ambv, to python
@ambv@mastodon.social avatar

The core․py podcast wasn't dead, it was just sleeping! Pablo Galindo and I return to talk about the upcoming new REPL in 3.13 ♒️

Join us for our signature professional takes just in time for Python 3.13 beta 1 and 🛜

https://podcasters.spotify.com/pod/show/corepy/episodes/Episode-10-The-Interactive-REPL-e2j788i/a-ab7uiak

danzin, to python
@danzin@mastodon.social avatar

PyPy v7.3.16 has been released:
https://www.pypy.org/posts/2024/04/pypy-v7316-release.html

"This release includes security fixes from upstream CPython, and bugfixes to the garbage collector, described in a gc bug-hunt blog post."

The mentioned blog post is https://www.pypy.org/posts/2024/03/fixing-bug-incremental-gc.html by @cfbolz

Changelog at https://doc.pypy.org/en/latest/release-v7.3.16.html#changelog

Some changes I like:

  • Faster, correct str.expandtabs()
  • Fix subprocess executable path on Windows (unlisted)
  • Faster normalize_exception
  • Correct cache size detection on aarch64

mgorny, to python Polish
@mgorny@pol.social avatar

2024-03-19, ogłoszono dwie dziury bezpieczeństwa na liście mailingowej, poświęconej problemom bezpieczeństwa Pythona: "quoted zip-bomb" i "TemporaryDirectory symlink dereference during cleanup". Obie miały dotykać wszystkich aktualnych wydań CPythona.

Tego samego dnia, wydano nowe wersje Pythona 3.10, 3.9 i 3.8. Co ciekawe, nie było wydań dla 3.11 i 3.12.

2024-04-02, w końcu otagowano Pythona 3.11.9. Początkowo, podpis dla archiwum się nie sprawdzał. Dziś już jest OK, ale wydania dalej nie ogłoszono. Co jednak najbardziej mnie zaskoczyło, to brak poprawek dla dwóch ogłoszonych wcześniej problemów! Czyżby nieudane wydanie?

Przyjrzałem się sprawie bliżej… i okazało się, że oba problemy były już poprawione w poprzednim wydaniu 3.11.8 (i 3.12.2), więc ogłoszenie było błędne. Wzdych.

Niestety, nadal jest dotknięte problemem.

https://mail.python.org/archives/list/security-announce@python.org/thread/XELNUX2L3IOHBTFU7RQHCY6OUVEWZ2FG/
https://mail.python.org/archives/list/security-announce@python.org/thread/Q5C6ATFC67K53XFV4KE45325S7NS62LD/
https://discuss.python.org/t/python-3-10-14-3-9-19-and-3-8-19-is-now-available/48993
https://bugs.gentoo.org/927299

83r71n, to Cybersecurity
@83r71n@ioc.exchange avatar

The Python Package Index (PyPI) repository experienced a malware upload attack, forcing maintainers to suspend new project creation and user registration to mitigate the threat. This incident involved malicious Python packages, likely uploaded using typo-squatting techniques, designed to steal sensitive information and credentials. The malware also included a persistence mechanism to remain active on compromised systems.

https://status.python.org/incidents/dc9zsqzrs0bv

mgorny, to RegEx Polish
@mgorny@pol.social avatar

Paczka Pythona (nie mylić z wbudowanym modułem re) zbudowana jest w oparciu o szczegóły implementacji CPythona i nie obsługuje poprawnie (i autor zapowiada, że może w końcu zablokować kompilację na PyPy). Jednakże wygląda na to, że wymagająca jej paczka działa bez problemów ze zwyczajnym re.

Dzisiaj przechodzi z łatania w sposób niedoskonały paczki regex, i ignorowania szczególnych przypadków, w których nie zadziała, na rzecz łatania re-assert. Chciałbym wysłać tę trywialną łatkę autorowi, ale — jak już wcześniej narzekałem — dostałem niegdyś bana, autor nie potrafi powiedzieć dlaczego, ale nie przeszkadza mu to uważać bana za sprawiedliwego. Może po prostu proaktywnie banuje devów dystrybucji Linuksa.

https://gitweb.gentoo.org/repo/gentoo.git/commit/?id=8413cf2c2955533fdf212fea3970c99cf193d4a1
https://github.com/mrabarnett/mrab-regex/issues/521
https://github.com/mrabarnett/mrab-regex/issues/404

hynek, to random
@hynek@mastodon.social avatar
hugovk,
@hugovk@mastodon.social avatar

@hynek Using M1 cuts about 25 minutes from the @pillow CI, some jobs are twice or thrice as fast! Especially helps PyPy 🚀

https://github.com/python-pillow/Pillow/pull/7766

danzin, to python
@danzin@mastodon.social avatar

PyPy v7.3.15 release

https://www.pypy.org/posts/2024/01/pypy-v7315-release.html

"The PyPy team is proud to release version 7.3.15 of PyPy.

This is primarily a bug-fix release, and includes work done to migrate PyPy to Git and Github."

PyPy deserves a lot more recognition and contributions.

At the time of this post, PyPy has 530 stars on GitHub. Please star it if you haven't yet.

Activity seems to have slightly increased: https://github.com/pypy/pypy/graphs/contributors?from=2023-01-06&to=2024-01-16&type=c

The wiki has been ported to https://github.com/pypy/pypy/wiki/

danzin, to python
@danzin@mastodon.social avatar

Wow, PyPy has passed 400 stars on GitHub already: https://github.com/pypy/pypy

If you haven't starred it yet, now is the time :)

Also, their Sponsors page could do with some more people: https://github.com/sponsors/pypy.

For detailed data on past contributions (and another place to help fund PyPy if GitHub sponsoring isn't your thing), visit https://opencollective.com/pypy.

governa, to github
@governa@fosstodon.org avatar
film_girl, to random
@film_girl@mastodon.social avatar

I too wish Mercurial had won but I’m ecstatic has moved to GitHub! https://www.pypy.org/posts/2023/12/pypy-moved-to-git-github.html

danzin, to python
@danzin@mastodon.social avatar

PyPy has moved to Git, GitHub:
https://www.pypy.org/posts/2023/12/pypy-moved-to-git-github.html

"PyPy has moved its canonical repo and issue tracker from https://foss.heptapod.net/pypy/pypy to https://github.com/pypy/pypy."

Even though the team prefers Mercurial, they believe being on GitHub will foster contributions.

If you know your way around GitHub Actions, Buildbots, GH Wikis, or are interested in contributing, now is a great time to get involved. Starring the repo helps with visibility.

Congrats @mattip @cfbolz and team!

danzin, to python
@danzin@mastodon.social avatar

v7.3.14: release of 2.7, 3.9, and 3.10

"The PyPy team is proud to release version 7.3.14 of PyPy.

Hightlights of this release are compatibility with HPy-0.9, cffi 1.16, additional C-API interfaces, and more python3.10 fixes.
[...]
We would like to thank our donors for the continued support of the PyPy project.
[...]
We would also like to thank our contributors and encourage new people to join the project. "

Pypy is a great project, contribute to it!

https://www.pypy.org/posts/2023/12/pypy-v7314-release.html

paulox, to python
@paulox@fosstodon.org avatar

RT @pypyproject

PyPy 7.3.14 has just been released! 🎉
The main feature is HPy 0.9 support, as well as various bug fixes and small improvements. ✨ Thanks to Matti for doing the release! 🙏

https://www.pypy.org/posts/2023/12/pypy-v7314-release.html

jedie, to random German
@jedie@chaos.social avatar
mgorny, to opensource Polish
@mgorny@pol.social avatar

Typowa sytuacja we współczesnym , na przykładzie ekosystemu języka .

Wiele projektów używa biblioteki , by nadpisywać wskazania zegara na potrzeby testów. FreezeGun powoli przestaje być rozwijany. W końcu zaczyna mieć problemy z nowymi wersjami Pythona. Dystrybucje, takie jak , są odporne na te problemy, bo mogą łatwo dodać lokalne łatki.

Tak więc projekty zaczynają korzystać z . Niestety, time-machine opiera się na hakowaniu detali implementacji CPythona (w imię wydajności, bo przecież nadpisywanie czasu w testach to wydajnościowe wąskie gardło), więc na nie działa w ogóle. Niektóre projekty wspierają FreezeGun i time-machine równocześnie, inne nie.

Czasem time-machine łapie segfaulty na CPythonie. Z czasem coraz więcej segfaultów zostaje zgłoszonych. Nie ma więc zaskoczenia, że nowe zgłoszenia błędów nie spotykają się z odpowiedzią. W międzyczasie, FreezeGun na nowo zaczyna być rozwijany. No i zgadnijcie, co teraz się dzieje…

https://github.com/aio-libs/aiohttp/commit/a7bc5e9eeae7c5c90898411962e9a74bf10a9cef
https://github.com/aio-libs/aiohttp/commit/43f92fae09bcc9692ee96ac1413eda884afa2f63

mgorny, to gentoo Polish
@mgorny@pol.social avatar

Za sprawą zlecenia z , Inc. (@cJ), sporo paczek Pythona w dorobiła się wsparcia , m.in.: , Notebook, , , .

Chciałbym także podziękować autorom PyPy za ich wsparcie, zarówno w kwestii poprawiania błędów w PyPy, jak również udzielaniu pomocy innym projektom, by poprawić ich zgodność z PyPy. Praca z wami jest przyjemnością!

Na koniec, poznałem ważny argument za pracą nad wsparciem PyPy w projektach: nawet jeśli dana paczka nie działa szybciej na PyPy, to może być zależnością w większym projekcie, w którym PyPy ogółem przynosi lepszą wydajność.

cfbolz, to random
@cfbolz@mastodon.social avatar

This weekend I landed a CPython PR that I'm very happy about (with the help of @ambv and Dennis Sweeney):

I switched the storage of all the names of Unicode code points in the unicodedata modules to using a different data structure, a "directed acyclic word graph". This makes the compiled module 440 KiB smaller. I did the same thing in PyPy a year ago, quite happy that it now made it to CPython too.

https://github.com/cfbolz/cpython/commit/9573d142157d8432f2772a109c304dafeaa454a5#diff-372ed45e01f260ef615543328bba7a8af29b0ded4e8b1f4142be55bd75ed54bf

paulox,
@paulox@fosstodon.org avatar

@cfbolz @ambv that's a great PR 👏
So many great additions from PyPy into CPython in recent years. 🐍

iritkatriel, to random

At the last Python Language Summit in April, after three back-to-back sessions on the C API, we agreed that our discussions about the future of the C API are lacking a shared understanding of its current state, its strengths and weaknesses.

We decided to work towards a document summarising a community consensus on that, and have now put together the draft of PEP 733.

https://peps.python.org/pep-0733/

larstiq,
@larstiq@mathstodon.xyz avatar

@iritkatriel Nice! As a irc channel lurker, this looks like the most comprehensive collection of concerns about the C API I have seen. Hope that it will help make the lives of alternative Python implementations easier in the future.

kubikpixel, to python German
@kubikpixel@chaos.social avatar

deleted_by_author

  • Loading...
  • MadMike77,
    @MadMike77@chaos.social avatar

    @kubikpixel Ich dachte ein no-GIL Python für den allgemeinen Gebrauch würde als erstes aus dem Projekt kommen: https://doc.pypy.org/en/latest/faq.html#does-pypy-have-a-gil-why

    Mal sehen ob CPython doch schneller damit fertig wird.

    danzin, to python
    @danzin@mastodon.social avatar

    PyPy v7.3.13: release of 2.7, 3.9, and 3.10

    """
    The PyPy team is proud to release version 7.3.13 of PyPy. This is primarily a security/bug-fix release. CPython released security patches, and this release also improves the ability to use type specifications via PyType_FromSpec and friends. There are also some small speed-ups.
    """

    PyPy is a great project, run by great people. I wish it had more recognition and more resources devoted to it.

    https://www.pypy.org/posts/2023/09/pypy-v7313-release.html

    castarco, to python
    @castarco@hachyderm.io avatar

    1/2

    At first, I was a bit skeptical of the new Modular's language.

    Having no binaries available (only a playground), and a long history of contenders such as or Pyston that never achieved full compatibility... was a huge turnoff.

    The amount of Python compilers that never reach 100% compatibility is almost hilarious.

    Having seen the tremendous amount of effort behind , and how little is its community compared to Python's... adds on top of that.

    paulox, to python
    @paulox@fosstodon.org avatar

    Everyone, welcome to the Fediverse for Antonio Cuni @antocuni , HPy founder, PyPy and PyScript core dev and a friend of mine. 👋

    sethmlarson, to random
    @sethmlarson@fosstodon.org avatar

    v2.0.4 is now available! This was the first release by our new maintainer Illia Volochii 🎉

    now uses stdlib hostname verification if possible
    – Union operator support for HTTPHeaderDict
    – Emits http.client.connect audit events

    https://github.com/urllib3/urllib3/releases/tag/2.0.4

    mgorny, to python

    language lawyers, we're seeing some interesting issue with / ijson.

    ijson is calling the async function's wait() method and iterating over the results until it gets StopIteration exception, and it uses it to get the return value. This works in Python both in CPython and PyPy but within C API, CPython tp_iternext() raises StopIteration while in PyPy it does not — is that a bug or is there some other way to get the return value?

    https://github.com/ICRAR/ijson/issues/101
    https://foss.heptapod.net/pypy/pypy/-/issues/3956

    hugovk, to python
    @hugovk@mastodon.social avatar

    The new release of is out, supporting 3.9 and 3.10!

    https://www.pypy.org/posts/2023/06/pypy-v7312-release.html

    If you're testing PyPy on in several projects, you can use all-repos to update many at the same time. See @jugmac00's blog for setup tips:

    https://jugmac00.github.io/til/how-to-update-all-zope-repositories-at-once/

    And here's an example command, run first with --dry-run, then remove to create real PRs:

    all-repos-sed --dry-run --commit-msg "CI: Replace pypy3.8 with pypy3.10" 'spypy3.8pypy3.10~g' -- '*.yml'

  • All
  • Subscribed
  • Moderated
  • Favorites
  • JUstTest
  • GTA5RPClips
  • DreamBathrooms
  • InstantRegret
  • magazineikmin
  • khanakhh
  • Youngstown
  • ngwrru68w68
  • slotface
  • everett
  • rosin
  • thenastyranch
  • kavyap
  • tacticalgear
  • megavids
  • cisconetworking
  • normalnudes
  • osvaldo12
  • ethstaker
  • mdbf
  • modclub
  • Durango
  • tester
  • provamag3
  • cubers
  • Leos
  • anitta
  • lostlight
  • All magazines