18

What are type projections in Scala useful for? Why does Scala's type system support both type projections and path dependent types? What was the rationale behind this design decision?

tan-theta
  • 181
  • 3

2 Answers2

13

Not a complete answer, but here are some uses for type projections that I have encountered:

  • Type level metaprogramming. For examples, see Michid's series (parts I, II, III), Jesper's implementation of HList, and the series at Apocalisp.

  • A workaround to enable type inference (for examples, here are some previous SO questions 1, 2, 3).

  • A way to bundle a bunch of types into a single type parameter. For example, in a matrix library I'm developing, I define trait Scalar { type A; type B; type C; ... } and then pass it as a single parameter to my matrix trait, trait Matrix[S <: Scalar] ... The individual types can be referred to as S#A, S#B, and so on. Between two matrices of type Matrix[S], for the same S, these types will be compatible (unlike what would be the case with path dependent types).

Community
  • 1
  • 1
Kipton Barros
  • 21,002
  • 4
  • 67
  • 80
  • Jesper's post you link states that the code doesn't work, but later post explain that Scala's newer version supports it: http://jnordenberg.blogspot.com/2009/09/type-lists-and-heterogeneously-typed.html – Blaisorblade Sep 02 '11 at 20:54
9

One thing type projections can be used for is partial type application:

({type λ[x]=Tuple2[Int,x]})#λ
michid
  • 10,536
  • 3
  • 32
  • 59