Question about Python 3.10+ support in BentoML (>=1.3.10) and its dependency simple_di

[oogou11]

I’m using the BentoML framework (version >= 1.3.10) and noticed that it explicitly states support for Python 3.10, 3.11, and 3.12. However, one of its dependencies, simple_di, seems to only declare support for Python 3.6 through 3.9 in its PyPI page and setup.py file.

Here are my questions:

1. Has simple_di been tested and confirmed to work with Python 3.10+?
2. Is it possible that the Python version range in setup.py hasn’t been updated, leading to this apparent mismatch with BentoML’s stated compatibility?
3. If simple_di hasn’t been fully validated for Python 3.10+, could there be potential compatibility issues?

Thank you for any insights or clarifications! I’d be happy to help with testing if needed.

[frostming]

1. simple-di doesn't cover Python 3.10+ in its own CI, but since it is a home-grown dependency used by BentoML, the latter is tested against Python 3.10+
2. I don't think it needs to be updated, the classifiers is an info-only metadata and what is important is the python_requires value.
3. Same as 1

[oogou11]

Thank you for your clarification.

[aarnphm]

quick chime in here, fwiw we should add test matrix to 3.10+, but from our end it doesn't seem like anything breaks for 3.11 and beyond.

[oogou11]

Thank you for the reply and support.