> On Apr 24, 2016, at 9:05 AM, Daniel Karch <danielka...@gmail.com> wrote: > > Hi, > > I recently started learning Racket and like it so far. Since I very much > prefer statically typed languages, I am leaning towards Typed Racket. > > In a small program I am writing I would like to have a function that accepts > a value of an abstract type and then calls a function on that type whose > implementation depends on the concrete type. I.e., what would be an interface > in Java with several classes implementing it or a type class in Haskell with > several instances. > In vanilla Racket it seems that I could use generic interfaces or class-based > interfaces, but in TR neither of these options seems to work. The following > works: > > #lang typed/racket > > ;; "Interface" > (define-type X > (Object [get-name (-> String)])) > > ;; Implementation 1 > (define A% > (class object% > (super-new) > (init-field [name : String]) > (: get-name (-> String)) > (define/public (get-name) > (~a "I am an A% and my name is " name ".")))) > > ;; Implementation 2 > (define B% > (class object% > (super-new) > (init-field [name : String]) > (: get-name (-> String)) > (define/public (get-name) > (~a "I am a B% and my name is " name ".")))) > > (: display-name (-> X Void)) > (define (display-name x) > (displayln (send x get-name))) > > (display-name (make-object A% "Alice")) ;; -> I am an A and my name is Alice. > (display-name (make-object B% "Bob")) ;; -> I am a B and my name is Bob. > > – but I am not completely happy about it. Is there an idiomatic way to > achieve what I described in TR?
Typed Racket does not support ML's abstract types. The goal is primarily to allow Racket programmers to port existing programs to the typed world in an incremental fashion. While I consider the existing system usable, there are many open questions and, frankly, some of them are open research questions, including how to add expressive power to Typed Racket’s type language that works with Racket programs. Typed Racket also doesn’t support Racket’s interfaces yet, another open research topic. What you have is as close as you can get to what you ask for. Alternatively experiment with structures (but you seem to come from the OO world, so the above is good) — Matthias -- You received this message because you are subscribed to the Google Groups "Racket Users" group. To unsubscribe from this group and stop receiving emails from it, send an email to racket-users+unsubscr...@googlegroups.com. For more options, visit https://groups.google.com/d/optout.
