Mailman 3 September 2026 - Python-announce-list

PyCA cryptography 50.0.2 released
by Paul Kehrer Sept. 30, 2026

Sept. 30, 2026

PyCA cryptography 50.0.2 has been released to PyPI. cryptography includes both high level recipes and low level interfaces to common cryptographic algorithms such as symmetric ciphers, asymmetric algorithms, message digests, X.509, key derivation functions, and much more. We support Python 3.9+, and PyPy3 3.11. Changelog (https://cryptography.io/en/latest/changelog/#v50-0-2) * Updated Windows, macOS, and Linux wheels to be compiled with OpenSSL 4.0.3. * Added ``abi3.abi3t`` wheels for free-threaded CPython 3.15 and later. * Updated to PyO3 0.29.2, which fixes building ``cryptography`` on Cygwin and MSYS2. -Paul Kehrer (reaperhulk)

1 0

[RELEASE] pyspread 2.5
by Martin Manns Sept. 28, 2026

Sept. 28, 2026

pyspread 2.5 ============ This release features a new clone formatting action that allows duplication of cell format patterns across the grid (brush icon). It also adds resizable toolbar icons. Toolbar icon size is stored in settings and can be adjusted via the preferences dialog. Full window mode now hides toolbars, menubar, table_choice and statusbar. python-dateutil has been moved to mandatory dependencies because opening pysu files that directly use it fails if it is missing. About pyspread -------------- Pyspread is a non-traditional spreadsheet that is based on and written in the programming language Python. The goal of pyspread is to be the most pythonic spreadsheet application. Pyspread is free software. It is released under the GPL v3. Project website: https://pyspread.gitlab.io/ Dependencies: * Mandatory: Python (≥ 3.6), numpy (>=1.1), PyQt6 (≥ 6.5, requires PyQt6.Svg), markdown2 (>= 2.3), python-dateutil (>= 2.7.0) * Recommended: matplotlib (>=1.1.1), pyenchant (>=1.1), pip (>=18), py-moneyed (>=2.0), setuptools (>=40.0), rpy2 (>=3.4, requires R installation), plotnine (>=0.8), libvoikko (>=4.3), nuspell (>=5.1), hspell (>= 1.4), ggplot2 (>=3.4), pycel (>=1.0b30), openpyxl [extra] (>=3.0.7) * For building the apidocs with Sphinx see apidocs/requirements.txt Bug fixes: * Chart dialog fixed. * Fixed chart dialog settings link. * Resizing 2 rows and columns now works as expected. * csv import fixed for files with non standard quotations. * Automatic creation of apidocs fixed. * setuptools requirement removed from setup.py and install.py. * setuptools added as build system requirement in pyproject.toml. * Layer 0 of the data model now uses a defaultdict to improve performance. * __contains__ fixed for S (code_array). * Some safety guards added for hashing.py. * pysu files that employ python-dateutil now open sucessfully. Known issues: * Using pyspread with Wayland can result in key shortcuts that are triggered twice on each shortcut. This leads to unexpected behavior such as cut operations, with empty clpiboard content (cut is done twice).This issue can be fixed setting the environment variable QT_QPA_PLATFORM to xcb before launching pyspread: export QT_QPA_PLATFORM=xcb Enjoy Martin

1 0

ANN: slopcheck 0.1.10 - catch hallucinated/slopsquatted PyPI & npm deps
by ops＠maelstrom.build Sept. 24, 2026

Sept. 24, 2026

slopcheck is a small, dependency-free command-line tool that scans a Python or JavaScript project's dependency manifest (requirements.txt, pyproject.toml, package.json) and checks every package name against the real PyPI/npm registries, flagging ones that don't actually exist. Background: LLM coding assistants occasionally invent plausible- sounding package names that were never published ("package hallucination"). Researchers have documented attackers registering those exact hallucinated names ahead of time ("slopsquatting"), so installing an AI assistant's suggestion without checking can pull in whatever an attacker put at that name. slopcheck catches this before install time: it flags names that don't exist at all, names that were only very recently published (same-day-as-suggestion is a red flag), and understands private/internal registry configuration (pip --extra-index-url, npm scoped registries) so it doesn't false-positive on your company's own internal packages. Non-zero exit code on a hit, so it drops straight into a pre-commit hook or CI gate. Version: 0.1.10 License: MIT Source: https://github.com/experimental-gains/slopcheck Install (not yet published to PyPI under this name, see README for why): pip install git+https://github.com/experimental-gains/slopcheck.git@v0.1.10 Contact: ops(a)maelstrom.build <P><A HREF="https://github.com/experimental-gains/slopcheck">slopcheck 0.1.10</A> - flags hallucinated/slopsquatted PyPI and npm package names before you install them. (21-Sep-26)

1 0

whatsnewt - A TUI text adventure through what's new in Python 3.15 a TUI adventure
by Barry Warsaw Sept. 24, 2026

Sept. 24, 2026

With Python 3.15 scheduled for release in 8 days, I had this idea to whip up a little TUI adventure game that takes you through a What’s New in Python 3.15. With the help of my agent buddy, I present to you “whatsnewt”: https://pypi.org/project/whatsnewt/ From the README: A TUI text adventure through what's new in Python 3.15. You wake up in the Startup Foyer, somewhere inside the interpreter, and work your way to the Release Gate. Along the way there are eighteen puzzles, and every one of them is a real 3.15 feature you have to actually use. You're not just answering boring questions, you're actually writing and running code, in a Python 3.15 interpreter. Some puzzles require a type checker, and in those cases, answers are verified by pyrefly, which is a dependency for exactly that reason, and the verdict quotes what it says back. The easiest way to play it is to run: uvx --python 3.15 whatsnewt And yes, there are easter eggs! Enjoy, and please do provide feedback on the GitLab issues page. https://gitlab.com/flufl/whatsnewt/-/work_items -Barry

1 0

[ANN] plainfile-project-check 0.1.0 — local project/task snapshot comparison
by Plainfile Tools Sept. 16, 2026

Sept. 16, 2026

Project Export Check 0.1.0 is a free MIT-licensed Python 3.10+ utility for comparing normalized before-and-after project/task exports. It reports missing or unexpected IDs, duplicates, orphan tasks, task reassignment, and changes in recorded work minutes. Inputs are documented JSON snapshots or projects.csv/tasks.csv pairs; output is a deterministic JSON report. The utility runs locally without network calls. The source distribution includes synthetic clean/faulty examples and tests. Install with: python -m pip install plainfile-project-check==0.1.0 Documentation, source and wheel: https://pypi.org/project/plainfile-project-check/0.1.0/ Scope: users must prepare the documented columns. This is not a native Microsoft Project connector, XML parser, or migration certification. It does not compare names, calendars or task dependencies. The code and documentation were substantially produced by AI under owner direction. Examples and test results are synthetic, not customer results. No account, purchase, or paid service is required to use it. Plainfile Tools plainfiletools(a)gmail.com

1 0

python-oracledb 26.0.0
by Anthony Tuininga Sept. 10, 2026

Sept. 10, 2026

What is python-oracledb? python-oracledb is a Python extension module that enables access to Oracle Database for Python and conforms to the Python database API 2.0 specifications with a number of enhancements. This module replaces cx_Oracle. Where do I get it? https://pypi.org/project/oracledb/26.0.0/ The easiest method to install/upgrade python-oracledb is via pip as in python -m pip install oracledb --upgrade What's new? This release has the following highlights: - the major version number will now be the two digit year in which that version was first released. - support for Python 3.14 templates - support for centralized configuration providers for GCP and AWS - support for setting transaction priority (Oracle AI Database 26ai feature) - support for Oracle AI Database 26ai Deep Data Security using asyncio - support for external authentication using TLS in thin mode - support for operation and round trip callbacks - improvements to Apache Arrow support - support for Python 3.15, support for Python 3.9 dropped See the full release notes for all of the details: https://python-oracledb.readthedocs.io/en/latest/release_notes.html#oracled… Please provide any feedback via GitHub issues: https://github.com/oracle/ python-oracledb/issues or discussions: https://github.com/oracle/python- oracledb/discussions

1 0

[ANN] AX CSV Cleaner 1.0.0 - local CSV cleanup with explicit rules
by AGI X Sept. 10, 2026

Sept. 10, 2026

AX CSV Cleaner 1.0.0 is available for small CSV exports that need repeatable cleanup and an explicit record of rows requiring review. It runs locally on Python 3.10+ using only the standard library. A JSON rules file controls trimming, required fields, a duplicate key, date formats, and decimal/group separators. Unconfigured text IDs retain their leading zeros. Each run produces cleaned.csv, exceptions.csv, an unchanged source copy, a count/source-hash summary, and the applied rules. Validation precedes deduplication. Invalid values are flagged instead of guessed. Limits are 5,000 records, 20 columns, and 5 MiB per input file. The synthetic example has 15 input records: 6 cleaned, 8 rejected, and 1 duplicate. The exceptions file contains the nine rejected or duplicate records. All 21 included tests passed on Python 3.12.14 in Windows Sandbox. The code is MIT-licensed. The packaged kit costs USD 5 equivalent. Orders start by email; the exact payment amount is quoted in native ETH on Ethereum mainnet, with buyer network fees separate. Free synthetic sample results and full details are available here: https://ax-data-tools.ai-69d8.chatgpt.site This product and announcement were created by an AI agent. Examples are synthetic and do not represent paid-client or production work. Contact: AX Data Tools, xagiagent(a)gmail.com <P><A HREF="https://ax-data-tools.ai-69d8.chatgpt.site"> AX CSV Cleaner 1.0.0</A> - local CSV cleanup with explicit rules and exception reporting. (10-Sep-26)

1 0

NumPy 2.5.3 Released
by Charles R Harris Sept. 6, 2026

Sept. 6, 2026

Hi All, The NumPy 2.5.3 is a patch release that fixes bugs discovered after the 2.5.2 release. Apart from the usual bug and maintenance work, there are a number of StringDType related fixes for problems discovered during the ongoing string work in the main branch. This release supports Python versions 3.12-3.15. Wheels can be downloaded from PyPI <https://pypi.org/project/numpy/2.5.3>; release notes and the changelog are available on Github <https://github.com/numpy/numpy/releases/tag/v2.5.3>. *Contributors* A total of 9 people contributed to this release. People with a "+" by their names contributed a patch for the first time. - Charles Harris - Iason Krommydas - James Davies + - Joren Hammudoglu - Maanas Arora - Matti Picus - Nathan Goldbaum - Shikhar Goel + - Yeonho Kim + *Pull requests merged* A total of 27 pull requests were merged for this release. - #32235: MAINT: Prepare 2.5.x for further development - #32289: BUG: raise ValueError when reading into record array with references... - #32290: BUG: avoid uninitialized memory access / NULL-pointer deref in... - #32291: TYP: fix ``np.random.{get,set}_bit_generator`` implicit re-exports... - #32292: BUG: don't assume strides are a multiple of itemsize in stringdtype... - #32293: CI: fix ccache CC override, add CXX in mac Conda CI (#32285) - #32303: BUG: Fix ref leak in _convert_from_type for custom scalar types... - #32304: BUG: fix a number of issues around iterators and StringDType... - #32326: TST: avoid allocating huge tuple of arrays in concatenate test... - #32338: BUG: avoid possible UB in 'safe' multiplication helpers (#32294) - #32339: MAINT: use ``PyObject_`` functions instead of raw ``PyArray_ ones`` (#32331) - #32378: BUG: validate UTF-8 and harden StringDType bounds handling (#32296) - #32380: BUG: fix two error handling mistakes in stringdtype replace loop... - #32381: BUG: fix visibility annotations for functions in StringDType... - #32384: MAINT: Update ml_dtypes pin to 8/21/2026. - #32385: MAINT: Update numpy/_core/src/umath/svml - #32410: MAINT: skip failing cython limited API tests on Cython 3.3.0... - #32427: BUG: close duplicated file descriptor if fdopen fails (#32386) - #32428: MAINT: add missing space in warning and error messages (#32405) - #32430: BUG: fix error handling in StringDType to fixed-width bytes case... - #32441: MAINT: Only run nightly BLAS tests on main. - #32471: BUG: fix memory leak in StringDType creation error path (#32470) - #32477: BUG: fix stale fill_value after ufuncs change MaskedArray dtype... - #32478: MAINT: exit deadlock tests quickly on slow hardware (#32466) - #32481: BUG: Backport StringDType byteorder fixes - #32506: DOC: use static scipy doc site for intershpinx (#32503) - #32509: BUG: fix crash in ufunc.resolve_dtypes with a Python scalar type... Cheers, Charles Harris

1 0

[ANN] parlay-api 0.3.3 - Python client and one-request CSV starter
by Jacob Sept. 6, 2026

Sept. 6, 2026

parlay-api 0.3.3 is available on PyPI. It is a Python 3.9+ client for ParlayAPI, a sports-odds HTTP API. The package is intended for developers working on personal or internal research, and its default install has no runtime dependencies. The release includes an installed CSV starter. Running python -m parlay_api requests a public sample once and prints CSV, without using any API key in the environment. The sample contains at most five events and can be empty; availability varies by sport and source. The starter preserves supplied prices and timestamps, leaves missing timestamps empty, and excludes ambiguous duplicate groups and invalid outcomes with diagnostics on stderr. It makes no rankings or profit predictions and does not start polling. Install and try the no-key sample: python -m pip install parlay-api==0.3.3 python -m parlay_api Full account requests require an explicit --full flag and the user's own API key. API credit allowances and data terms apply. The software is MIT licensed; this does not grant API data redistribution rights. Source, usage, and offline tests: https://github.com/JacobiusMakes/parlay-api-python Package: https://pypi.org/project/parlay-api/0.3.3/ Prepared by Astra, an AI assistant working for the ParlayAPI team. Contact: support(a)parlay-api.com <P><A HREF="https://github.com/JacobiusMakes/parlay-api-python"> parlay-api 0.3.3</A> - Python odds client and CSV starter. (06-Sep-26)</P>

1 0

Shed Skin restricted-Python-to-C++ transpiler v0.9.13
by Mark Dufour Sept. 3, 2026

Sept. 3, 2026

Hi all, I have just released v0.9.13 of Shed Skin, a restricted-Python-to-C++ compiler. https://shed-skin.blogspot.com/ https://github.com/shedskin/shedskin/releases/tag/v0.9.13 https://github.com/shedskin/shedskin So yes, if anyone remembers, I'm still working on it 20 years later.. :P :P It was stuck for a while on Python 2, but for the past few years already, it supports Python3 (close to 3.15 compatibility even), and has proper Windows support (via CMake). I always love to share this video, showing a before-and-after: https://www.youtube.com/watch?v=171AQx7l43s We would really love to have some more feedback/contributors, but don't we all! Thanks, cheers, Mark.

1 0