Hello and happy new year! :slightly_smiling_face: If i need to synchronise this calls: ```container...

# orbit-mvi

[Guilherme Delgado]

Hello and happy new year! 🙂
If i need to synchronise this calls:

containerHost.intent {
        reduce { state.copy(...) }
        reduce { state.copy(...) }
    }
}

is there any way I can assure that the second reduce happens after the first one?

Use case: I’m doing an “undo” swipe to dismiss. The androidx.compose.material.SwipeToDismiss removes/hides the item from the UI (LazyColumn) before removing it from the UIState.
Therefore we need to simulate that removal from the UIState by emitting a new one without the item, and then, we emit a new state again with the item back in the list.
My question appears because I can only succeed if I do something like:

fun undoDelete(id: Long) {
    containerHost.intent {       
         val last = state.clients.toList()
         reduce { state.copy(clients = state.clients.filterNot { it.id == id }.toList()) }
         delay(50)
         reduce { state.copy(clients = last) }
    }
}

but it feels like 🔨 😅

[Mikolaj Leszczynski]

I'm not sure I follow... 🤔

Generally if you reduce like the above it should be in sequence, as it all runs within the same coroutine.

Is this a case of state updates happening so quickly they get conflated by the StateFlow?

[Mikolaj Leszczynski]

PS for UI-bound intents ideally you should use blockingIntent

[Guilherme Delgado]

maybe, because the delay does the trick

[Mikolaj Leszczynski]

i.e. if you don't do any background work at all

[Guilherme Delgado]

I’l try

[Mikolaj Leszczynski]

yeah try blockingIntent, if it doesn't help then it's likely the conflation that's the problem

[Mikolaj Leszczynski]

I'm not sure we can do anything about it other than what you already did

[Guilherme Delgado]

btw, how can I have access to blockingIntent ?

[Guilherme Delgado]

containerHost.blockingIntent ?

[Guilherme Delgado]

can’t find it 🤔

[Mikolaj Leszczynski]

Are you on the latest build?

[Guilherme Delgado]

orbitMviCore = "4.4.0"

[Guilherme Delgado]

gonna check it if there’s a new one 🙂

[Guilherme Delgado]

is it 4.5.0 ?

[Guilherme Delgado]

(yes it is ^^)

[Guilherme Delgado]

the problem is with the conflation i guess, blockingIntent by itself didn’t solved

[Mikolaj Leszczynski]

Unless you can change your logic then, I think the delay is acceptable..