Why Kotlin has "postfix" syntax for type declaration?

1

For example I has class CoolMessageFactory. In some method I want to declare variable of this type.

In Java (with Idea) I can type:
  CMF <enter> <space> <enter>
And I get
  CoolMessageFactory coolMessageFactory

But in  Kotlin I should type:
  val  coolMessageFactory:CMF<enter>

Too long even if I omit last “CMF<enter>”.

Why Kotlin has “postfix” syntax for type declaration?

2

The main reasons for this kind of syntax lie in the area of the crucial new features we have added: type inference and higher-order funcitons. The Java-style syntax is pretty hostile to these features: they just don't fit in. Also, we find it easier to read the name of the entity before the type, and the code is read much more than written.

I understand your concern about the IDE experience, and we will address it on the IDE level: the idea is that you say

val cMF <Ctrl+Space> <Enter> = CMF()

and get

val coolMessageFactory = CoolMessageFactory()

3

If you're assigning a value you can use the introduce variable refactoring:

CMF <enter> -> CoolMessageFactory()

introduce variable -> val coolMessageFactory = CoolMessageFactory()

I use this all the time in Java.

If you’re not assigning a value it should also be possible to use live templates. e.g.

var $NAME$: $TYPE$

with the variables in the order TYPE, NAME.

That means the type of the variable would be known before the name and IntelliJ could use the type to suggest a name. Unfortunately it doesn’t work at the moment.

http://youtrack.jetbrains.com/issue/KT-2024

Chris