I have some formatting defined in common code between ios and android target and just noticed that t...

# multiplatform

[Michal Klimczak]

I have some formatting defined in common code between ios and android target and just noticed that they treat special chars differently which messes my impl. I can't even properly paste it into slack, so I'll use screenshots to make it clearer, but take twenti̩ as example. This i̩ with a little dash at the bottom is treated by kotlin as two separate chars, but swift treats it as one. So when I defined formatting which spans across the whole word in shared code with text.length, it calculates it as 7 chars. When I use it in Swift, which thinks of it as 6 chars (via *let* characters = Array(text)), I get Fatal error: Index out of range. Any ideas how to work with these?

[Michal Klimczak]

it has something to do with how both systems treat utf16 but I have no idea yet how to approach this

[russhwolf]

Try text.codePointAt(index) and text.codePointCount(), but note they're JVM-only

[Michal Klimczak]

iunfortunately it returns the same thing:

println("twenti̩: ${text.codePointCount(0, text.length)} | ${text.length}")

//twenti̩: 7 | 7

[Michal Klimczak]

For anyone wondering - this helped https://kotlinlang.slack.com/archives/C0922A726/p1627401848232700. The problem is that kotlin represents strings with utf16 and swift with so called 21-bit unicode scalar value.

Unfortunately I didn't find any better solution than to just create an interface

interface TextLengthProvider {
    fun length(text: String) : Int
}

and implement it on both platforms (i.e. delegate completely to the swift impl). Could not have used expect/actual, because we're dealing with pure swift here.