[Caml-list] Comparison of functional values

[Caml-list] Comparison of functional values

[Olivier Andrieu]

Hello list, 

I wanted to build a Set of functional value. But generic equality
can't be used for functional values (raises an exception ...). It is
therefore difficult to build a Set of such values because the Set.Make
functor requires a comparison function and Pervasives.compare relies
on (=).

Is there a way around this ? I thought I could use physical equality
(wich is OK wrt functional values) but there's no such thing as
`physical inequality'.

-- 
Olivier Andrieu
-------------------
To unsubscribe, mail caml-list-request@inria.fr.  Archives: http://caml.inria.fr

(no subject)

[Juergen Pfitzenmaier]

Olivier Andrieu wanted to build a Set of functional values but had problems
with the comparison of these values.

One solution would be to have a Set of pairs. The first element would be a
functional value and the second something of a simple type (like integer).
Then you define your comparison function only to use this second element.
The only inconvenience is that you have always to select explicitly the first
element of such pairs to get the functional value; but you can hide this
inside a module build around Set.
hope this is of some help

pfitzen

-------------------
To unsubscribe, mail caml-list-request@inria.fr.  Archives: http://caml.inria.fr

Re: [Caml-list] Comparison of functional values

[Alan Schmitt]

>Is there a way around this ? I thought I could use physical equality
>(wich is OK wrt functional values) but there's no such thing as
>`physical inequality'.
>

As stated in http://caml.inria.fr/ocaml/htmlman/manual031.html, != is
the negation of ==.

Alan Schmitt

--
The hacker: someone who figured things out and made something cool happen.
-------------------
To unsubscribe, mail caml-list-request@inria.fr.  Archives: http://caml.inria.fr

Re: [Caml-list] Comparison of functional values

[Olivier Andrieu]

 Alan Schmitt [Wednesday 7 March 2001] :
 > As stated in http://caml.inria.fr/ocaml/htmlman/manual031.html, != is
 > the negation of ==.

Salut Alan !

I known. But to use the Set functor, you need a `compare' function of
type ('a -> 'a -> int) whereas != and == have type ('a -> bool). 
So != does not seem very helpful here.

What I meant actually, was a «physical compare». I found out that 
this (crude) C function works :

value phys_comp(value a, value b)
{
 if(a<b)
   return Val_int(-1) ;
 else if(a>b)
   return Val_int(+1) ;
 else 
   return Val_int(0);
}

Olivier
-------------------
To unsubscribe, mail caml-list-request@inria.fr.  Archives: http://caml.inria.fr

Re: [Caml-list] Comparison of functional values

[Xavier Leroy]

> What I meant actually, was a «physical compare». I found out that 
> this (crude) C function works :
> 
> value phys_comp(value a, value b)
> {
>  if(a<b)
>    return Val_int(-1) ;
>  else if(a>b)
>    return Val_int(+1) ;
>  else 
>    return Val_int(0);
> }

No, it does not:  the GC can copy heap blocks and change the relative
ordering of their memory addresses.  You really need to pair your
functions with unique identifying integers.

- Xavier Leroy
-------------------
To unsubscribe, mail caml-list-request@inria.fr.  Archives: http://caml.inria.fr

Re: [Caml-list] Comparison of functional values

[Arturo Borquez]

On Tue, 06 March 2001, Olivier Andrieu wrote:

> 
> 
> Hello list, 
> 
> I wanted to build a Set of functional value. But generic equality
> can't be used for functional values (raises an exception ...). It is
> therefore difficult to build a Set of such values because the Set.Make
> functor requires a comparison function and Pervasives.compare relies
> on (=).
> 
> Is there a way around this ? I thought I could use physical equality
> (wich is OK wrt functional values) but there's no such thing as
> `physical inequality'.
> 
> -- 
> Olivier Andrieu
> -------------------
> To unsubscribe, mail caml-list-request@inria.fr.  Archives: http://caml.inria.fr

Hi Olivier:
Perhaps this help

let a = fun f x -> x = x;;
let b = fun f x -> x = x;;
(a == b) = true;;
-: bool = false
not (a == b) = true;;
-: bool = true

Regards
Arturo Borquez


Find the best deals on the web at AltaVista Shopping!
http://www.shopping.altavista.com
-------------------
To unsubscribe, mail caml-list-request@inria.fr.  Archives: http://caml.inria.fr