Skip to content

Keep directory refreshes running past invalid repositories - #95

Merged
RaviTejaKNTS merged 1 commit into
mainfrom
fix/self-host-directory-automations
Nov 2, 2026
Merged

RaviTejaKNTS merged 1 commit into
mainfrom
fix/self-host-directory-automations

Conversation

@RaviTejaKNTS

Copy link
Copy Markdown
Collaborator

Published apps marked as GitHub-hosted but missing a repository URL entered the oldest-first queue and aborted the refresh when failure bookkeeping could not find an identity. Exclude unlinked apps with missing or malformed URLs, and make failure bookkeeping log and continue instead of terminating the batch. Linked projects can supply the repository identity when an app URL is absent.

GitHub, GitLab, and Codeberg now fail only above 20% failed rows or when a non-empty batch refreshes nothing. Every selected failure remains in the job summary. Empty queues succeed. This keeps isolated dead repositories visible without making every scheduled run red.

Validation:

  • npm run quality: TypeScript and 116 core tests passed.
  • 35 focused tests passed without skips, including isolated PostgreSQL queue selection and CLI exit-status checks at and above the 20% boundary.
  • Read-only production CLI run: --published-only --limit 500 --dry-run completed with 457 successful checks and 43 failures (8.6%). PostgreSQL connections had default_transaction_read_only=on; no production data was written.
  • Merged origin/main into the existing branch before applying the hotfix.

@RaviTejaKNTS
RaviTejaKNTS merged commit 7ddc5ce into main Nov 2, 2026
4 checks passed

This branch was previously deployed

1 inactive deployment
production — 05c89e59 Deployed Nov 2, 2026 by RaviTejaKNTS via verify-production-build #90
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