[m-dev.] diff: type_name/1 et al

[Tyson Richard DOWD]

> On Thu, 24 Apr 1997, Fergus Henderson wrote:
> 
> > > Why shouldn't type_of(in, out) just be multi?  Why not allow it to
> > > generate the names of equivalent types?
> > 
> > Because that would be difficult to implement, inefficient, and not useful.
> 
> Ok, then why not just implement
> 
> 	:- mode type_of(in, in) is semidet?
> 
> The type_of(in, out) is cc_multi version seems like it's not going to do
> anything useful very often.  If you want to get "the" type name, you
> probably want the canonical one, rather than just picking one of the
> equivalent names at random.

Just because the declarative semantics are that one of the equivalent
names is chosen at random, doesn't mean that operationally, you won't
(usually) get a reasonable one. I'd like to see

	map lookup failed - map type was map(var, goal_info).

of course, I wouldn't be too unhappy if it said

	map lookup failed - map type was map(int, goal_info).

I think that one reason you want the *canonical* type_of is for
efficiency (so you don't have to collapse equivalences whenever you do
something with the type_info), and if you want the type_name, you either
	- want to be as high up the equivalence chain as you can,
	  to get the most information about the type.
	- want the canonical type, so you get the most basic information
	  about the type.

-- 
       Tyson Dowd           #
                            #             Sign on refrigerator:
     trd at cs.mu.oz.au        #           Refrigerate after opening. 
http://www.cs.mu.oz.au/~trd #                  - C. J. Owen.