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

Swapping the navigation rail and the bottom bar passes the whole suite, and Medium is in no test at all #173

New Issue
Closed
opened 2026-09-02 02:15:06 +00:00 by JMR-dev · 0 comments
JMR-dev commented 2026-09-02 02:15:06 +00:00 (Migrated from github.com)
Copy Link
Copy Source

The adaptive shell: both branches run, nothing checks which one rendered

MainActivity.kt:113-149 (useRail, the NavigationRail-in-a-Row vs
Scaffold(bottomBar = { NavigationBar { … } })) and :154-159 (Content).

This is an assertion gap, not a coverage gap, and that is the point. JaCoCo is green on the
layout branches — AppRootRestorationTest drives Compact and Expanded — but
grep -rn 'NavigationRail\|NavigationBar\|useRail' app/src/test app/src/androidTest returns
nothing. Swapping the two bodies passes the entire suite. This is the shape CLAUDE.md records
from the mutation review: green is not evidence.

Two more facts in the same place:

  • WindowWidthSizeClass.Medium appears in no test at all. The code folds it into the rail
    branch via != Compact; changing that to == Expanded would silently break tablet and foldable
    layouts with nothing failing.
  • Content is 0% (MainActivityKt.Content, 6 lines, 3 of 3 branches). Every existing test
    passes a stub content lambda, so the CONVERT→ConverterScreen / JOIN→JoinScreen mapping is
    never rendered by anything.

This is what AppRoot's own KDoc (:85-96) says the shell exists for: "it has to lay out properly
at every width", citing targetSdk 37's forced resizing.

Feasibility: checked, and it works

Rendering AppRoot with its default content reaches ConverterScreen's
viewModel: ConversionViewModel = viewModel(), which needs a ViewModelStoreOwner — and every
existing test uses createComposeRule, which supplies none.

Measured on main before filing: createAndroidComposeRule<ComponentActivity>() works under
Robolectric here. A spike rendering AppRoot(WindowWidthSizeClass.Compact) with default content
found converter.chooseFile, clicked the Join tab, found join.chooseFiles, and passed in 10.4s —
then went red when Content's two arms were transposed. So Content is in scope; no seam is
needed.

Two things make it work, both already true: both ViewModels are
@JvmOverloads constructor(app: Application, …) with defaults, so AndroidViewModelFactory builds
them; and app/build.gradle.kts states that ui-test-manifest's debugImplementation entry
already supplies the ComponentActivity to the merged manifest the unit tests build against.

Use the v2 rule (androidx.compose.ui.test.junit4.v2.createAndroidComposeRule) to match the
repo's existing v2 createComposeRule; the non-v2 one compiles with a deprecation warning. The
test needs installTestWorkManager and the ConversionDependencies fakes in @Before, as
ScreenWiringTest does.

Acceptance: three mutations, all must go red

  1. Swap the NavigationRail and NavigationBar bodies.
  2. Change widthSizeClass != WindowWidthSizeClass.Compact to == WindowWidthSizeClass.Expanded
    (this is what the Medium case is for).
  3. Transpose Content's two arms.
## The adaptive shell: both branches run, nothing checks which one rendered `MainActivity.kt:113-149` (`useRail`, the `NavigationRail`-in-a-`Row` vs `Scaffold(bottomBar = { NavigationBar { … } })`) and `:154-159` (`Content`). This is an **assertion** gap, not a coverage gap, and that is the point. JaCoCo is green on the layout branches — `AppRootRestorationTest` drives `Compact` and `Expanded` — but `grep -rn 'NavigationRail\|NavigationBar\|useRail' app/src/test app/src/androidTest` returns nothing. **Swapping the two bodies passes the entire suite.** This is the shape `CLAUDE.md` records from the mutation review: green is not evidence. Two more facts in the same place: - **`WindowWidthSizeClass.Medium` appears in no test at all.** The code folds it into the rail branch via `!= Compact`; changing that to `== Expanded` would silently break tablet and foldable layouts with nothing failing. - **`Content` is 0%** (`MainActivityKt.Content`, 6 lines, 3 of 3 branches). Every existing test passes a stub `content` lambda, so the CONVERT→`ConverterScreen` / JOIN→`JoinScreen` mapping is never rendered by anything. This is what `AppRoot`'s own KDoc (`:85-96`) says the shell exists for: *"it has to lay out properly at every width"*, citing targetSdk 37's forced resizing. ## Feasibility: checked, and it works Rendering `AppRoot` with its **default** content reaches `ConverterScreen`'s `viewModel: ConversionViewModel = viewModel()`, which needs a `ViewModelStoreOwner` — and every existing test uses `createComposeRule`, which supplies none. Measured on `main` before filing: `createAndroidComposeRule<ComponentActivity>()` works under Robolectric here. A spike rendering `AppRoot(WindowWidthSizeClass.Compact)` with default content found `converter.chooseFile`, clicked the Join tab, found `join.chooseFiles`, and passed in 10.4s — then went red when `Content`'s two arms were transposed. So `Content` is **in scope**; no seam is needed. Two things make it work, both already true: both ViewModels are `@JvmOverloads constructor(app: Application, …)` with defaults, so `AndroidViewModelFactory` builds them; and `app/build.gradle.kts` states that `ui-test-manifest`'s `debugImplementation` entry already supplies the `ComponentActivity` to the merged manifest the unit tests build against. Use the **v2** rule (`androidx.compose.ui.test.junit4.v2.createAndroidComposeRule`) to match the repo's existing v2 `createComposeRule`; the non-v2 one compiles with a deprecation warning. The test needs `installTestWorkManager` and the `ConversionDependencies` fakes in `@Before`, as `ScreenWiringTest` does. ## Acceptance: three mutations, all must go red 1. Swap the `NavigationRail` and `NavigationBar` bodies. 2. Change `widthSizeClass != WindowWidthSizeClass.Compact` to `== WindowWidthSizeClass.Expanded` (this is what the `Medium` case is for). 3. Transpose `Content`'s two arms.
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
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#173