That did it! I... never would've guessed that, from the error.
greay
Users receive reputation points by contributing quality content to the forums.
Posts include post and replies published on the forums.
Post authors can mark replies as Accepted to indicate successful solutions.
Apple Developer Forums admins can mark replies as Apple Recommended to indicate an approved solution
Last seen
User for
Post
Replies
Boosts
Views
Activity
Hey, greay, did you ever find a solution for your problem? I'm having a similar problem with an app I'm working on where I have a one-and-done binding that functions as expected the first time, but then doesn't fire again upon changes. Any guidance would be greatly appreciated. :) No, I ended up moving completely away from SwiftUI because I need to support older versions of macOS.
I should add/clarify that both of these apps are already for sale on the store. We are NOT trying to change anything under "App Information" – like the localizable name. We want that to stay as-is. All we're trying to do now is add localized copy for the stuff in the update: What's new, promotional text, description, etc.
Hmm, ok, that suggests to me that the problem isn't in this construction itself, but something further up.As I hinted, but should've made more explicit, the [Check] is a property on another struct. The larger structure has some pretty deeply-nested subviews, too much to put in a forum post, so I was hoping this small snippet would be enough to highlight what was going wrong. It wasn't – but since this works, I'm thinking I might be able to slowly grow out from this & poinpoint the problem.
It's in an existing application, so the change actually breaks some functionality.In the app on the store, certain on-screen buttons are disabled when the keyboard appears. They're re-enabled when you press submit, dismissing the keyboard. Since the notifications are no longer sent, those buttons never get re-enabled.Basically, the text entry is a modal operation, and the existing code relies on those notifications to exit that mode. Might not be the best design, but I didn't write the original code.
But it's not always visible. I would still expect it to send the will/did show notifications when it appears. And send the will/did hide notifications when it's dismissed. If this is in fact by design, it should be documented.Having to check whether the starting/ending frame's bounds is zero seems incredibly backwards to me.