Skip to content

Cite external upstreams instead of the internal research/ tree#425

Open
xmap wants to merge 1 commit into
mainfrom
worktree-scrub-research-links
Open

Cite external upstreams instead of the internal research/ tree#425
xmap wants to merge 1 commit into
mainfrom
worktree-scrub-research-links

Conversation

@xmap

@xmap xmap commented Jun 29, 2026

Copy link
Copy Markdown
Owner

Summary

research/ is now local-only and gitignored (per merged #420/#421) — the developer's private upstream practice, not shared. But tracked public pages still linked into it with github.com/xmap/cora/.../research/... URLs that now 404, and the NSRRC research brief still lived in the published docs tree. Both leak private research into the public site. This fixes the leak by citing the real external upstream each descriptor was reverse-engineered from.

  • APS (4-ID / 8-ID / 9-ID): point provenance at the BCDA-APS/*-bits instrument repos.
  • ESRF (ID19 / ID16B): point at the gitlab.esrf.fr/<bl>/beamline_configuration Beacon configs.
  • SYRMEP / FAXTOR: point at the elettra.eu / cells.es facility pages.
  • NSRRC: remove docs/deployments/nsrrc/_research_brief.md from the published tree and scrub its inbound links (TPS-05A/07A pages + the deployments index). The brief is preserved locally under the gitignored research/nsrrc/.
  • Tooling: point scripts/reverse_engineer defaults at research/aps/ (the canonical local layout) instead of the old research/aps-reverse-engineering/.

After this, no public page references research/.

Context

Supersedes the closed #422 (which tried to make research/ a tracked tree — the opposite of the local-only direction main settled on). The disciplined two-tier research structure from #422 is preserved on-disk locally, gitignored.

Test plan

  • pytest tests/unit/deployments/ — 541 passed
  • make docs-build (mkdocs --strict) — exit 0, no orphan-page or broken-link regressions
  • git grep confirms zero research/ path references in tracked files (only the .gitignore rule remains)
  • Reviewer: confirm the external-citation provenance reads right on each page

🤖 Generated with Claude Code

…search/ tree

The research/ tree is now local-only and gitignored (#420/#421): it is the
developer's private upstream practice, not shared. But tracked, public
deployment pages still linked into it with github.com/.../research/... URLs that
404 now that research/ is out of the repo, and the NSRRC research brief still
lived in the published docs tree. Both leak private research into the public
site.

Fix the leak by citing the real external upstream the descriptor was
reverse-engineered from, and dropping the internal research/ pointer:
- APS beamlines (4-ID/8-ID/9-ID) -> the BCDA-APS/*-bits instrument repos
- ESRF beamlines (ID19/ID16B) -> the gitlab.esrf.fr Beacon configs
- SYRMEP/FAXTOR -> the elettra.eu / cells.es facility pages
- Remove docs/deployments/nsrrc/_research_brief.md from the published tree and
  scrub its inbound links (the TPS pages + deployments index); the brief is
  preserved locally under the gitignored research/nsrrc/.
- Point scripts/reverse_engineer defaults at research/aps/ (the canonical
  local layout) instead of the old research/aps-reverse-engineering/ path.

No public page references research/ any more. Deployment tests (541) pass and
mkdocs --strict builds with no orphan-page or broken-link regressions.

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