PR template — items to address before "Ready for review"

Warnings — informational, address before merging:

• Description section is empty. Describe what changed and why.
• Related issues section is empty. Add Fixes: #123 / Closes: <URL> / Refs: <Jira key>, or write a short rationale after the colon.
• Manual testing steps still contain template content (the Gherkin example title or a [...] placeholder). Replace with real steps, or write N/A — <reason>.
• Screenshots/Recordings section is empty. Add an image/video for user-facing changes, logs/console output for non-user-facing changes, or write N/A if no evidence is applicable.
• Pre-merge author checklist has unchecked items (e.g. "I've followed MetaMask Contributor Docs and MetaMask Mobile Coding Standards."). Every box must be consciously checked — see docs/readme/ready-for-review.md.

See docs/readme/ready-for-review.md for the full Definition of Ready for Review.

🧪 Flaky unit test detection

Run history flaky detection

View recent run history

Historical failure rate is a hint, not proof — review each suggestion in context. See the flaky-test-detection skill for the full pattern reference and manual audit workflow.

Failures / runs sampled per window:

AI-detected flaky patterns

app/components/Views/TrendingView/feeds/perps/usePerpsLiveMovers.test.ts

• J8 — jest.useFakeTimers() combined with waitFor (polling conflict / missing clearAllTimers) (high)
  • In every test inside the recomputeIntervalMs describe block, jest.useFakeTimers() is called mid-test (after an initial await waitFor(...) that resolves under real timers). The afterEach only calls jest.useRealTimers() but never jest.clearAllTimers(). Two risks arise:

1. Pending timer leak on test failure: If a test throws after jest.useFakeTimers() but before jest.advanceTimersByTime(...) fires the pending 10 s recompute timer, that timer remains in the fake queue. jest.useRealTimers() in afterEach switches the clock back but does NOT drain the fake queue. Depending on the Jest version, those pending callbacks can fire unexpectedly in the next test, causing order-dependent failures.

2. Mid-test timer activation: Calling jest.useFakeTimers() inside the test body (rather than in beforeEach) means the fake-timer state is not guaranteed to be clean at the start of each test. Moving the call to beforeEach (and pairing it with jest.clearAllTimers() in afterEach) makes the timer lifecycle explicit and symmetric.

The fix moves jest.useFakeTimers() to beforeEach and adds jest.clearAllTimers() before jest.useRealTimers() in afterEach. Because the initial await waitFor(...) in each test must run under real timers, the tests are restructured to call jest.useRealTimers() for that initial setup phase and then switch to fake timers — or, more cleanly, the initial subscription/mount assertion is extracted into a shared beforeEach that runs before fake timers are installed.

• Suggested fix in app/components/Views/TrendingView/feeds/perps/usePerpsLiveMovers.test.ts:340:

  -  describe('recomputeIntervalMs', () => {
  -    afterEach(() => {
  -      jest.useRealTimers();
  -    });
  -
  -    it('batches multiple ticks inside the interval into a single recompute using the freshest percents', async () => {
  -      // ...
  -      await waitFor(() => {
  -        expect(result.current.map((item) => item.market.symbol)).toEqual([
  -          'HIGH_GAINER',
  -          'LOW_GAINER',
  -        ]);
  -      });
  -
  -      jest.useFakeTimers();
  -      // ...
  -    });
  +  describe('recomputeIntervalMs', () => {
  +    beforeEach(() => {
  +      jest.useFakeTimers();
  +    });
  +
  +    afterEach(() => {
  +      jest.clearAllTimers();
  +      jest.useRealTimers();
  +    });
  +
  +    // In each test, replace the initial `await waitFor(...)` block that
  +    // previously ran under real timers with an `act`-based flush, since
  +    // fake timers are now active from the start:
  +    //
  +    //   Before:
  +    //     await waitFor(() => {
  +    //       expect(result.current.map(...)).toEqual([...]);
  +    //     });
  +    //     jest.useFakeTimers(); // ← remove this line from every test
  +    //
  +    //   After:
  +    //     await act(async () => { jest.runAllTimersAsync(); });
  +    //     expect(result.current.map(...)).toEqual([...]);
  +    //
  +    // This keeps the timer lifecycle symmetric (fake timers always active
  +    // inside the describe block) and prevents pending-timer leaks on
  +    // test failure.

• J9 — Module-level mutable let bindings not reset in beforeEach (high)
  • The module-level mockSubscribeToSymbols jest.fn() is shared across all tests. The beforeEach correctly calls jest.clearAllMocks(), which resets call counts and clears mock implementations set via mockReturnValue/mockImplementation. However, jest.clearAllMocks() does NOT restore the default implementation — it only clears call history and removes per-call return values. Tests that call mockSubscribeToSymbols.mockImplementation(...) or mockSubscribeToSymbols.mockReturnValue(...) set a persistent implementation that clearAllMocks will clear, but the inner usePerpsStream factory function (the outer jest.fn(() => ({ prices: { subscribeToSymbols: mockSubscribeToSymbols } }))) is also a jest.fn() whose implementation is set at mock-definition time and is never reset.

More concretely: usePerpsStream is a jest.fn() whose factory is defined once in jest.mock(...). If any test calls jest.mocked(usePerpsStream).mockImplementation(...) to override it, that override persists into subsequent tests because clearAllMocks only clears call counts, not implementations (that requires resetAllMocks). The current beforeEach uses clearAllMocks, which is insufficient to guard against implementation bleed if a future test overrides usePerpsStream directly.

The safest fix is to change jest.clearAllMocks() to jest.resetAllMocks() in beforeEach and re-apply the default mockSubscribeToSymbols return value there, so every test starts from a known state regardless of what the previous test did.

• Suggested fix in app/components/Views/TrendingView/feeds/perps/usePerpsLiveMovers.test.ts:13:

  -const mockSubscribeToSymbols = jest.fn();
  -
  -jest.mock('../../../../UI/Perps/providers/PerpsStreamManager', () => ({
  -  usePerpsStream: jest.fn(() => ({
  -    prices: {
  -      subscribeToSymbols: mockSubscribeToSymbols,
  -    },
  -  })),
  -}));
  +  beforeEach(() => {
  +    jest.resetAllMocks(); // resets implementations AND call counts
  +    // Re-apply the default implementation so every test starts from a
  +    // known state without relying on the module-factory default.
  +    jest.mocked(
  +      require('../../../../UI/Perps/providers/PerpsStreamManager').usePerpsStream,
  +    ).mockReturnValue({
  +      prices: {
  +        subscribeToSymbols: mockSubscribeToSymbols,
  +      },
  +    });
  +  });

app/components/Views/TrendingView/tabs/NowTab.test.tsx

• J3 — Missing jest.clearAllMocks() / jest.resetAllMocks() in beforeEach (high)
  • jest.clearAllMocks() is called inside the arrangeMocks() helper rather than in a beforeEach hook. Every describe block delegates to arrangeMocks() (via arrangePerpsMoversMocks(), arrangeCryptoMoversMocks(), etc.), so in the current test suite this works. However, this is a fragile pattern: any future test that is added without calling arrangeMocks() — or that calls it after some side-effectful setup — will inherit stale call counts and mock implementations from the previous test. Jest runs tests in file order by default, but --randomize or parallel sharding can expose the ordering dependency. The fix is to move jest.clearAllMocks() (and the default mock implementations) into a top-level beforeEach so isolation is guaranteed unconditionally.
  • Suggested fix in app/components/Views/TrendingView/tabs/NowTab.test.tsx:248:

    -const arrangeMocks = () => {
    -  jest.clearAllMocks();
    -  mockUseIsFocused.mockReturnValue(true);
    -
    -  const mockUseSelector = useSelector as jest.MockedFunction<
    -    typeof useSelector
    -  >;
    -  const mockUseTokensFeed = useTokensFeed as jest.MockedFunction<
    -    typeof useTokensFeed
    -  >;
    -
    -  mockUseSelector.mockImplementation(
    -    createMockSelectorImpl({
    -      perpsEnabled: false,
    -      predictEnabled: false,
    -      whatsHappeningEnabled: false,
    -    }),
    -  );
    -  // ... (rest of mock setup)
    -};
    +// Add a top-level beforeEach that always clears mocks and restores defaults.
    +// Keep arrangeMocks() for the per-describe overrides, but remove the
    +// jest.clearAllMocks() call from inside it.
    +
    +beforeEach(() => {
    +  jest.clearAllMocks();
    +  // Restore module-level mock defaults so every test starts from a known state.
    +  mockUseIsFocused.mockReturnValue(true);
    +  mockSubscribeToSymbols.mockReturnValue(jest.fn());
    +  mockWhatsHappeningRefresh.mockReset();
    +  mockUseWhatsHappening.mockReturnValue({
    +    items: [{ id: 'trend-0' }],
    +    isLoading: false,
    +    error: null,
    +    refresh: mockWhatsHappeningRefresh,
    +  });
    +});
    +
    +// Inside arrangeMocks(), remove the jest.clearAllMocks() line:
    +const arrangeMocks = () => {
    +  // jest.clearAllMocks(); ← remove this line
    +  const mockUseSelector = useSelector as jest.MockedFunction<typeof useSelector>;
    +  // ... rest of per-describe setup unchanged
    +};

• J9 — Module-level mutable let bindings not reset in beforeEach (high)
  • All module-level mock functions (mockNavigate, mockUseIsFocused, mockSubscribeToSymbols, mockWhatsHappeningRefresh, mockUseWhatsHappening, mockNavigateToPerpsMarketList, mockPredictionsCarouselSection) are declared at module scope and their call history / implementations are only reset inside arrangeMocks(), which is called manually per test. If a test is skipped, fails early, or is added without calling arrangeMocks(), the accumulated call counts and overridden implementations from the previous test bleed into the next one. For example, mockUseIsFocused is set to false in one 'live movers subscription gating' test and is never explicitly restored to true unless arrangeMocks() is called next. Moving the reset to beforeEach eliminates this ordering dependency.
  • Suggested fix in app/components/Views/TrendingView/tabs/NowTab.test.tsx:5:

    -const mockNavigate = jest.fn();
    -const mockUseIsFocused = jest.fn(() => true);
    -
    -// ...
    -
    -const mockSubscribeToSymbols = jest.fn(() => jest.fn());
    -
    -// ...
    -
    -const mockWhatsHappeningRefresh = jest.fn();
    -const mockUseWhatsHappening = jest.fn(() => ({
    -  items: [] as { id: string }[],
    -  isLoading: false,
    -  error: null as string | null,
    -  refresh: mockWhatsHappeningRefresh,
    -}));
    -
    -// ...
    -
    -const mockNavigateToPerpsMarketList = jest.fn();
    -const mockPredictionsCarouselSection = jest.fn(/* ... */);
    +// At the top level of the describe block (or file), add:
    +beforeEach(() => {
    +  jest.clearAllMocks();
    +  // Re-apply default implementations for every module-level mock:
    +  mockUseIsFocused.mockReturnValue(true);
    +  mockSubscribeToSymbols.mockImplementation(() => jest.fn());
    +  mockUsePerpsFeed.mockReturnValue({
    +    data: [],
    +    isLoading: false,
    +    refetch: jest.fn(),
    +    defaultSortOptionId: 'priceChange' as const,
    +  });
    +  mockUseWhatsHappening.mockReturnValue({
    +    items: [],
    +    isLoading: false,
    +    error: null,
    +    refresh: mockWhatsHappeningRefresh,
    +  });
    +});

This check is informational only and does not block merging.

Codecov Report

❌ Patch coverage is 94.80519% with 4 lines in your changes missing coverage. Please review.
✅ Project coverage is 84.32%. Comparing base (9aad058) to head (eb320cb).
⚠️ Report is 5 commits behind head on main.

@@            Coverage Diff             @@
##             main   #33150      +/-   ##
==========================================
+ Coverage   84.31%   84.32%   +0.01%     
==========================================
  Files        6080     6084       +4     
  Lines      162378   162522     +144     
  Branches    39577    39616      +39     
==========================================
+ Hits       136905   137050     +145     
+ Misses      16021    16010      -11     
- Partials     9452     9462      +10     

☔ View full report in Codecov by Harness.
📢 Have feedback on the report? Share it here.

🔍 Smart E2E Test Selection

• Selected E2E tags: SmokeWalletPlatform, SmokePerps, SmokeConfirmations
• Selected Performance tags: @PerformancePreps
• Risk Level: medium
• AI Confidence: 85%

View GitHub Actions results

Quality Gate passed

Issues
0 New issues
0 Accepted issues

Measures
0 Security Hotspots
97.1% Coverage on New Code
0.0% Duplication on New Code

See analysis details on SonarQube Cloud

⚡ Performance Test Results

ℹ️ Performance test results are currently non-blocking and will not block this PR.

✅ All tests passed · 2 tests · 1 device

Branch: fix/oversubscription-perps-top-movers · Build: Normal · Commit: e2207c2 · View full run