test(ios): test iOS in release mode in addition to debug

[mikehardy]

Description

This PR adds a release mode check to our CI to reproduce a functions and firestore crash that shows in release mode only on iOS

The actual issue is/was upstream, and releases are rolling out now, but the issue we can respond to here is that this showed up in release mode easily but we never saw it since we were not testing release mode.

The main change here is just to add release mode build/test scripts for iOS, then augment the iOS CI check to run a buildmode matrix

Related issues

• Related 🔥 Functions / Firestore iOS Release Mode crash, w/react-native 0.77.1, 0.78.0 (solution: use react-native 0.78.1, or apply patch, or disable new arch) #8369

Release Summary

No release will be triggered by any commits in this PR, the actual fix was upstream and I filed pick requests that have already been integrated for future upstream releases

Checklist

• I read the Contributor Guide and followed the process outlined there for submitting PRs.
  • Yes
• My change supports the following platforms;
  • Android
  • iOS
  • Other (macOS, web)
• My change includes tests;
  • e2e tests added or updated in packages/\*\*/e2e
  • jest tests added or updated in packages/\*\*/__tests__
• I have updated TypeScript types that are affected by my change.
• This is a breaking change;
  • Yes
  • No

Test Plan

With the commit on this PR you can run these and see the crash, until the upstream patch is integrated in our e2e app

yarn test:ios:build:release
yarn test:ios:test:release

---

Think react-native-firebase is great? Please consider supporting the project with any of the below:

• 👉 Star this repo on GitHub ⭐️
• 👉 Follow React Native Firebase and Invertase on Twitter

[vercel]

The latest updates on your projects. Learn more about Vercel for Git ↗︎

Name	Status	Preview	Comments	Updated (UTC)
react-native-firebase	✅ Ready (Inspect)	Visit Preview	💬 Add feedback	Mar 21, 2025 4:33pm

[mikehardy]

Perfect! Before adding the last commit that includes the upstream patch that fixes the issue, I wanted to make sure it reproduced the crash in testing since that is now the whole point of this PR. It did, here's what happened in the iOS release mode test:

  error: The app has crashed, see the details below:
  
  Exception in HostFunction: -[NSNull length]: unrecognized selector sent to instance 0x1e4b8f140
  
  Error: Exception in HostFunction: -[NSNull length]: unrecognized selector sent to instance 0x1e4b8f140
  (
  	0   Detox                               0x0000000103452514 +[NSThread(DetoxUtils) dtx_demangledCallStackSymbols] + 36
  	1   Detox                               0x0000000103455520 __DTXHandleCrash + 668
  	2   Detox                               0x0000000103456048 __dtx_terminate() + 376
  	3   libc++abi.dylib                     0x00000001802a6c50 std::__terminate(void (*)()) + 12
  	4   libc++abi.dylib                     0x00000001802a6c00 std::terminate() + 52
  	5   libdispatch.dylib                   0x000000018017198c _dispatch_client_callout + 36
  	6   libdispatch.dylib                   0x0000000180179b10 _dispatch_lane_serial_drain + 960
  	7   libdispatch.dylib                   0x000000018017a688 _dispatch_lane_invoke + 388
  	8   libdispatch.dylib                   0x0000000180185a84 _dispatch_root_queue_drain_deferred_wlh + 276
  	9   libdispatch.dylib                   0x00000001801850d0 _dispatch_workloop_worker_thread + 448
  	10  libsystem_pthread.dylib             0x0000000102c07b74 _pthread_wqthread + 284
  	11  libsystem_pthread.dylib             0x0000000102c06934 start_wqthread + 8
  )

So these won't slip through again

[mikehardy]

iOS release mode e2e appears to have become stuck:

Thu, 20 Mar 2025 14:24:21 GMT
          ✔ resolves a started instance of a ScreenTrace
Thu, 20 Mar 2025 14:39:38 GMT
Error: The operation was canceled.

Going to check logs and see if I can determine why (was it a crash?) while re-running (was it a flake?)

edit: unable to check logs, artifacts were uploaded without matrix name in them so they collided and release mode artifact wouldn't upload, no simulator log to check. Re-running now with a fix for that so we can inspect in future

[mikehardy]

Test failure is sendSignInLinkToEmail - instrumented with RNFBTest searchable string in Simulator log for future runs here

Then a secondary is that during the re-run it hangs on test after ✔ resolves a started instance of a ScreenTrace - that appears to be the final test in the perf.e2e.js suite and was successful, next suite is messaging.e2e.js and I'm not sure why it didn't work as there was no simulator log uploaded yet due to an error in github workflow yaml, hopefully next Simulator.log has more info