Re: [Caml-list] First order compile time functorial polymorphism in Ocaml

[Jun.Furuse@inria.fr]

Hello,

At Mon, 23 Jun 2003 04:25:18 +1000,
John Skaller wrote:
> 
> In ML style functional programming languages like Ocaml,
> we have what is termed data polymorphism. This provides
> a kind of code reuse we're all familiar with.
> 
> However, there is another kind of polymophism
> which Ocaml does not provide. Two things to consider here:
> 
> 1. Every data structure has a map function.
> 2. User defined algebraic type require a hand written map function

Yes, I agree that writing map or fold function over again and again is 
trivial and boring.

Your approach reminds me polytypic programming or so-called 
"generic programming"[1] in Haskell. I have also considered a bit about 
the possibility of this "generic programming" in O'Caml. Actually, 
I think our "generics" (= G'Caml) has already had allmost of all 
the internal functionalities for so-called "generics" in Haskell
community. Only one possible problem of "generics" on Caml is that
type constructors of Caml is not so "mathematically pure" as Haskell.
I still do not have clear idea how to declare generic case for n-ary
tuples, objects and labeled function types of Caml...

Anyway, "generics" are too general solution for your purpose.
This is not something I can prepare for you in this week. :-)

[1]: Generic Haskell project
	http://www.cs.uu.nl/research/projects/generic-haskell/

--
Jun

-------------------
To unsubscribe, mail caml-list-request@inria.fr Archives: http://caml.inria.fr
Bug reports: http://caml.inria.fr/bin/caml-bugs FAQ: http://caml.inria.fr/FAQ/
Beginner's list: http://groups.yahoo.com/group/ocaml_beginners