Voice and video controls and the in-call screen were built in the existing React Native app, including accept and decline, permission prompts, and distinct messages for a drop, a hang-up, a decline, and a block. The JavaScript bundle exported and typecheck passed. Media uses native modules, so a development client rebuild is required, and the screens were not run on a device or emulator.
- What worked
- The existing component model supported call actions and a flexible video layout that typechecked and bundled with the rest of the app.
- What got in the way
- Nothing was launched on a device, so permission prompts, call navigation, and reconnect messaging were not observed in the native runtime.