Skip to content

Fix stall detection to allow failure detection on first working rep #260

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
9thLevelSoftware merged 1 commit into from
Mar 5, 2026
Merged

Fix stall detection to allow failure detection on first working rep #260

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
14 changes: 7 additions & 7 deletions ...rc/commonMain/kotlin/com/devil/phoenixproject/presentation/manager/ActiveSessionEngine.kt
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -540,18 +540,18 @@ class ActiveSessionEngine(

/**
* Standard set stall guard:
* - Ignore fail/stall detection before first confirmed working rep.
* - Defer stall detection only when no working reps are confirmed AND no rep is pending.
*
* Issue #256: Removed hasPendingRep guard. Previously, stall detection was deferred
* while a rep was pending at TOP (to avoid false triggers during brief lockout pauses).
* However, a stalled pending rep IS the failure scenario (e.g., failed bench press).
* The velocity hysteresis band (STALL_VELOCITY_LOW=2.5, STALL_VELOCITY_HIGH=10.0 mm/s)
* already provides adequate protection against false triggers during brief pauses.
* Issue #256: Removed hasPendingRep guard from deferral. A stalled pending rep IS the
* failure scenario (e.g., failed bench press at TOP of first rep). The velocity hysteresis
* band (STALL_VELOCITY_LOW=2.5, STALL_VELOCITY_HIGH=10.0 mm/s) already provides adequate
* protection against false triggers during brief pauses. When workingReps == 0 but
* hasPendingRep is true, the user is mid-first-rep and must be protected.
*/
private fun shouldDeferStandardSetStall(params: WorkoutParameters, repCount: RepCount): Boolean {
val isStandardSet = !params.isJustLift && !params.isAMRAP && !coordinator.isCurrentTimedCableExercise
if (!isStandardSet) return false
return repCount.workingReps == 0
return repCount.workingReps == 0 && !repCount.hasPendingRep
}
Comment on lines +543 to 555

Copilot AI Mar 5, 2026

Copy link

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

The KDoc contradicts the implementation: it states the hasPendingRep guard was removed from deferral, but the updated logic explicitly uses !repCount.hasPendingRep to decide deferral. Please rewrite the Issue #256 note to reflect the actual change (i.e., deferral was narrowed to only when no rep is pending), and clarify what 'must be protected' means here (protected from deferral vs. protected from false positives) to avoid confusion for future maintainers.

Copilot uses AI. Check for mistakes.

/**
Expand Down
22 changes: 10 additions & 12 deletions ...mmonTest/kotlin/com/devil/phoenixproject/presentation/manager/DWSMWorkoutLifecycleTest.kt
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -510,21 +510,20 @@ class DWSMWorkoutLifecycleTest {
advanceUntilIdle()
assertIs<WorkoutState.Active>(harness.dwsm.coordinator.workoutState.value)

// Complete warmup and first working rep
// Complete warmup
completeWarmupReps(harness, warmupTarget = 3, workingTarget = 8)
completeFirstWorkingRep(harness, warmupTarget = 3, workingTarget = 8)
advanceUntilIdle()
assertTrue(harness.dwsm.coordinator.repCount.value.isWarmupComplete)
assertEquals(1, harness.dwsm.coordinator.repCount.value.workingReps)
assertEquals(0, harness.dwsm.coordinator.repCount.value.workingReps)

// Simulate starting a second rep (pending at TOP = failed bench press scenario)
// Simulate starting the first rep (pending at TOP = failed bench press scenario)
harness.fakeBleRepo.emitRepNotification(
RepNotification(
topCounter = 5, // warmup(3) + working(2) = 5th up counter
completeCounter = 4, // Only 4 downs (second working rep not completed)
topCounter = 4, // warmup(3) + working(1) = 4th up counter
completeCounter = 3, // Only 3 downs (first working rep not completed)
repsRomCount = 3,
repsRomTotal = 3,
repsSetCount = 1, // Still 1 completed working rep
repsSetCount = 0, // Still 0 completed working reps
repsSetTotal = 8,
rangeTop = 800f,
rangeBottom = 0f,
Expand Down Expand Up @@ -571,17 +570,16 @@ class DWSMWorkoutLifecycleTest {
assertIs<WorkoutState.Active>(harness.dwsm.coordinator.workoutState.value)

completeWarmupReps(harness, warmupTarget = 3, workingTarget = 8)
completeFirstWorkingRep(harness, warmupTarget = 3, workingTarget = 8)
advanceUntilIdle()

// Simulate pending rep (stalled mid-concentric)
// Simulate pending first rep (stalled mid-concentric)
harness.fakeBleRepo.emitRepNotification(
RepNotification(
topCounter = 5,
completeCounter = 4,
topCounter = 4,
completeCounter = 3,
repsRomCount = 3,
repsRomTotal = 3,
repsSetCount = 1,
repsSetCount = 0,
Comment on lines +575 to +582

Copilot AI Mar 5, 2026

Copy link

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

The test comment says 'stalled mid-concentric' but the event fields use topCounter/completeCounter values that appear to represent a rep that has reached TOP but is not completed (pending at TOP). Please align the comment with what the counters represent in this test scenario (or adjust the counters to reflect a true mid-concentric stall if that’s the intent) so the test remains self-explanatory.

Copilot uses AI. Check for mistakes.
repsSetTotal = 8,
rangeTop = 800f,
rangeBottom = 0f,
Expand Down