How would you model a state emited from viewmodel/presenter, if you need to render a "Hello Joe"? `...

# compose

[ursus]

How would you model a state emited from viewmodel/presenter, if you need to render a "Hello Joe"?

State(val greeting: String) and just render thr greeting

State(val name: String) render "Hello $state.name"

State(val user: User) render "Hello $state.user.name"

TLDR; do you format in viewmodel or in composables?

[alorma]

I ALWAYS format on ViewModel / Presenter.

UI must be responsible of displaying content, but not define which content to show.

Think of it about dates... would you format the date on UI?

[ursus]

Currently I do that yea. Trouble is that I often need to keep the User instance in state for some other parts to read it, like user.id and what not

And this way if timestamp: Long is part of your presenter state, formatted value doesnt really needs to be a backed val, since its a calculation, giving u a 1 source of truth

[alorma]

yes, long, or DateTime is logic.. but formatted date is presentation layer..

[ursus]

you dont consider viewmodel a presentation layer?

[alorma]

it's a layer between domain and presentation.

it receive data form UseCases / dataspurces... but domain data, Numbers DateTime boolean enum ... and it maps them to the presentation data, strings with params, strings based on given param, ...