Repository navigation
Conversation
… root inference, and longpress duration fallback Covers the four 0-33%-covered daemon functions from #3424 (G1) through their public entry points: - applyRequestLockPolicy: fresh-session target vs lock-platform matrix (targetSelectorsConflict, 33% -> 83%) - the shared auxiliary HTTP auth gate: hook rejection rendered as the flat REST error at the hook's status (sendAuxiliaryAuthHookRejection, 0% -> 100%) - extractTarInstallableArtifact: top-level root inference for ios and android without an expectedRootName (inferArchiveRootName, 0% -> 100%) - SessionScriptWriter.write: longpress duration recovery from recorded positionals, dropping blank/non-numeric tails (readLongPressDurationFromPositionals, 0% -> 100%) Each new test was observed to fail under a targeted mutation of its function before landing.
Size Report
Startup median (7 runs, lower is better):
|
![cubic-dev-ai[bot]](https://avatars.githubusercontent.com/in/1082092?s=60&v=4){kind=small}
There was a problem hiding this comment.
All reported issues were addressed across 4 files
Reply with feedback, questions, or to request a fix.
View guided diff | Turn on auto-fix | Re-trigger cubic
Cubic review: the fresh-session matrix only pinned rejections; a change that also rejected the valid mobile target would have passed. Adds the accepted-target case for each mobile-family lock platform.
|
This PR is ready at 9d82198. The new tests read correctly against the code, and all 18 checks are green. Since the diff only touches tests, no production route can cause failures here. There are no conflicts, and I did not run the tests or mutations locally. Not blocking, and you can take or leave these: the auth-hook harness in http-authorization.test.ts copies the one in http-server-tenant-trust.test.ts, so extract it if a third caller appears; the rejection tests match message text and could also assert The cubic-dev-ai thread on accepted mobile cases per lock (#3447 (comment)) is fixed at this head, because the loop at request-lock-policy.test.ts:302-315 covers android, harmonyos and ios. Please resolve it. |
thymikee
added
the
ready-for-human
This branch has not been deployed
1 participant
Summary
Covers the four low-coverage CRAP findings from #3424 (G1 of 3) through their owning modules' public entry points; no production changes, no splits.
request-lock-policy.tstargetSelectorsConflictserver/http-authorization.tssendAuxiliaryAuthHookRejectionartifact-archive.tsinferArchiveRootNamesession-script-writer.tsreadLongPressDurationFromPositionalsRoutes:
applyRequestLockPolicyfresh-session target/lock-platform matrix; the shared aux HTTP auth gate rejecting a hook via the diagnostics route;extractTarInstallableArtifactwithoutexpectedRootName;SessionScriptWriter.writelongpress selector rewrite. Each new test was observed to fail under a targeted mutation of its function, then pass again.Part of #3424, part of #3414.
Validation
8c6bbd8a0(rebased onto currentmain).pnpm check:affected --run: all runnable checks passed (format, lint, typecheck, fallow, related tests) under the machine-wide lock.pnpm check:coverage-crap --allon a per-file map: no function at or above 30 among the four owned files (was 4).