2ac32114313bae519db3ccc21fe45c982 --

[issue40863] bytes.decode changes/destroys line endings on windows

Paul Moore added the comment: Because if you open a file in text mode (without "b" in the mode), Python writes \n (newline) characters as \r\n (carriage return, line feed) which are the Windows textfile representation of "Newline". >From the documentation of the built in open() function,

[issue40863] bytes.decode changes/destroys line endings on windows

Change by Eryk Sun : -- stage: -> resolved status: open -> closed ___ Python tracker ___ ___ Python-bugs-list mailing list

[issue40865] Remove unused insint() macro in SHA512 module

Christian Heimes added the comment: The other builtin hashes have the same unused macro. $ grep -R insint Modules/*.c Modules/md5module.c:#define insint(n,v) { PyModule_AddIntConstant(m,n,v); } Modules/sha1module.c:#define insint(n,v) { PyModule_AddIntConstant(m,n,v); }

[issue40867] Remove unused include in Module/_randommodule.c

New submission from Erlend Egeberg Aasland : _Py_bswap32() is no longer used in _randommodule.c (removed in commit 2d87577), so including pycore_byteswap.h is not necessary. -- components: Library (Lib) messages: 370732 nosy: erlendaasland priority: normal severity: normal status:

[issue40521] [subinterpreters] Make free lists and unicode caches per-interpreter

Change by STINNER Victor : -- pull_requests: +19856 pull_request: https://github.com/python/cpython/pull/20636 ___ Python tracker ___

[issue40521] [subinterpreters] Make free lists and unicode caches per-interpreter

STINNER Victor added the comment: New changeset 2ba59370c3dda2ac229c14510e53a05074b133d1 by Victor Stinner in branch 'master': bpo-40521: Make float free list per-interpreter (GH-20636) https://github.com/python/cpython/commit/2ba59370c3dda2ac229c14510e53a05074b133d1 --

[issue40802] AbstractEventLoop.shutdown_default_executor breaks backwards compatibility

Kyle Stanley added the comment: The main issue is that shutdown_default_executor() (or something that does the same thing) is necessary to ensure the resources of the event loop's ThreadPoolExecutor are finalized in a safe manner. Otherwise, it frequently results in dangling threads when

[issue40521] [subinterpreters] Make free lists and unicode caches per-interpreter

STINNER Victor added the comment: New changeset 88ec9190105c9b03f49aaef601ce02b242a75273 by Victor Stinner in branch 'master': bpo-40521: Make list free list per-interpreter (GH-20642) https://github.com/python/cpython/commit/88ec9190105c9b03f49aaef601ce02b242a75273 --

[issue36724] Clear _PyRuntime at exit

STINNER Victor added the comment: > _PyRuntime.warnings is not cleared at Python exit: 3 objects are kept alive > even after Py_Finalize(). This issue has been fixed. Moreover, the GC state is now per-interpreter. -- resolution: -> fixed stage: patch review -> resolved status: open

[issue40807] Codeop: Show warnings once during _maybe_compile

Change by miss-islington : -- pull_requests: +19867 pull_request: https://github.com/python/cpython/pull/20647 ___ Python tracker ___

[issue40807] Codeop: Show warnings once during _maybe_compile

Terry J. Reedy added the comment: New changeset a5d6aba318ead9cc756ba750a70da41f5def3f8f by Terry Jan Reedy in branch '3.8': [3.8] bpo-40807: Backport test_codeop change [GH-19670] https://github.com/python/cpython/commit/a5d6aba318ead9cc756ba750a70da41f5def3f8f --

[issue40521] [subinterpreters] Make free lists and unicode caches per-interpreter

STINNER Victor added the comment: New changeset 3744ed2c9c0b3905947602fc375de49533790cb9 by Victor Stinner in branch 'master': bpo-40521: Make frame free list per-interpreter (GH-20638) https://github.com/python/cpython/commit/3744ed2c9c0b3905947602fc375de49533790cb9 --

[issue40521] [subinterpreters] Make free lists and unicode caches per-interpreter

Change by STINNER Victor : -- pull_requests: +19863 pull_request: https://github.com/python/cpython/pull/20643 ___ Python tracker ___

[issue40521] [subinterpreters] Make free lists and unicode caches per-interpreter

STINNER Victor added the comment: New changeset e005ead49b1ee2b1507ceea94e6f89c28ecf1f81 by Victor Stinner in branch 'master': bpo-40521: Make context free list per-interpreter (GH-20644) https://github.com/python/cpython/commit/e005ead49b1ee2b1507ceea94e6f89c28ecf1f81 --

[issue40807] Codeop: Show warnings once during _maybe_compile

Terry J. Reedy added the comment: This is a spinoff of #37824. My above comments about compile are wrong. The example nests '' within ''. And, with freshly compiled master, I found an example where Deprecation warning, but not SyntaxWarning, accompanies SyntaxError, and which requires

[issue40807] Codeop: Show warnings once during _maybe_compile

Terry J. Reedy added the comment: New changeset 12d3061c7819a73d891dcce44327410eaf0e1bc2 by Miss Islington (bot) in branch '3.7': [3.8] bpo-40807: Backport test_codeop change [GH-19670] https://github.com/python/cpython/commit/12d3061c7819a73d891dcce44327410eaf0e1bc2 --

[issue40521] [subinterpreters] Make free lists and unicode caches per-interpreter

STINNER Victor added the comment: New changeset 7daba6f221e713f7f60c613b246459b07d179f91 by Victor Stinner in branch 'master': bpo-40521: Make slice cache per-interpreter (GH-20637) https://github.com/python/cpython/commit/7daba6f221e713f7f60c613b246459b07d179f91 --

[issue40514] [subinterpreters] Add --experimental-isolated-subinterpreters build option

STINNER Victor added the comment: I started to *remove* a few"#ifdef EXPERIMENTAL_ISOLATED_SUBINTERPRETERS": * commit 88ec9190105c9b03f49aaef601ce02b242a75273: Make list free list per-interpreter * commit 3744ed2c9c0b3905947602fc375de49533790cb9: Make frame free list per-interpreter * PR

[issue40514] [subinterpreters] Add --experimental-isolated-subinterpreters build option

Change by STINNER Victor : -- versions: +Python 3.10 -Python 3.9 ___ Python tracker ___ ___ Python-bugs-list mailing list

[issue40868] io.TextIOBase.buffer is not necessarily a buffer

New submission from Manuel Jacob : https://docs.python.org/dev/library/io.html#io.TextIOBase.buffer says: "The underlying binary buffer (a BufferedIOBase instance) that TextIOBase deals with. This is not part of the TextIOBase API and may not exist in some implementations." It is not

[issue39325] Original window focus when opening IDLE by double clicking Python file Mac

Terry J. Reedy added the comment: Irv also posted to idledev, where he explained his particular need: He teaches a class with assignment to write a .py file with certain behavior. Submissions are in 1 directory. He would like to double-click to load a file and F5 to run it. Currently, he

[issue40807] Codeop: Show warnings once during _maybe_compile

Terry J. Reedy added the comment: New changeset 052d3fc0907be253cfd64b2c737a0b0aca586011 by Cheryl Sabella in branch 'master': bpo-40807: Show warnings once from codeop._maybe_compile (#20486) https://github.com/python/cpython/commit/052d3fc0907be253cfd64b2c737a0b0aca586011 --

[issue40807] Codeop: Show warnings once during _maybe_compile

Change by miss-islington : -- pull_requests: +19860 pull_request: https://github.com/python/cpython/pull/20640 ___ Python tracker ___

[issue40807] Codeop: Show warnings once during _maybe_compile

Change by miss-islington : -- nosy: +miss-islington nosy_count: 3.0 -> 4.0 pull_requests: +19859 pull_request: https://github.com/python/cpython/pull/20639 ___ Python tracker

[issue40807] Codeop: Show warnings once during _maybe_compile

Change by miss-islington : -- pull_requests: +19861 pull_request: https://github.com/python/cpython/pull/20641 ___ Python tracker ___

[issue40521] [subinterpreters] Make free lists and unicode caches per-interpreter

Change by STINNER Victor : -- pull_requests: +19862 pull_request: https://github.com/python/cpython/pull/20642 ___ Python tracker ___

[issue40807] Codeop: Show warnings once during _maybe_compile

miss-islington added the comment: New changeset 3dfe549bfa08a099b88efd030669359199f31064 by Miss Islington (bot) in branch '3.9': bpo-40807: Show warnings once from codeop._maybe_compile (GH-20486) https://github.com/python/cpython/commit/3dfe549bfa08a099b88efd030669359199f31064 --

[issue40521] [subinterpreters] Make free lists and unicode caches per-interpreter

Change by STINNER Victor : -- pull_requests: +19864 pull_request: https://github.com/python/cpython/pull/20644 ___ Python tracker ___

[issue40521] [subinterpreters] Make free lists and unicode caches per-interpreter

Change by STINNER Victor : -- pull_requests: +19865 pull_request: https://github.com/python/cpython/pull/20645 ___ Python tracker ___

[issue40521] [subinterpreters] Make free lists and unicode caches per-interpreter

Change by STINNER Victor : -- pull_requests: +19858 pull_request: https://github.com/python/cpython/pull/20638 ___ Python tracker ___

[issue38160] [subinterpreters] Add a "PyInterpreterState *" field to PyTypeObject.

STINNER Victor added the comment: Which kind of code pattern is supposed to be more efficient if it would be possible to get a "PyInterpreterState *" from a "PyTypeObject*"? I don't understand which problem this issue is supposed to solve. -- ___

[issue40521] [subinterpreters] Make free lists and unicode caches per-interpreter

Change by STINNER Victor : -- pull_requests: +19857 pull_request: https://github.com/python/cpython/pull/20637 ___ Python tracker ___

[issue40521] [subinterpreters] Make free lists and unicode caches per-interpreter

STINNER Victor added the comment: New changeset 78a02c2568714562e23e885b6dc5730601f35226 by Victor Stinner in branch 'master': bpo-40521: Make async gen free lists per-interpreter (GH-20643) https://github.com/python/cpython/commit/78a02c2568714562e23e885b6dc5730601f35226 --

[issue40807] Codeop: Show warnings once during _maybe_compile

Change by Terry J. Reedy : -- pull_requests: +19866 pull_request: https://github.com/python/cpython/pull/20646 ___ Python tracker ___

[issue40869] errno missing descriptions

New submission from YoSTEALTH : `errno` https://docs.python.org/3/library/errno.html is missing description for symbols like `ECANCELED` https://www.gnu.org/software/libc/manual/html_node/Error-Codes.html There might be others missing description as well, i haven't investigated further,

[issue40837] email.utils.encode_rfc2231(string, None, None) returns broken value

Change by Karthikeyan Singaravelan : -- components: +email nosy: +barry, maxking, r.david.murray ___ Python tracker ___ ___

[issue40864] spec_set/autospec/spec seems to not be reading attributes defined in class body

Karthikeyan Singaravelan added the comment: mock uses dir to iterate through the attributes that needs to be specced [0]. Unless the variable is initialized it's not listed in dir. Below is an example where age is initialized and name is not. name is not present in dir(Person) and hence

[issue40679] show class name in method invocation TypeError

STINNER Victor added the comment: New changeset 232dda6cbc10860328a83517a6e3ea238ff4147f by Victor Stinner in branch 'master': bpo-40679: Fix _PyEval_EvalCode() crash if qualname is NULL (GH-20615) https://github.com/python/cpython/commit/232dda6cbc10860328a83517a6e3ea238ff4147f --

[issue40679] show class name in method invocation TypeError

STINNER Victor added the comment: > Thanks a lot, Victor. You're welcome. Using the qualified name instead of the "short" name in error messages is nice enhancement! -- resolution: -> fixed stage: patch review -> resolved status: open -> closed

[issue40860] Exception in multiprocessing/context.py under load

Arkady added the comment: The problem is likely in the call to multiprocessing.Process.join() with timeout. If I use timeout=None the code works. -- ___ Python tracker ___

[issue25377] Mention octal format of mode argument of os.chmod

Ama Aje My Fren added the comment: I think I was finally able to do a PR for this on 3.10. Requesting a triager to please check because it does not appear as a Documentation change on Github. Thanks -- ___ Python tracker

[issue31254] WeakKeyDictionary/Mapping doesn't call __missing__

Nathaniel Smith added the comment: Just found this while searching to see if we had an existing way to combine WeakValueDictionary + defaultdict. The use case I've run into a few times is where you need a keyed collection of mutexes, like e.g. asyncio.Lock objects. You want to make sure

[issue40826] PyOS_InterruptOccurred() now requires to hold the GIL: PyOS_Readline() crash

Eryk Sun added the comment: > Perhaps the child interpreter used by test_repl doesn't inherit > that behavior. The OS error mode is inherited, unless the child is created with CREATE_DEFAULT_ERROR_MODE flag. But the CRT error mode isn't inherited. (For the spawn* family, in principle the

[issue40861] On Windows, liblzma is always built without optimization

New submission from Nikita Nemkin : Windows build system always builds liblzma with optimizations disabled (/Od), even in Release configuration. Compared to optimized build (/O2), compression speed is 2-2.5x slower and module size is 30% larger. -- components: Build messages: 370702

[issue40856] IDLE line numbering should be light gray

Terry J. Reedy added the comment: When the question of changing default-x.def has come up before (and there are things I would like to change) the answer has generally been that we should not, a) because it would be too disruptive, including affecting exiting customizations, and b) because

[issue40855] statistics.stdev ignore xbar argument

Matti added the comment: If we estimate the mean using a sample we loose one degree of freedom so it will be divided by N-1, while if we have the mean independent of the sample it should be divided by N to be unbiased. i.e. example 1 sqrt(((1-1.5)²+(2-1.5)²)/(2-1)) = 0.7... example 3

[issue40860] Exception in multiprocessing/context.py under load

larytet added the comment: This code reproduces the problem https://github.com/larytet-py/multiprocess I assume that my use of join() is not correct. -- ___ Python tracker

[issue40862] argparse.BooleanOptionalAction accept and silently its the const argument

New submission from Rémi Lapeyre : The action is used to store None, True or False when an argument like --foo or --no-foo is given to the cli so it has no used for this action, but it is accepted without warning: Python 3.10.0a0 (heads/bpo-wip:6e23a9c82b, Jun 4 2020, 13:41:35) [Clang

[issue1294959] Add sys.platlibdir and configure --with-platlibdir to use /usr/lib64 on Fedora and SuSE

Miro Hrončok added the comment: Note: https://docs.python.org/3.9/install/index.html probably needs some update wrt this change. -- ___ Python tracker ___

[issue40862] argparse.BooleanOptionalAction accept and silently discard its the const argument

Change by Rémi Lapeyre : -- title: argparse.BooleanOptionalAction accept and silently its the const argument -> argparse.BooleanOptionalAction accept and silently discard its the const argument ___ Python tracker

[issue17258] multiprocessing.connection challenge implicitly uses MD5

miss-islington added the comment: New changeset b022e5cffbd3ff51ae361cf80f2a3b660be8b1ee by Christian Heimes in branch 'master': bpo-17258: Add requires_hashdigest to multiprocessing tests (GH-20412) https://github.com/python/cpython/commit/b022e5cffbd3ff51ae361cf80f2a3b660be8b1ee

[issue40862] argparse.BooleanOptionalAction accept and silently its the const argument

Change by Rémi Lapeyre : -- keywords: +patch pull_requests: +19848 stage: -> patch review pull_request: https://github.com/python/cpython/pull/20623 ___ Python tracker ___