AxelNoun commented on issue #921: URL: https://github.com/apache/tsfile/issues/921#issuecomment-6036408307
Follow-up on the release status, checked on 2026-10-07. - `datasets` 5.1.0, published on PyPI on 2026-10-05, requires `pyarrow>=24.0.0` (huggingface/datasets#8502). - The latest stable `tsfile` release on PyPI is still 2.4.0, which requires `pyarrow<20` on Python 3.10 to 3.13 and `pyarrow<24` on Python 3.14. As a result, `datasets==5.1.0` and `tsfile==2.4.0` no longer resolve together on Python 3.14 either, although the bounds still overlapped there when this issue was opened. I checked this with `uv pip install --dry-run` on Python 3.12 and 3.14. - The workaround given above first installs `pyarrow>=21,<24`, a range that `datasets` 5.1.0 no longer accepts. The Python 3.14 CI job of `datasets` currently installs `tsfile` with `--no-deps` alongside `pyarrow>=24.0.0`. - The pre-releases 2.5.0.dev0 and 2.5.0.dev1 carry the relaxed bounds from #922, and `datasets==5.1.0` resolves with `tsfile==2.5.0.dev1` on both Python versions. Since pip and uv do not select pre-releases by default, a requirement such as `tsfile>=2.3.0` still fails to resolve against `datasets` 5.1.0. Is there a target date for the 2.5.0 release? -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
