JMR-dev/LibreMediaConverter
1
0
Fork 0
Code Issues 16 Pull Requests Actions Packages Projects Releases Wiki Activity

R36 — AppRootRestorationTest's KDoc repeats the claim R14 corrected #46

New Issue
Closed
opened 2026-08-23 04:03:21 +00:00 by JMR-dev · 1 comment
JMR-dev commented 2026-08-23 04:03:21 +00:00 (Migrated from github.com)
Copy Link
Copy Source

Found by the docs-correction stream while working R14 (#23); outside that stream's scope (app/src was owned by another agent), so it was reported rather than fixed.

R36 — A test's KDoc repeats the claim R14 corrected in the audit

severity: low
verdict: CONFIRMED
where: app/src/test/java/org/libremediaconverter/AppRootRestorationTest.kt:38-39
scenario: The KDoc states "The instrumented tests cannot run on the development host at all (see CLAUDE.md)". That is now false — tools/local-emulator/run-e2e.sh runs API 33–36 locally, 49 tests each. A reader trusting this comment concludes the local emulator path does not exist.
evidence: Same family as R14 (#23), which corrected the identical claim in docs/defect-audit.md. The docs stream verified the site but could not edit it under its scope rules.
fix: Restate on speed/determinism rather than impossibility, matching the wording R14 settled on. Leave the API 37 Pixel rule standing.
risk: None — comment only, no behaviour.

Cut: above — same family as R14/R15, factual correction, verifiable by inspection. Deferred to a follow-up because both owning streams had already committed by the time it surfaced.

🤖 Generated with Claude Code

_Found by the docs-correction stream while working R14 (#23); outside that stream's scope (`app/src` was owned by another agent), so it was reported rather than fixed._ ### R36 — A test's KDoc repeats the claim R14 corrected in the audit severity: low verdict: CONFIRMED where: app/src/test/java/org/libremediaconverter/AppRootRestorationTest.kt:38-39 scenario: The KDoc states *"The instrumented tests cannot run on the development host at all (see CLAUDE.md)"*. That is now false — `tools/local-emulator/run-e2e.sh` runs API 33–36 locally, 49 tests each. A reader trusting this comment concludes the local emulator path does not exist. evidence: Same family as R14 (#23), which corrected the identical claim in `docs/defect-audit.md`. The docs stream verified the site but could not edit it under its scope rules. fix: Restate on speed/determinism rather than impossibility, matching the wording R14 settled on. Leave the API 37 Pixel rule standing. risk: None — comment only, no behaviour. --- **Cut:** `above` — same family as R14/R15, factual correction, verifiable by inspection. Deferred to a follow-up because both owning streams had already committed by the time it surfaced. 🤖 Generated with [Claude Code](https://claude.com/claude-code)
JMR-dev commented 2026-08-25 03:07:23 +00:00 (Migrated from github.com)
Copy Link
Copy Source

Sharper now than when this was filed, and wider by one file.

PR #73 corrected CLAUDE.md — it no longer says instrumented tests cannot run locally; it says the opposite, with the renderer table, because tools/local-emulator/run-e2e.sh runs API 33-36 on this host. The claim in this KDoc cites CLAUDE.md as its authority:

 * Robolectric rather than the instrumented suite, deliberately. The instrumented tests
 * cannot run on the development host at all (see CLAUDE.md), and a red test nobody can
 * execute is not a loop anyone can work in.

So it is not merely stale any more — it points at a document that now states the reverse. A reader who follows the reference finds the contradiction immediately, which is the worst version of this defect: the citation makes it look verified.

It is in two files, not one. This issue names AppRootRestorationTest; the same sentence is also in app/src/test/java/org/libremediaconverter/convert/OutputPublisherStagingTest.kt. Fixing only the file named here would leave the other one asserting it, and the next reader has no way to know which is current.

grep -rln "cannot run on the development host" app/src --include='*.kt'
  app/src/test/java/org/libremediaconverter/AppRootRestorationTest.kt
  app/src/test/java/org/libremediaconverter/convert/OutputPublisherStagingTest.kt

The reasoning underneath both KDocs is still sound and should survive the correction — Robolectric genuinely is the right choice for these two tests, and "a red test nobody can execute is not a loop anyone can work in" is a good argument. What has changed is only the premise that nobody can execute them. The honest replacement says Robolectric was chosen because it runs in the same ./gradlew invocation as everything else, not because the alternative is impossible.

That command is also the acceptance check: after the fix it should return nothing.

Sharper now than when this was filed, and wider by one file. **PR #73 corrected `CLAUDE.md`** — it no longer says instrumented tests cannot run locally; it says the opposite, with the renderer table, because `tools/local-emulator/run-e2e.sh` runs API 33-36 on this host. The claim in this KDoc cites `CLAUDE.md` as its authority: ``` * Robolectric rather than the instrumented suite, deliberately. The instrumented tests * cannot run on the development host at all (see CLAUDE.md), and a red test nobody can * execute is not a loop anyone can work in. ``` So it is not merely stale any more — **it points at a document that now states the reverse.** A reader who follows the reference finds the contradiction immediately, which is the worst version of this defect: the citation makes it look verified. **It is in two files, not one.** This issue names `AppRootRestorationTest`; the same sentence is also in `app/src/test/java/org/libremediaconverter/convert/OutputPublisherStagingTest.kt`. Fixing only the file named here would leave the other one asserting it, and the next reader has no way to know which is current. ``` grep -rln "cannot run on the development host" app/src --include='*.kt' app/src/test/java/org/libremediaconverter/AppRootRestorationTest.kt app/src/test/java/org/libremediaconverter/convert/OutputPublisherStagingTest.kt ``` **The reasoning underneath both KDocs is still sound** and should survive the correction — Robolectric genuinely is the right choice for these two tests, and "a red test nobody can execute is not a loop anyone can work in" is a good argument. What has changed is only the premise that nobody can execute them. The honest replacement says Robolectric was chosen because it runs in the same `./gradlew` invocation as everything else, not because the alternative is impossible. That command is also the acceptance check: after the fix it should return nothing.
Sign in to join this conversation.
No Branch/Tag Specified
Branches Tags
Labels
Clear labels
above-cut
Worked autonomously overnight: local, JVM-verifiable, no product decision
 accessibility
Barrier affecting people with disabilities
 backlog
Held for manual review: product/UX call, CI/workflow, hardware, or unverifiable here
 bug
Something isn't working
 confirmed
Reviewer demonstrated the defect
 documentation
Improvements or additions to documentation
 duplicate
This issue or pull request already exists
 enhancement
New feature or request
 good first issue
Good for newcomers
 help wanted
Extra attention is needed
 invalid
This doesn't seem right
 plausible
Reviewer could not fully demonstrate it; treat as unproven
 question
Further information is requested
 sev:high
High severity
 sev:low
Low severity
 sev:medium
Medium severity
 wontfix
This will not be worked on
No labels above-cut confirmed documentation sev:low
Milestone
No items
No Milestone
Projects
Clear projects
No projects
Assignees
Clear assignees
JMR-dev
No Assignees
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: JMR-dev/LibreMediaConverter#46