Skip to content

minor debugging for publishing#2739

Merged
awni merged 2 commits intoml-explore:mainfrom
madrob:publishing
Nov 12, 2025
Merged

minor debugging for publishing#2739
awni merged 2 commits intoml-explore:mainfrom
madrob:publishing

Conversation

@madrob
Copy link
Copy Markdown
Contributor

@madrob madrob commented Nov 6, 2025

Proposed changes

Print hashes when uploading to PyPI
Skip already existing artifacts for Test-PyPI

Checklist

Put an x in the boxes that apply.

  • I have read the CONTRIBUTING document
  • I have run pre-commit run --all-files to format my code / installed pre-commit prior to committing changes
  • I have added tests that prove my fix is effective or that my feature works
  • I have updated the necessary documentation (if needed)

@madrob madrob requested review from angeloskath and awni November 6, 2025 13:10
outputs:
pypi_env: ${{ github.event_name == 'push' && 'pypi' || 'test-pypi' }}
pypi_url: ${{ github.event_name == 'push' && 'https://upload.pypi.org/legacy/' || 'https://test.pypi.org/legacy/' }}
pypi_dups: ${{ github.event_name == 'push' && 'true' || 'false' }}
Copy link
Copy Markdown
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I'm a bit confused by this variable. How is it used to determine if there is already a package of the same name on pypi?

Copy link
Copy Markdown
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I actually got it backwards! When we are publishing to pypi (on a tag pushed event) we want to fail on duplicates, when publishing to test-pypi (on a manual workflow launch event) it's fine to skip when there are duplicates.

For regular releases there shouldn't be duplicates, but for the test releases since we're not incrementing the version number in between then the releases could be repeating (until we delete them I guess?)

Copy link
Copy Markdown
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Tried to make it more clear in 88a9841

@madrob
Copy link
Copy Markdown
Contributor Author

madrob commented Nov 6, 2025

I'm pretty sure the failure at https://github.com/ml-explore/mlx/actions/runs/19140474034/job/54704171529 is not related to this change set, but I don't know if it is a hardware issue or machine setup issue or just a flaky test.

@awni
Copy link
Copy Markdown
Member

awni commented Nov 6, 2025

I agree that failure seems unrelated. Very strange. Those kinds of failures that only happen on circle are almost impossible to debug without ssh access :(

@zcbenz
Copy link
Copy Markdown
Collaborator

zcbenz commented Nov 6, 2025

For GitHub-hosted runners people can use https://github.com/mxschmitt/action-tmate to ssh into the machine. Not sure if it is usable for self-hosted runners or whether Apple's policy allows it.

Copy link
Copy Markdown
Member

@awni awni left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

LGTM!

@awni awni merged commit 3f866be into ml-explore:main Nov 12, 2025
13 of 14 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants