[issue46070] _PyImport_FixupExtensionObject() regression causing a crash in subintepreters

Change by Miro Hrončok : -- nosy: +hroncok ___ Python tracker ___ ___ Python-bugs-list mailing list Unsubscribe: https://mail.python

[issue46148] Optimize pathlib

New submission from Kumar Aditya : Issue for tracking improving performance of pathlib module. -- components: Library (Lib) messages: 409015 nosy: gvanrossum, kumaraditya303 priority: normal severity: normal status: open title: Optimize pathlib type: performance versions: Python 3.11 _

[issue46148] Optimize pathlib

Change by Kumar Aditya : -- keywords: +patch pull_requests: +28449 stage: -> patch review pull_request: https://github.com/python/cpython/pull/30226 ___ Python tracker ___ ___

[issue41222] POpen bufsize=0 ignored with universal_newlines=True

Antoine Pitrou added the comment: Hmm, sorry for not responding earlier. Buffering is necessary for implementing the universal_newlines behaviour (I don't know how we could do otherwise?). This has the unavoidable side effect that the Python buffered file object is not in sync with the under

[issue46149] FIPS usedforsecurity flag is no longer functional with OpenSSL 3.0.0

New submission from Florin Spătar : I added christian.heimes to the nosy list; I hope that's OK usedforsecurity flag is based on setting the EVP_MD_CTX_FLAG_NON_FIPS_ALLOW flag. However this flag has no effect in OpenSSL 3.0.0. [root@lambada ~]# /opt/opsware/bin/python3 Python 3.10.0 (

[issue46146] Python IDLE fails to start (tk font issue?)

Matej Cepl added the comment: $ rpm -qR python38-idle /usr/bin/python3 /usr/bin/python3.8 config(python38-idle) = 3.8.12-3.1 python(abi) = 3.8 python38 = 3.8.12 python38-tk rpmlib(CompressedFileNames) <= 3.0.4-1 rpmlib(FileDigests) <= 4.6.0-1 rpmlib(PartialHardlinkSets) <= 4.0.4-1 rpmlib(Payloa

[issue46149] FIPS usedforsecurity flag is no longer functional with OpenSSL 3.0.0

Christian Heimes added the comment: This is a known issue, see #40479. OpenSSL 3.0.0 new provider system requires a major redesign of hashlib module's internals. It's not a trivial change. My first attempt GH-19878 has some flaws and introduces a performance regression. -- superseder

[issue46146] Python IDLE fails to start (tk font issue?)

Matej Cepl added the comment: And I am not sure what’s going on, but when I run the test suite you indicated, it kills my Wayland session immediately. How is Tk doing with X11->Wayland switch? -- ___ Python tracker

[issue46146] Python IDLE fails to start (tk font issue?)

Matej Cepl added the comment: And yes, idle (both with 3.10 and 3.8) works for me just fine all the time (it doesn't for my reporter, but that's another point). -- ___ Python tracker ___

[issue46148] Optimize pathlib

Christian Heimes added the comment: Please provide benchmarks and reasons in the ticket. A BPO issue should contain all relevant information for a change. -- nosy: +christian.heimes ___ Python tracker _

[issue1284670] Allow to restrict ModuleFinder to get "direct" dependencies

Change by mike mcleod : -- pull_requests: +28450 pull_request: https://github.com/python/cpython/pull/30228 ___ Python tracker ___ ___

[issue37295] Possible optimizations for math.comb()

Serhiy Storchaka added the comment: long long is at least 64 bit, so we can safely use PyLong_FromLongLong() for int64_t. -- ___ Python tracker ___ __

[issue46148] Optimize pathlib

Kumar Aditya added the comment: Optimizations include: - Remove redundant check for functools lru_cache - Replace attrgetter as it is slower than direct access -- ___ Python tracker

[issue46148] Optimize pathlib

Kumar Aditya added the comment: Benchmarks for attrgetter is provided in the PR. -- ___ Python tracker ___ ___ Python-bugs-list mai

[issue20392] Inconsistency with uppercase file extensions in MimeTypes.guess_type

Change by Kumar Aditya : -- keywords: +patch nosy: +kumaraditya303 nosy_count: 4.0 -> 5.0 pull_requests: +28451 stage: -> patch review pull_request: https://github.com/python/cpython/pull/30229 ___ Python tracker

[issue46149] FIPS usedforsecurity flag is no longer functional with OpenSSL 3.0.0

Florin Spătar added the comment: I was able to get past this issue with minimal changes: https://github.com/florinspatar/cpython/commit/3b16c65eb3e54c0be40413ebabf504356e303e8a I see the issue you linked, handles multiple OpenSSL 3.0.0 problems. AFAIK, python 3 no longer supports OpenSSL 1.0.2

[issue20823] [doc] Clarify copyreg.pickle() documentation

Change by Kumar Aditya : -- keywords: +patch nosy: +kumaraditya303 nosy_count: 3.0 -> 4.0 pull_requests: +28452 pull_request: https://github.com/python/cpython/pull/30230 ___ Python tracker __

[issue7262] [doc] codecs.open() + eol (windows)

Change by Kumar Aditya : -- versions: -Python 3.10, Python 3.9 ___ Python tracker ___ ___ Python-bugs-list mailing list Unsubscribe:

[issue7262] [doc] codecs.open() + eol (windows)

Change by Kumar Aditya : -- versions: +Python 3.10, Python 3.9 ___ Python tracker ___ ___ Python-bugs-list mailing list Unsubscribe:

[issue7262] [doc] codecs.open() + eol (windows)

Change by Kumar Aditya : -- nosy: +kumaraditya303 nosy_count: 5.0 -> 6.0 pull_requests: +28453 stage: needs patch -> patch review pull_request: https://github.com/python/cpython/pull/30231 ___ Python tracker _

[issue37295] Possible optimizations for math.comb()

Mark Dickinson added the comment: [Tim] > The justification for the shift count isn't self-evident, and > appears to me to be an instance of the generalization of Kummer's > theorem to multinomial coefficients. Not sure there's any generalisation here: I think it *is* just Kummer's theorem.

[issue46149] FIPS usedforsecurity flag is no longer functional with OpenSSL 3.0.0

Christian Heimes added the comment: Your patch gets the work done, but it's even slower than my WIP patch set. Hashing is a performance critical path. The new fetch() API in OpenSSL 3.0.0 is substantially slower than the old OpenSSL 1.1.1 APIs. Python 3.9 and earlier still support OpenSSL 1.

[issue46150] test_pathlib assumes "fakeuser" does not exist as user

New submission from Thomas Wouters : test_pathlib contains, in PosixPathTest.test_expanduser, a check that expanduser on a nonexistent user will raise RuntimeError. Leaving aside the question why that's a RuntimeError (which is probably too late to fix anyway), the test performs this check by

[issue41222] POpen bufsize=0 ignored with universal_newlines=True

Eryk Sun added the comment: > Buffering is necessary for implementing the universal_newlines Why is that? I can see that it requires newline state tracking, and the allowance to make two read(fd, &c, 1) system calls for a single read(1) method call, in case a "\n" has to be ignored. testpr

[issue46140] Some API methods could take const Py_buffer* instead of Py_buffer *

Serhiy Storchaka added the comment: New changeset 31ff96712e8f89ac1056c2da880b44650002219f by David Hewitt in branch 'main': bpo-46140: take more Py_buffer arguments as const * (GH-30217) https://github.com/python/cpython/commit/31ff96712e8f89ac1056c2da880b44650002219f -- nosy: +ser

[issue46140] Some API methods could take const Py_buffer* instead of Py_buffer *

Serhiy Storchaka added the comment: Thank you for your contribution David. -- resolution: -> fixed stage: patch review -> resolved status: open -> closed ___ Python tracker _

[issue35037] PYLONG_BITS_IN_DIGIT differs between MinGW and MSVC

Mark Dickinson added the comment: > This should probably be a separate issue, Specifically, issue 45569. -- ___ Python tracker ___ ___

[issue20369] concurrent.futures.wait() blocks forever when given duplicate Futures

Change by Mark Dickinson : -- nosy: -mark.dickinson ___ Python tracker ___ ___ Python-bugs-list mailing list Unsubscribe: https://

[issue46151] SimpleCookie.js_output is vulnerable to HTML injection

New submission from Trung Pham : In /Lib/http/cookies.py, the output from SimpleCookie.js_output might be parsed as HTML if it contained < and >. ``` from http import cookies c = cookies.SimpleCookie() c["fig"] = "newtonalert(document.domain)"; // c.js_output()