Mailman 3 April 2021 - Python-Dev

Summary of Python tracker Issues
by Python tracker 16 Apr '21

16 Apr '21

ACTIVITY SUMMARY (2021-04-09 - 2021-04-16) Python tracker at https://bugs.python.org/ To view or respond to any of the issues listed below, click on the issue. Do NOT respond to this message. Issues counts and deltas: open 7522 ( -2) closed 48086 (+78) total 55608 (+76) Open issues with patches: 2993 Issues opened (52) ================== #37251: Mocking a MagicMock with a function spec results in an AsyncMo https://bugs.python.org/issue37251 reopened by gregory.p.smith #40066: Enum: modify __repr__, __str__; update docs https://bugs.python.org/issue40066 reopened by ethan.furman #40432: Pegen regenerate project for Windows not working https://bugs.python.org/issue40432 reopened by pablogsal #43764: Turning off generation of __match_args__ for dataclasses https://bugs.python.org/issue43764 reopened by eric.smith #43802: Seg fault on macOS using multiprocessing.JoinableQueue https://bugs.python.org/issue43802 opened by jacobtylerwalls #43803: ctypes string_at/wstring_at - bad argument name used in docs a https://bugs.python.org/issue43803 opened by talhayon1 #43804: Add more info about building C/C++ Extensions on Windows using https://bugs.python.org/issue43804 opened by shreyanavigyan #43805: multiprocessing.Queue hangs when process on other side dies https://bugs.python.org/issue43805 opened by kormang #43806: asyncio.StreamReader hangs when reading from pipe and other pr https://bugs.python.org/issue43806 opened by kormang #43807: JSONDecodeError: Extra Data Raised on Long Valid JSON https://bugs.python.org/issue43807 opened by UnknownRetiredGuy #43811: Run GHA CI with multiple OpenSSL versions https://bugs.python.org/issue43811 opened by christian.heimes #43813: Denial of service on http.server module with large request met https://bugs.python.org/issue43813 opened by demonia #43814: Fix the error message for disallowed __weakref__ slots https://bugs.python.org/issue43814 opened by maggyero #43815: documentation for types.new_class() mention misleading default https://bugs.python.org/issue43815 opened by eric.smith #43817: Add typing.get_annotations() https://bugs.python.org/issue43817 opened by larry #43818: Email does not apply policy to multipart messages with defects https://bugs.python.org/issue43818 opened by jishac #43819: ExtensionFileLoader Does Not Implement invalidate_caches https://bugs.python.org/issue43819 opened by Ian.H #43821: Undocumented behavior of sleep functions and asyncio delayed e https://bugs.python.org/issue43821 opened by Josef Havr��nek #43826: Resource warnings in test_subprocess https://bugs.python.org/issue43826 opened by xtreak #43827: abc conflicts with __init_subclass__ https://bugs.python.org/issue43827 opened by vladhoi #43829: MappingProxyType cannot hash a hashable underlying mapping https://bugs.python.org/issue43829 opened by andymaier #43832: asyncio + multiprocessing = core dump in sem_trywait https://bugs.python.org/issue43832 opened by Andrei Pozolotin #43833: Unexpected Parsing of Numeric Literals Concatenated with Boole https://bugs.python.org/issue43833 opened by sco1 #43834: Use context manager in StringIO example https://bugs.python.org/issue43834 opened by John Hagen #43835: Dataclasses don't call base class __init__ https://bugs.python.org/issue43835 opened by Paul Pinterits #43837: Operator precedence documentation could be more clear https://bugs.python.org/issue43837 opened by aidan.feldman #43838: There is a way to access an underlying mapping in MappingProxy https://bugs.python.org/issue43838 opened by serhiy.storchaka #43843: libregrtest: mark a test as failed if a thread logs an unexpec https://bugs.python.org/issue43843 opened by vstinner #43845: test_concurrent_futures leaks many dangling threads on FreeBSD https://bugs.python.org/issue43845 opened by vstinner #43846: Control stack usage in large expressions https://bugs.python.org/issue43846 opened by Mark.Shannon #43847: [Windows] ntpath.realpath() of bytes root directory may raise https://bugs.python.org/issue43847 opened by 9001 #43848: gzip.py: explain optional argument mtime https://bugs.python.org/issue43848 opened by jwuttke #43851: Optimise SQLite builds on macOS and Windows https://bugs.python.org/issue43851 opened by erlendaasland #43852: [sqlite3] Improve tuple creation https://bugs.python.org/issue43852 opened by erlendaasland #43853: [sqlite3] Fix sqlite3_value_text() usage https://bugs.python.org/issue43853 opened by erlendaasland #43854: curses: returns incorrect chars after resuming a suspended pro https://bugs.python.org/issue43854 opened by darrikonn #43855: test_ssl: test_msg_callback_deadlock_bpo43577() failed on macO https://bugs.python.org/issue43855 opened by vstinner #43856: Docs for importlib.metadata should mention Python version https://bugs.python.org/issue43856 opened by pitrou #43857: Fix the AttributeError message for deletion of a missing attri https://bugs.python.org/issue43857 opened by maggyero #43858: Provide method to get list of logging level names https://bugs.python.org/issue43858 opened by andylowry #43859: Improve syntax error for indentation errors https://bugs.python.org/issue43859 opened by pablogsal #43860: unittest increment tests executed https://bugs.python.org/issue43860 opened by chelmite #43861: A potential double free in list_sort_impl https://bugs.python.org/issue43861 opened by Yunlongs #43862: warnings: -W option and PYTHONWARNINGS now use the message as https://bugs.python.org/issue43862 opened by vstinner #43863: [Windows] test_distutils logs: DeprecationWarning: bdist_msi https://bugs.python.org/issue43863 opened by vstinner #43864: [Windows] test_importlib logs: DeprecationWarning: WindowsRegi https://bugs.python.org/issue43864 opened by vstinner #43866: Installation files of the Python https://bugs.python.org/issue43866 opened by xmha97 #43868: Remove PyOS_ReadlineFunctionPointer from the stable ABI list https://bugs.python.org/issue43868 opened by petr.viktorin #43869: Fix documentation of epoch/time.time https://bugs.python.org/issue43869 opened by Ofekmeister #43870: C API Functions Bypass __import__ Override https://bugs.python.org/issue43870 opened by Ian.H #43871: urllib.parse.urlparse doesn't check port https://bugs.python.org/issue43871 opened by palik #43872: Unable to compile cpython in Visual Studio 2019 https://bugs.python.org/issue43872 opened by shreyanavigyan Most recent 15 issues with no replies (15) ========================================== #43870: C API Functions Bypass __import__ Override https://bugs.python.org/issue43870 #43864: [Windows] test_importlib logs: DeprecationWarning: WindowsRegi https://bugs.python.org/issue43864 #43863: [Windows] test_distutils logs: DeprecationWarning: bdist_msi https://bugs.python.org/issue43863 #43861: A potential double free in list_sort_impl https://bugs.python.org/issue43861 #43859: Improve syntax error for indentation errors https://bugs.python.org/issue43859 #43857: Fix the AttributeError message for deletion of a missing attri https://bugs.python.org/issue43857 #43855: test_ssl: test_msg_callback_deadlock_bpo43577() failed on macO https://bugs.python.org/issue43855 #43853: [sqlite3] Fix sqlite3_value_text() usage https://bugs.python.org/issue43853 #43848: gzip.py: explain optional argument mtime https://bugs.python.org/issue43848 #43838: There is a way to access an underlying mapping in MappingProxy https://bugs.python.org/issue43838 #43834: Use context manager in StringIO example https://bugs.python.org/issue43834 #43832: asyncio + multiprocessing = core dump in sem_trywait https://bugs.python.org/issue43832 #43827: abc conflicts with __init_subclass__ https://bugs.python.org/issue43827 #43826: Resource warnings in test_subprocess https://bugs.python.org/issue43826 #43821: Undocumented behavior of sleep functions and asyncio delayed e https://bugs.python.org/issue43821 Most recent 15 issues waiting for review (15) ============================================= #43868: Remove PyOS_ReadlineFunctionPointer from the stable ABI list https://bugs.python.org/issue43868 #43862: warnings: -W option and PYTHONWARNINGS now use the message as https://bugs.python.org/issue43862 #43859: Improve syntax error for indentation errors https://bugs.python.org/issue43859 #43857: Fix the AttributeError message for deletion of a missing attri https://bugs.python.org/issue43857 #43856: Docs for importlib.metadata should mention Python version https://bugs.python.org/issue43856 #43853: [sqlite3] Fix sqlite3_value_text() usage https://bugs.python.org/issue43853 #43852: [sqlite3] Improve tuple creation https://bugs.python.org/issue43852 #43851: Optimise SQLite builds on macOS and Windows https://bugs.python.org/issue43851 #43848: gzip.py: explain optional argument mtime https://bugs.python.org/issue43848 #43846: Control stack usage in large expressions https://bugs.python.org/issue43846 #43843: libregrtest: mark a test as failed if a thread logs an unexpec https://bugs.python.org/issue43843 #43834: Use context manager in StringIO example https://bugs.python.org/issue43834 #43827: abc conflicts with __init_subclass__ https://bugs.python.org/issue43827 #43826: Resource warnings in test_subprocess https://bugs.python.org/issue43826 #43814: Fix the error message for disallowed __weakref__ slots https://bugs.python.org/issue43814 Top 10 most discussed issues (10) ================================= #43833: Unexpected Parsing of Numeric Literals Concatenated with Boole https://bugs.python.org/issue43833 15 msgs #43725: Create a release branch ABI stability regression test https://bugs.python.org/issue43725 12 msgs #14243: tempfile.NamedTemporaryFile not particularly useful on Windows https://bugs.python.org/issue14243 11 msgs #43817: Add typing.get_annotations() https://bugs.python.org/issue43817 10 msgs #42238: Deprecate suspicious.py? https://bugs.python.org/issue42238 9 msgs #43803: ctypes string_at/wstring_at - bad argument name used in docs a https://bugs.python.org/issue43803 9 msgs #40137: TODO list when PEP 573 "Module State Access from C Extension M https://bugs.python.org/issue40137 8 msgs #43761: Documenting dataclass and namedtuple changes for structural pa https://bugs.python.org/issue43761 8 msgs #43835: Dataclasses don't call base class __init__ https://bugs.python.org/issue43835 8 msgs #35056: Test leaks of memory not managed by Python allocator https://bugs.python.org/issue35056 6 msgs Issues closed (78) ================== #5758: fileinput.hook_compressed returning bytes from gz file https://bugs.python.org/issue5758 closed by methane #14811: decoding_fgets() truncates long lines and fails with a SyntaxE https://bugs.python.org/issue14811 closed by vstinner #20364: Rename & explain sqlite3.Cursor.execute 'parameters' param https://bugs.python.org/issue20364 closed by berker.peksag #32466: Fix missing test coverage for fractions.Fraction.__new__ https://bugs.python.org/issue32466 closed by mark.dickinson #34311: locale.format() and locale.format_string() cast Decimals to fl https://bugs.python.org/issue34311 closed by matrixise #37741: importlib.metadata docs not showing up in the module index https://bugs.python.org/issue37741 closed by barry #38530: Offer suggestions on AttributeError and NameError https://bugs.python.org/issue38530 closed by pablogsal #39774: Missing documentation on how to make package executable as scr https://bugs.python.org/issue39774 closed by igo95862 #40006: enum: Add documentation for _create_pseudo_member_ and composi https://bugs.python.org/issue40006 closed by ethan.furman #40107: pathlib: make `_Accessor.open()` return a file object and not https://bugs.python.org/issue40107 closed by steve.dower #40701: tempfile mixes str and bytes in an inconsistent manner https://bugs.python.org/issue40701 closed by lukasz.langa #40804: Bug report in python3.6.8 using argparse module https://bugs.python.org/issue40804 closed by iritkatriel #41098: Deprecating PyUnicodeEncodeError_Create https://bugs.python.org/issue41098 closed by methane #41163: test_weakref hangs https://bugs.python.org/issue41163 closed by iritkatriel #41506: Inclusion or documentation of extended with syntax in 3.9 https://bugs.python.org/issue41506 closed by pablogsal #41515: typing.get_type_hints generates KeyError https://bugs.python.org/issue41515 closed by vstinner #41661: os.path.relpath does not document ValueError on Windows with d https://bugs.python.org/issue41661 closed by steve.dower #42124: FAIL test_asyncio test_sendfile_close_peer_in_the_middle_of_re https://bugs.python.org/issue42124 closed by vstinner #42152: Use PyDict_Contains() and PyDict_SetDefault() instead of PyDic https://bugs.python.org/issue42152 closed by serhiy.storchaka #42248: Raised exception in Enum keeping user objects alive unnecessar https://bugs.python.org/issue42248 closed by ethan.furman #42578: Add tip when encountering UnicodeDecode/EncodeError in open() https://bugs.python.org/issue42578 closed by methane #42904: get_type_hints does not provide localns for classes https://bugs.python.org/issue42904 closed by kj #43080: pprint for dataclass instances https://bugs.python.org/issue43080 closed by eric.smith #43105: [Windows] Can't import extension modules resolved via relative https://bugs.python.org/issue43105 closed by steve.dower #43265: [sqlite3] Improve backup error handling https://bugs.python.org/issue43265 closed by berker.peksag #43296: [sqlite3] Fix sqlite3_value_blob() usage https://bugs.python.org/issue43296 closed by berker.peksag #43351: `RecursionError` during deallocation https://bugs.python.org/issue43351 closed by andrewvaughanj #43426: test_importlib.test_windows emits deprecation warning over usa https://bugs.python.org/issue43426 closed by vstinner #43505: [sqlite3] Explicitly initialise and shut down sqlite3 https://bugs.python.org/issue43505 closed by berker.peksag #43539: test_asyncio: test_sendfile_close_peer_in_the_middle_of_receiv https://bugs.python.org/issue43539 closed by vstinner #43680: Deprecate undocumented io.OpenWrapper and _pyio.OpenWrapper https://bugs.python.org/issue43680 closed by vstinner #43682: Make static methods created by @staticmethod callable https://bugs.python.org/issue43682 closed by vstinner #43686: re.match appears to hang with certain combinations of pattern https://bugs.python.org/issue43686 closed by gregory.p.smith #43708: Tkinter theme settings object schema is missing https://bugs.python.org/issue43708 closed by terry.reedy #43712: PEP 597: fileinput uses locale encoding https://bugs.python.org/issue43712 closed by methane #43723: Deprecate camelCase aliases from threading.py https://bugs.python.org/issue43723 closed by vstinner #43731: PEP 597: logging.fileConfig() uses locale encoding. https://bugs.python.org/issue43731 closed by methane #43739: Fixing the example code in Doc/extending/extending.rst to decl https://bugs.python.org/issue43739 closed by rhettinger #43744: enum: Adding a member named _classname__ raises IndexError https://bugs.python.org/issue43744 closed by ethan.furman #43751: await anext() returns None when default is given https://bugs.python.org/issue43751 closed by pablogsal #43752: [sqlite3] Fetching an empty value from date column raises Valu https://bugs.python.org/issue43752 closed by berker.peksag #43772: Minor repr error in typing.TypeVar.__ror__() https://bugs.python.org/issue43772 closed by gvanrossum #43777: Remove description of "pip search" command from tutorial https://bugs.python.org/issue43777 closed by methane #43782: Failure to build from source on ppc64le on ubuntu xenial https://bugs.python.org/issue43782 closed by zach.ware #43783: Make ParamSpec.args/kwargs more useful objects https://bugs.python.org/issue43783 closed by gvanrossum #43785: Remove RLock from BZ2File https://bugs.python.org/issue43785 closed by methane #43787: Optimize BZ2File, GzipFile, and LZMAFile __iter__ method. https://bugs.python.org/issue43787 closed by methane #43790: CLA check fails with a 500 error https://bugs.python.org/issue43790 closed by Mariatta #43792: Cython is not compatible with Python 3.10 yet: "SystemError: b https://bugs.python.org/issue43792 closed by vstinner #43796: "install" package on PyPI https://bugs.python.org/issue43796 closed by terry.reedy #43797: Improve syntax error for invalid comparisons https://bugs.python.org/issue43797 closed by pablogsal #43798: Add position metadata to alias AST type https://bugs.python.org/issue43798 closed by lys.nikolaou #43799: OpenSSL 3.0.0: define OPENSSL_API_COMPAT 1.1.1 https://bugs.python.org/issue43799 closed by christian.heimes #43800: os.fwalk is not exposed on macOS https://bugs.python.org/issue43800 closed by ned.deily #43801: Carriage Return problem in version v3.9.0:9cf6752 https://bugs.python.org/issue43801 closed by steven.daprano #43808: Add .isfloat() method to str https://bugs.python.org/issue43808 closed by gregory.p.smith #43809: Improve mismatching parentheses error https://bugs.python.org/issue43809 closed by pablogsal #43810: os.path.abspath returns invalid path (resolves symbolic link) https://bugs.python.org/issue43810 closed by eryksun #43812: Tuple unpacking in getitem https://bugs.python.org/issue43812 closed by pablogsal #43816: Missing 'extern "C"' for _Py_ctype_table https://bugs.python.org/issue43816 closed by vstinner #43820: Remove namespace copy from dataclasses.make_dataclass() https://bugs.python.org/issue43820 closed by eric.smith #43822: Improve syntax errors for missing commas https://bugs.python.org/issue43822 closed by pablogsal #43823: Improve syntax errors for invalid dictionary literals https://bugs.python.org/issue43823 closed by pablogsal #43824: array.array.__deepcopy__() accepts a parameter of any type https://bugs.python.org/issue43824 closed by serhiy.storchaka #43825: Deprecation warnings in test_cmd_line and test_collections https://bugs.python.org/issue43825 closed by vstinner #43828: MappingProxyType accepts string https://bugs.python.org/issue43828 closed by rhettinger #43830: (-1) ** 0.5 returns (6.123233995736766e-17+1j) instead of 1j https://bugs.python.org/issue43830 closed by serhiy.storchaka #43831: sqlite: convert_timestamp raises ValueError for empty columns https://bugs.python.org/issue43831 closed by petr.viktorin #43836: range.index doesn't support start/stop optional arguments https://bugs.python.org/issue43836 closed by ammar2 #43839: [easy] test_cmd_line: DeprecationWarning: invalid escape seque https://bugs.python.org/issue43839 closed by serhiy.storchaka #43840: [easy] test_distutils logs 3 DeprecationWarning warnings https://bugs.python.org/issue43840 closed by vstinner #43841: [easy] test_collections: DeprecationWarning: Please use assert https://bugs.python.org/issue43841 closed by serhiy.storchaka #43842: test_logging: "OSError: [Errno 9] Bad file descriptor" logged https://bugs.python.org/issue43842 closed by vstinner #43844: [easy] test_lib2to3 logs a PendingDeprecationWarning: lib2to3 https://bugs.python.org/issue43844 closed by vstinner #43849: Typo in calendar doc https://bugs.python.org/issue43849 closed by JohnCC330 #43850: unreproducible bytecode: set order depends on random seed for https://bugs.python.org/issue43850 closed by benjamin.peterson #43865: win32_ver and getwindowsversion return different Windows 10 bu https://bugs.python.org/issue43865 closed by eryksun #43867: multiprocess: catch explicitly SystemExit in the Server class https://bugs.python.org/issue43867 closed by vstinner

1 0

Can we change python -W option and PYTHONWARNINGS to use a regex for the message?
by Victor Stinner 16 Apr '21

16 Apr '21

Hi, I propose to change the -W command line option and the PYTHONWARNINGS environment variable to use the message as a regular expression in Python 3.10. Or does anyone have a reason to keep the current behavior as it is? I created https://bugs.python.org/issue43862 for this change. -- Python provides two ways to specify warnings filters: * -W command line option: can be used multiple times * PYTHONWARNINGS environment variable: can contain multiple options separated by commas While the Python API warnings.filterwarnings(action, message="", ...) uses the message as a regular expression, -W and PYTHONWARNINGS require to match *exactly* the *whole* message. For example, if you only want to ignore the new distutils deprecation warning, you must write exactly: $ ./python -X dev -W 'ignore:The distutils package is deprecated and slated for removal in Python 3.12. Use setuptools or check PEP 632 for potential alternatives:DeprecationWarning' -c 'import distutils' I use -X dev to show DeprecationWarning, or you can also use -Wdefault if you prefer. If the deprecation warning changes in Python or if you have a single typo, the warning is not ignored. Example with a typo ("3.13" rather than "3.12"): $ ./python -X dev -W 'ignore:The distutils package is deprecated and slated for removal in Python 3.13. Use setuptools or check PEP 632 for potential alternatives:DeprecationWarning' -c 'import distutils' <string>:1: DeprecationWarning: The distutils package is deprecated and slated for removal in Python 3.12. Use setuptools or check PEP 632 for potential alternatives The PYTHONWARNINGS has another limitation: you cannot specify a message if it contains a comma (","). Hopefully, Python doesn't raise warnings containing comma, right? Well... Just one example: Lib/distutils/sysconfig.py:554: warnings.warn('SO is deprecated, use EXT_SUFFIX', DeprecationWarning, 2) You cannot only ignore the message: $ PYTHONWARNINGS='ignore:SO is deprecated, use EXT_SUFFIX:DeprecationWarning' ./python -c 'import sys; print(sys.warnoptions); print(len(sys.warnoptions))' Invalid -W option ignored: invalid action: 'use EXT_SUFFIX' ['ignore:SO is deprecated', ' use EXT_SUFFIX:DeprecationWarning'] 2 You can only try to use "module" and "lineno" parameters of a warning filter, which are more fragile and hard to use to use. Victor -- Night gathers, and now my watch begins. It shall not end until my death.

3 3

Re: In support of PEP 649
by Tin Tvrtković 16 Apr '21

16 Apr '21

Guido: > It seems a little disingenuous to claim discussions about annotations don’t > concern you when you’re actively using them (for typing, no less, in the > case of pydantic). And I am sure a project as popular (by their own > description) as pydantic will find a way forward if PEP 649 is rejected, > despite overdramatized claims. I also maintain a library that uses type annotations in a runtime context (cattrs), but in a slightly different way than Pydantic. My project is much less popular than Pydantic (to be expected, since it only deals with de/serialization and the class definition layer is attrs/dataclasses), and the first issue filed on my bug tracker about future annotations was created at the end of 2019, so that's when I started thinking about it. To be honest, supporting stringified annotations was a lot of work to do in an efficient way (and that's the main reason cattrs has two converter classes now), but there was plenty of time and I simply did the work. The new design is even faster than the old one.

1 0

Boundaries between numbers and identifiers
by Serhiy Storchaka 15 Apr '21

15 Apr '21

In Python 2.5 `0or[]` was accepted by the Python parser. It became an error in 2.6 because "0o" became recognizing as an incomplete octal number. `1or[]` still is accepted. On other hand, `1if 2else 3` is accepted despites the fact that "2e" can be recognized as an incomplete floating point number. In this case the tokenizer pushes "e" back and returns "2". Shouldn't it do the same with "0o"? It is possible to make `0or[]` be parseable again. Python implementation is able to tokenize this example: $ echo '0or[]' | ./python -m tokenize 1,0-1,1: NUMBER '0' 1,1-1,3: NAME 'or' 1,3-1,4: OP '[' 1,4-1,5: OP ']' 1,5-1,6: NEWLINE '\n' 2,0-2,0: ENDMARKER '' On other hand, all these examples look weird. There is an assymmetry: `1or 2` is a valid syntax, but `1 or2` is not. It is hard to recognize visually the boundary between a number and the following identifier or keyword, especially if numbers can contain letters ("b", "e", "j", "o", "x") and underscores, and identifiers can contain digits. On both sides of the boundary can be letters, digits, and underscores. I propose to change the Python syntax by adding a requirement that there should be a whitespace or delimiter between a numeric literal and the following keyword.

8 10

gzip.py: allow deterministic compression (without time stamp)
by Joachim Wuttke 15 Apr '21

15 Apr '21

gzip compression, using class GzipFile from gzip.py, by default inserts a timestamp to the compressed stream. If the optional argument `mtime` is absent or None, then the current time is used [1]. This makes outputs non-deterministic, which can badly confuse unsuspecting users: If you run "diff" over two outputs to see whether they are unaffected by changes in your application, then you would not expect that the *.gz binaries differ just because they were created at different times. I'd propose to introduce a new constant `NO_TIMESTAMP` as possible value of `mtime`. Furthermore, if policy about API changes allows, I'd suggest that `NO_TIMESTAMP` become the new default value for `mtime`. How to proceed from here? Is this the kind of proposals that has to go through a PEP? - Joachim [1] https://github.com/python/cpython/blob/6f1e8ccffa5b1272a36a35405d3c4e4bbba0…

9 13

Re: Typing syntax and ecosystem, take 2
by Hugh Fisher 15 Apr '21

15 Apr '21

> Date: Wed, 14 Apr 2021 09:57:49 +1000 > From: Chris Angelico <rosuav(a)gmail.com> > Subject: [Python-Dev] Re: Typing syntax and ecosystem, take 2 > > You're advocating an approach that absolutely mandates running the > type checker, but then caches the results in an executable file (the > JS file built from the TS source). Python already has a caching system > - the .pyc files - so if you're importing the same file more than > once, the cost of parsing type annotations is pretty much equivalent > to the cost of doing it in TypeScript. Thinking about it, if mypy compiled to .pyc for execution by the Python interpreter most of what I'm complaining about would go away. -- cheers, Hugh Fisher

2 1

Typing syntax and ecosystem, take 2
by Hugh Fisher 15 Apr '21

15 Apr '21

In any Python 3.6 or later, type >>> x : float = 1 >>> isinstance(x, float) or replace the second line with >>> type(x) As someone who has programmed in FORTRAN, Pascal, C/C++, Java, and Go this is not at all what I consider reasonable. I do not believe other programmers with experience in typed languages would expect this behaviour either. Having a type checker run before the Python interpreter in our current day continuous build/integration environment adds a second step and therefore the chance for it to be forgotten, for version mismatches, for warning/error reports to be misdirected. From an environmental point of view we're also doing the read source, lexical scan, syntax parse twice rather than once. If typing is the future of Python, and the number of PEPs being worked on suggests that it is, then type checking should be integrated into CPython itself. An alternative is the TypeScript/JavaScript model, where typed Python becomes a distinct programming language that cannot be executed directly by the Python interpreter. -- cheers, Hugh Fisher

7 11

Re: Typing syntax and ecosystem
by Hugh Fisher 14 Apr '21

14 Apr '21

> Date: Wed, 14 Apr 2021 11:24:12 +0200 > From: Antoine Pitrou <antoine(a)python.org> > Subject: [Python-Dev] Re: Typing syntax and ecosystem > > Can you explain why you think C++ typing is based on structural > equivalence? I'd rather not have a detailed discussion of C++ typing on python-dev. My example was Ada style type equivalence by name only, in contrast to Java/C++ style type checking which AFAIK is called "structural equivalence". But if you use another term, it won't matter. My point is that sometimes people have argued that the type checker shouldn't be part of the Python interpreter because we're not sure about the best approach and need to have freedom to experiment. I don't think that is still true. Some type system choices, such as equivalence by name only, have already been effectively ruled out. And there's now so much code already annotated assuming the Java/C++ style that I don't think a different type scheme checking can replace it. We've chosen the type scheme for Python (although there may be some fine details to do) and now we're just working on the implementation aspects. Travelling for a few days, so this is probably a good place to stop on. Thank you everybody for responding. -- cheers, Hugh Fisher

1 0

Re: Typing syntax and ecosystem, take 2
by Hugh Fisher 14 Apr '21

14 Apr '21

> > Date: Tue, 13 Apr 2021 19:55:38 -0400 > From: Ned Batchelder <ned(a)nedbatchelder.com> > > In another message, you alluded to PEP 649 being a solution to run-time > type checking. Maybe I'm misunderstanding something: type annotations > never do type checking at runtime. As I understand it, PEP 649 is about > avoiding the overhead of evaluating the type expression itself, not > about checking the types of values. Yes, and I see that as a problem. Here are two versions of code V1 # Apocalypse may occur if foo is not <complicated-type> foo = "coconut" V2 # Apocalypse may occur if foo is not <complicated-type> foo : <complicated-type> = "coconut" To the Python interpreter both have the same semantics and runtime behaviour. But V2 has type annotations, which are effectively syntactically constrained comments within individual statements. As code they do absolutely nothing. However, the performance hit from just being present is so great that talented programmers are working hard to make the Python interpreter ... ignore them more efficiently. -- cheers, Hugh Fisher

1 0

Re: Revive PEP 396 -- Module Version Numbers ?
by Christopher Barker 14 Apr '21

14 Apr '21

Simon, I'm not sure your note made it to the list -- bringing it back on. On Wed, Apr 14, 2021 at 1:15 AM Simon Cross <hodgestar+pythondev(a)gmail.com> wrote: > I would be +1 on having a small PEP like this one to give __version__ > an official blessing and some well-defined expected behaviour. > > I don't have a deep opinion on the specifics of the specification, but > I'm guessing they should be thought about again even if the answers > end up being the same. Quite a lot has changed in the Python ecosystem > in the last 13 years. > Yes indeed. Critically, over on the packaging side, standardization of versioning specs. -CHB -- Christopher Barker, PhD (Chris) Python Language Consulting - Teaching - Scientific Software Development - Desktop GUI and Web Development - wxPython, numpy, scipy, Cython

1 0