[m-dev.] for review: add type_ctor_reps for user-defined equality

[Zoltan Somogyi]

Tyson wrote:
> +** Note: next time the binary compatibility version number is increased,
> +** we should remove the MR_get_new_type_ctor_rep() function in
> +** runtime/mercury_type_info.c.  

It would be a good idea to do this before the next release, in order to
remove the overhead of the calls to MR_get_new_type_ctor_rep.

> +		case MR_TYPECTOR_REP_ENUM_USEREQ:
>  		case MR_TYPECTOR_REP_DU:
> +		case MR_TYPECTOR_REP_DU_USEREQ:
>  		case MR_TYPECTOR_REP_ARRAY:
>  		case MR_TYPECTOR_REP_NOTAG:
> +		case MR_TYPECTOR_REP_NOTAG_USEREQ:
>  		case MR_TYPECTOR_REP_EQUIV:
>  		case MR_TYPECTOR_REP_EQUIV_VAR:

This should enable you to undo Fergus's action in disabling the use of this
code. You definitely want to check that this would work; you may want to wait
committing that until we can delete MR_get_new_type_ctor_rep, or redefine it
as an identity macro.

However, I think you should treat ENUM_USEREQ conservatively (i.e. call the
procedure in the type_ctor_info) not just for unifications but also for
index and comparison operations as well. For example, the auto-generated
comparison predicate for types with user-defined equality calls error;
it does not return.

I did not see any other problems in the diff, but then I did not look too hard.

Zoltan.
--------------------------------------------------------------------------
mercury-developers mailing list
Post messages to:       mercury-developers at cs.mu.oz.au
Administrative Queries: owner-mercury-developers at cs.mu.oz.au
Subscriptions:          mercury-developers-request at cs.mu.oz.au
--------------------------------------------------------------------------