Skip to content

Update URLs for the madsuite-org rename; codecov upload over OIDC - #638

Open
sshin23 wants to merge 5 commits into
masterfrom
robin/org-sweep
Open

sshin23 wants to merge 5 commits into
masterfrom
robin/org-sweep

Conversation

@sshin23

@sshin23 sshin23 commented Aug 11, 2026

Copy link
Copy Markdown
Member

Follow-up to the organization rename (MadNLP → madsuite-org) and the org-site move to madsuite-org.github.io:

  • README badges and links, the Documenter source/deploy targets, and the HybridKKT link in the GPU tutorial point at the new home. GitHub Pages does not redirect across the rename, so the docs links were the ones that had actually broken.
  • The codecov badge drops the repo-scoped graphing token and pins the madsuite-org/MadNLP.jl slug.
  • Coverage uploads switch from the shared CODECOV_TOKEN secret to OIDC: the workflow authenticates itself, attribution follows the repo that ran (which also populates codecov's new slug on this PR's first run), and there is no token whose codecov-side binding can go stale after a rename. The same convention is already in place in CNLPModels.jl and cnlpmodels-py.

🤖 Generated with Claude Code

sshin23 and others added 2 commits August 11, 2026 14:19
The MadNLP organization was renamed to madsuite-org and the org Pages
site moved to madsuite-org.github.io (GitHub Pages does not redirect, so
the docs links were the ones that had actually broken). README badges and
links, the Documenter source/deploy targets, and the HybridKKT link in the
GPU tutorial now point at the new home; the codecov badge drops the
repo-scoped graphing token and pins the new slug.

Coverage uploads switch from the shared CODECOV_TOKEN secret to OIDC:
the workflow authenticates itself, attribution follows the repo that ran
(which also populates codecov's new madsuite-org slug on the first run),
and there is no token whose codecov-side org binding can go stale after
a rename.
Measured while this PR was open: OIDC uploads are rejected org-wide
(the ExaModels/CNLPModels coverage blackout — 'Repository not found'
behind fail-open green steps), while token-authenticated uploads via
the org-level CODECOV_TOKEN work — this repository's own run tonight
uploaded clean with it (88.91% posted). The URL and badge updates
stand unchanged.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
@sshin23

sshin23 commented Aug 16, 2026

Copy link
Copy Markdown
Member Author

One commit added (0bbeab18), reverting only the codecov hunk: the OIDC switch is dropped and the three upload sites stay on token: ${{ secrets.CODECOV_TOKEN }} (org-level). Measured while this PR sat open: OIDC uploads are rejected org-wide — the mechanism behind the ExaModels/CNLPModels coverage blackout (Repository not found behind fail-open green steps) — while the org token works, as this repository's own upload tonight demonstrated (clean, 88.91% posted). The URL, badge, and Documenter changes are untouched and remain the substance of the PR.

🤖 Generated with Claude Code

@codecov

codecov Bot commented Aug 16, 2026 •

Copy link
Copy Markdown

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 88.87%. Comparing base (9947ee1) to head (b7213ae).
⚠️ Report is 3 commits behind head on master.

Additional details and impacted files
@@            Coverage Diff             @@
##           master     #638      +/-   ##
==========================================
- Coverage   88.91%   88.87%   -0.04%     
==========================================
  Files          55       55              
  Lines        5015     5015              
==========================================
- Hits         4459     4457       -2     
- Misses        556      558       +2     

☔ View full report in Codecov by Harness.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:
  • ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

sshin23 and others added 3 commits August 15, 2026 22:51
The project/patch checks keep reporting coverage and its delta on every
PR but never fail the rollup; a coverage drop is information, not a
broken build.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
The cuda13-labelled jobs (the GPU test leg and the docs build) queue
forever against a label no runner in the org carries; the live GPU
runners are labelled cuda. Repointed rather than removed, since the
docs build executes the GPU tutorial and needs a GPU runner at all.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
…ot imply a repository

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>

This branch has not been deployed

No deployments
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.

1 participant