[racket] inherit, inherit/super, and inherit/inner in class syntax

From: Asumu Takikawa (asumu at ccs.neu.edu)
Date: Fri Jun 14 18:13:20 EDT 2013

On 2013-06-14 14:39:57 -0600, Christopher wrote:
> So I have been trying to learn the ins and outs of Racket's class
> system.  I've a little puzzled when it comes to the "inherit"-forms.
> I've poured over the Racket Reference and fiddled with some
> experimental classes, but I'm still not clear.
> My questions specifically are, What is the difference between inherit
> and inherit/super, and how does inherit/inner work?

Here's an example that hopefully shows the difference:

  #lang racket

  (define point%
    (class object%
      (inspect #f) ; to make example object easier to understand
      (init-field [x 0] [y 0])
      (define/public (move-x dx)
        (new this% [x (+ x dx)] [y y]))))

  (define fast-point%
    (class point%
      (inherit/super move-x)
      ;; or you can inherit
      ;(inherit move-x)
      (define/public (move-fast dx)
        ;; only with inherit/super or override
        (super move-x (* dx 10))
        ;; with inherit, inherit/super, or override
        ;(move-x (* dx 10))

  (send (new fast-point% [x 0] [y 2]) move-fast 3)

Notice that with `inherit/super`, you can use `super` on the method name
that you inherit from the superclass. Normally, you can only call
`super` on a method name that you are overriding.

On the other hand, with either `inherit` or `inherit/super`, you can
call the superclass method by just using the name.

I have never found the need to use `inherit/super` or `inherit/inner` in
my programs though. I always use `inherit`.

(If you wanted to know the rationale of why `inherit/super` exists, I'm
 not sure. The commit log says it was added to replace `rename-super`


Posted on the users mailing list.