linkedin / linkedin/Liger-Kernel

pyproject.toml has no requires-python — pip installs on any interpreter; publish workflows disagree (3.8 vs 3.10)

Open
#1,302 1 comment 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Dominant language
Python
Stars
6.6k
Forks
603
Avg merge
1d 20h
Merged PRs (30d)
47

Description

Hi, and thank you for Liger Kernel.

While packaging-reviewing the repo (at HEAD `72a4ed4`) I noticed `pyproject.toml` has **no `requires-python`** in `[project]`. Consequences:

- `pip` will happily resolve and install liger-kernel on any interpreter, however old — users on unsupported Pythons get runtime failures instead of a clean "requires Python >= X" resolver message.
- The two publish workflows don't agree on a floor either: `publish-nightly.yml` builds with Python **3.8**, `publish-release.yml` with **3.10** — so the effective supported floor is ambiguous even inside CI.

I'd have sent a one-line PR adding `requires-python = ">=3.10"`, but the floor is a support commitment only you can make (and it hard-blocks installs below it), so asking first: **what floor do you want?** Happy to send the PR (plus aligning the nightly workflow's Python) once you name it.

For transparency (per AGENTS.md): I used AI assistance to spot and draft this; I verified pyproject and both workflows myself.

Contributor guide

Open the contributing guide

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

Start by reading the [project] metadata in pyproject.toml and compare the Python versions in publish-nightly.yml and publish-release.yml. Confirm the maintainer-approved minimum Python version, then update the package metadata and align the nightly workflow; done means the resolver floor and both publish workflows agree.

Written by the indexing model from the issue text.

Assessment

Tech stack
github-actions, python
Domain
build-system, ci-cd, release
Issue type
Bug
Difficulty
2/5
Estimated time
1-3 hours
Activity status
Quiet
Clarity
Mostly clear
Newbie friendliness
55/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.