Conversation

Actually, turns out I shouldn't use that on launch anyway! For now I'm abstracting the few calls to the SDK into my own object under a Swift "namespace" (an enum). This file will probably change a lot in the near future, but for now it gives me hooks to call on sign in/out.
Image
2
1
Since I'm working in a hybrid app most of the UI is rendered on the web. So I can send a JavaScript message to the client when the web-based purchase button is clicked. This Stimulus controller grabs the iOS plan identifier to pass along to RevenueCat.
Image
1
4
Up next is integrating the webhooks from RevenueCat into my Rails app. Then I can associate iOS purchases with users in the Rails database. But first… lunch!
Image
1
4
Progress progress! I now have a small Rails controller that handles authorization and kicks off a background job. Now I can unit test the actual sync logic in a PORO and not deal with integration/controller tests.
Image
2
3
Back at integrating with my Rails app! The more I think about this the more confused I get. Why is it simpler to have a single method/class do EVERYTHING when we are already getting the "diff" with each webhook?
Image
3
3