condition-based-waiting

Condition-based waiting eliminates flaky tests by polling for specific conditions instead of using arbitrary sleeps. Replace explicit timeouts with a waitFor mechanism that checks your desired state, event, or count until it occurs or a timeout threshold is reached, reducing race condition failures.

Fixed timeouts cause flakiness because they assume work completes in a predetermined window, leading to failures on slower runs. Polling checks whether your condition is met repeatedly, adapting to actual execution speed and eliminating timing-dependent failures.

Yes. Condition-based waiting supports multiple scenarios: waiting for events to fire, state transitions to occur, event counts to reach a threshold, and file existence checks. Helper functions like waitForEvent, waitForEventCount, and waitForEventMatch cover these common patterns.

Flakiness tracking monitors test reliability metrics as you replace timeouts with polling. Integration with a Shannon-based flakiness framework lets you measure stability improvements and identify remaining race conditions in your test suite.

No prior framework dependency is required. The waitFor polling mechanism works with any test suite experiencing race conditions or asynchronous work. Start by replacing sleeps with condition-based waits in your existing tests.