registry: the open-request table counted a closed issue as an open one (#743) #1916
No reviewers
Labels
No milestone
No project
No assignees
1 participant
Notifications
Due date
No due date set.
Dependencies
No dependencies set.
Reference
jaylouisw/packetforge!1916
Loading…
Add table
Add a link
Reference in a new issue
No description provided.
Delete branch "fix/1117-2306"
Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?
Refs #743.
What was wrong
PR #1902 landed
devices/huawei/oceanprotect-x8100.jsonand-x9100.json(the management plane, with the front-end census deliberately unmodelled and named in each spec's notes), and its closing reference closed the request at 2026-11-17 22:05 UTC - which is the order the issue's own last comment asked for.The row stayed in
research/open-device-requests.json, so two records went on lying:docs/DEVICE_REGISTRY.mdsaid "Seven hardware requests are open and cannot be modelled yet" and listed a closed issue as one of them, in the section that issue threads cite as the queue's record.scripts/check-closed-issues.jsprinted a contradiction for it on every merge ("closed against our own record"), and the message saidno spec existsabout two files that are indevices/.Nothing mechanical joined the two files: the ledger is keyed on the issue number, the catalogue on the model, and the only comparison was a human reading both.
What changed
research/open-device-requests.json: the X8100/X9100 row is out (six requests left). The E8000 row keeps the shared missing field and now says where the X8100/X9100 gap is recorded instead of pointing at a closed request.docs/DEVICE_REGISTRY.md: regenerated withnpm run regen:registry-doc- "Six hardware requests are open", minus that row, nothing else touched.test/closed-issues-audit.test.js: the join that was missing. Every ledger row's model list is split on the slash and compared, case- and punctuation-insensitively, against each spec'smodelandaliases; a row naming hardwaredevices/already holds fails the test. On the pre-fix tree it fails with#743 -> oceanprotect-x9100.Gates
node scripts/syntax-check.js- 400 files, all OK.node test/closed-issues-audit.test.js- 34 passed, 0 failed (32/2 before the fix).node test/doc-catalog-counts.test.js- 43 passed.npm test- 327 suites, 11,393 assertions, 0 failed, exit 0.4a8729903c49d404a7b0