in ran into an issue with default arguments ```my...

# getting-started

[tipsy]

in ran into an issue with default arguments

myFunc(p1: String, p2: Consumer<*>, p3: String = "") {}
// is not the same as 
myFunc(p1: String, p2: Consumer<*>) { myFunc(p1, p2, "") }
myFunc(p1: String, p2: Consumer<*>, p3: String) {}

the compiler does not allow you to use trailing lambdas if you use default arguments

i'm not really sure if it's an issue, but did seem odd to me

[nanodeath]

I'm confused by the code snippet. why does only one have a body?

[tipsy]

the body is irrelevant, but i added curly braces now

[nanodeath]

what if you put p2 last?

[tipsy]

unless i declare myFunc as two separate functions, i'm not able to call

myFunc("test") { }

[tipsy]

that will work

[nanodeath]

trailing lambdas always require the lambda be the last argument tho

[tipsy]

it is the last argument, since the third argument is omitted (it has a default value)

[tipsy]

i don't see why the compiler cares if it's defined as two different functions, or one function with a default value

[nanodeath]

if there's an argument after it, then it's not the last argument

[nanodeath]

default arguments don't completely obviate the need for overloads unfortunately

[tipsy]

it would seem so

[tipsy]

just a small gotcha after auto-converting some java code

[Ruckus]

I don't understand your issue with putting the consmer last. It works fine for me: https://pl.kotl.in/B_F3K_Auz

typealias Consumer<T> = (T) -> Unit

fun myFunc(p1: String, p2: String = "", p3: Consumer<String>) {
    p3(p1 + p2)
}

fun main() {
    myFunc("Test") { println(it) }
    myFunc("Test", "2") { println(it) }
}

[Ruckus]

(You may have a problem using Consumer<*> as that will accept only Nothing.)

[nanodeath]

it's not always an option to change the signature 🙂

[Ruckus]

Yes, that is true, but unless I missed something the discussion was:

Max Aller: what if you put p2 last?

david: unless i declare myFunc as two separate functions, i'm not able to call

myFunc("test") { }

[Ruckus]

I guess this is more of a discussion about what "last argument" means, and I'm just fixating on the wrong thing. Feel free to ignore me 🙂

[nanodeath]

it's understandable that it's a point of confusion 🙂

[tipsy]

@Ruckus the two lines you quote was not a question+response, it was two messages that were sent at the same time. a more accurate representation of the conversation would be

Max Aller: what if you put p2 last?

david: that will work

[tipsy]

but i can't change the signature

[nanodeath]

you could add an extension method with the arguments in the correct order perhaps

[Ruckus]

Ah, got it. My bad.