[m-dev.] for review: type specialisation

[Fergus Henderson]

On 30-Aug-1998, Simon Taylor <stayl at cs.mu.OZ.AU> wrote:
> Rework the handling of types in higher_order.m.
> - Fix bugs in higher_order.m that stopped it working with --typeinfo-liveness.
> - Perform type and typeclass specialisation.
...
> @@ -836,13 +1224,12 @@
>          NextHOid is NextHOid0 + 1,
>          string__append_list([Name0, "__ho", IdStr], PredName),
>          pred_info_typevarset(PredInfo0, TypeVars),
> -        remove_listof_higher_order_args(Types0, 1, HOArgs, Types),
>          pred_info_context(PredInfo0, Context),
>          pred_info_get_markers(PredInfo0, MarkerList),
>          pred_info_get_goal_type(PredInfo0, GoalType),
>                  % When we start specialising class method calls, this
> -                % context will need to be updated. 
> -                % cf.  remove_listof_higher_order_args.
> +                % context will need to be updated -dgj?
> +                % XXX Is this needed by anything after here? -stayl
>          pred_info_get_class_context(PredInfo0, ClassContext),

There might well be something that currently depends on it --
I don't know off-hand.

I think the class context here should be updated,
even if there is currently nothing that depends on it.
It will just make maintenance easier if higher_order.m preserves
type-correctness.

> +        % Substitute the types in a higher_order_arg.
> +:- pred substitute_higher_order_arg(substitution::in, higher_order_arg::in, 
> +                higher_order_arg::out) is det.

If the argument there is supposed to be a type substitution,
please use the type `tsubst' instead of `substitution'.

Otherwise, that looks OK.

-- 
Fergus Henderson <fjh at cs.mu.oz.au>  |  "I have always known that the pursuit
WWW: <http://www.cs.mu.oz.au/~fjh>  |  of excellence is a lethal habit"
PGP: finger fjh at 128.250.37.3        |     -- the last words of T. S. Garp.