OpenConceptLab/ocl_online#517 | treat pro as community

[snyaggarwal]

Linked Issue

Ref OpenConceptLab/ocl_online#517

siblings
OpenConceptLab/oclapi2#945
OpenConceptLab/oclweb3#66

[snyaggarwal]

Verdict: Request changes

Issue complete: no — the Pro behavior and exact-name tests are correct, but the helper extraction leaves a live consumer broken.

Finding:

1. src/components/map-projects/MapProject.jsx still imports hasAuthGroup from ../../common/utils and calls it at line 616, while this PR removes that export from common/utils.js. Update the consumer to import from common/authGroups.js (or re-export the helper from utils.js). Otherwise the map-project screen receives an undefined binding and fails when it evaluates the core-user check.

Verification: npm test passes all 508 tests. The production webpack build reached emission, but the sandbox prevented writing dist/index.html; the existing unit suite does not exercise this component import.

Security blockers: none.

(snyaggarwal • Codex/GPT-5)

[snyaggarwal]

Verdict: Approve on re-review

The prior blocker is resolved: MapProject.jsx now imports hasAuthGroup from common/authGroups, and common/utils.js re-exports it for compatibility with any remaining consumers. Pro receives the intended early-access request limits and output-language choice, while exact matching prevents mapper-approved from matching pro.

Verification: all 508 unit tests pass; PR lint/test and Snyk checks are green.
Security blockers: none.

GitHub does not allow authors to formally approve their own PR, so this verdict is posted as a review comment.

(snyaggarwal • Codex/GPT-5)

[paynejd]

I recommend starting over

[snyaggarwal]

Closing without merging. Plan names and the limits they grant stay out of open code, so this PR's approach is withdrawn. The reworked approach is tracked on the private ticket (ocl_online#517).