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

R7 — A present, willing user has no way to make a foreground-denied retry happen now #16

New Issue
Open
opened 2026-08-23 03:44:44 +00:00 by JMR-dev · 0 comments
JMR-dev commented 2026-08-23 03:44:44 +00:00 (Migrated from github.com)
Copy Link
Copy Source

Finding R7 from the overnight max-effort review (Fable lead, Opus sub-agents). Full report: scratchpad/overnight/REVIEW.md.

R7 — A present, willing user has no way to make a foreground-denied retry happen now

severity: medium
verdict: CONFIRMED
where: app/src/main/java/org/libremediaconverter/work/FailureOutcome.kt:56-59; ConverterScreen.kt:171-188; JoinScreen.kt:119-133
scenario: Transcode interrupted by process death at 23:00; every WorkManager recovery is denied at setForeground (app in background); backoff doubles to 4h16m. User opens the app at 09:00 (runAttemptCount 9): screen says "Paused ... keeping the app open helps it along" and offers only Cancel. The next attempt may be hours away; the one thing that would fix the job — the user's foreground presence — cannot be spent on it. Close the app first and the job hits the bound and dies silently (see D16 note below).
evidence: Lead re-ran the greps: zero hits for ProcessLifecycleOwner/observers/enqueueUniqueWork/setBackoffCriteria in app/src/main; Waiting branch renders only Cancel. Reviewer verified against work-runtime 2.11.2 sources that no supported "run this one-time job now" API exists (setNextScheduleTimeOverride is periodic-only, WorkSpec.kt:423-434), so a fix must cancel + re-enqueue.
fix: "Try now" on the Waiting state: cancel old, enqueue equivalent, observe new. Must decide (a) reattached jobs have no recoverable input Uri (JobTags deliberately drops it) — offer only for jobs this ViewModel started, or carry the URI and own the dead-grant risk; (b) re-enqueue resets runAttemptCount, granting a fresh budget of 10 — defensible (the user asked) but must be decided, not inherited.
risk: Cancel must complete before enqueue or two live jobs share NOTIFICATION_ID 1001. Dispatch timing is not JVM-verifiable; the ViewModel contract (cancel->enqueue->observe, activeWorkId moves once) is, via WorkManagerTestInitHelper.

Cut: below — held for manual review: product decision, CI/workflow, hardware, or PLAUSIBLE verdict.

🤖 Generated with Claude Code

_Finding **R7** from the overnight max-effort review (Fable lead, Opus sub-agents). Full report: `scratchpad/overnight/REVIEW.md`._ ### R7 — A present, willing user has no way to make a foreground-denied retry happen now severity: medium verdict: CONFIRMED where: app/src/main/java/org/libremediaconverter/work/FailureOutcome.kt:56-59; ConverterScreen.kt:171-188; JoinScreen.kt:119-133 scenario: Transcode interrupted by process death at 23:00; every WorkManager recovery is denied at setForeground (app in background); backoff doubles to 4h16m. User opens the app at 09:00 (runAttemptCount 9): screen says "Paused ... keeping the app open helps it along" and offers only Cancel. The next attempt may be hours away; the one thing that would fix the job — the user's foreground presence — cannot be spent on it. Close the app first and the job hits the bound and dies silently (see D16 note below). evidence: Lead re-ran the greps: zero hits for ProcessLifecycleOwner/observers/enqueueUniqueWork/setBackoffCriteria in app/src/main; Waiting branch renders only Cancel. Reviewer verified against work-runtime 2.11.2 sources that no supported "run this one-time job now" API exists (setNextScheduleTimeOverride is periodic-only, WorkSpec.kt:423-434), so a fix must cancel + re-enqueue. fix: "Try now" on the Waiting state: cancel old, enqueue equivalent, observe new. Must decide (a) reattached jobs have no recoverable input Uri (JobTags deliberately drops it) — offer only for jobs this ViewModel started, or carry the URI and own the dead-grant risk; (b) re-enqueue resets runAttemptCount, granting a fresh budget of 10 — defensible (the user asked) but must be decided, not inherited. risk: Cancel must complete before enqueue or two live jobs share NOTIFICATION_ID 1001. Dispatch timing is not JVM-verifiable; the ViewModel contract (cancel->enqueue->observe, activeWorkId moves once) is, via WorkManagerTestInitHelper. --- **Cut:** `below` — held for manual review: product decision, CI/workflow, hardware, or PLAUSIBLE verdict. 🤖 Generated with [Claude Code](https://claude.com/claude-code)
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 backlog bug confirmed sev:medium
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#16