Maude-help mailing list

[Steven Eker <eker AT csl.sri.com>]

That is correct. I would say the predefined SET/LIST etc modules are most
useful if you need the algorithms - since they are carefully written to make
best use of Maude's fancy AC(U)/A(U) term representations and 
matching/replacement
algorithms for low complexity.

If you just need the signature, it's usually easier just to declare the 
sorts and
operators yourself. You'll notice this approach in the META-LEVEL, which 
contains
numerous ad hoc set and list sorts.

Steven

On 12/5/14, 12:50 AM, Konrad Hinsen wrote:
> Hi Steven,
>
>   >  Core Maude doesn't support parameterized views, though in principle it's
>   >  not a difficult feature to add. I believe Full Maude does, so there is a
>   >  worked out design. On-the-fly views would certainly be doable but there
>   >  are no plans to add them.
>
>   >  Now I think of it, Maude's immediate predecessor, OBJ3 had the notional
>   >  of principal sorts, and the ability to guess views. I imagine something
>   >  similar could be done in Maude by taking the largest user sort in a
>   >  target module as principal.
>
> Thanks for your comments. I had a quick look at OBJ3 which indeed has
> a mechanism for automatically generating simple views, in particular
> from TRIV. That covers 95% of the views I have used so far, so from
> my perspective such a mechanism is very useful.
>
> Back to my immediate problem: Do I understand correctly that for
> using lists whose elements are sorts from a parametrized module,
> I need to use Full Maude? In that case, I think it's simpler to
> re-implement a specialized version of LIST for this case.
>
> Konrad.
> _______________________________________________
> Maude-help mailing list
> Maude-help AT cs.uiuc.edu
> http://lists.cs.uiuc.edu/mailman/listinfo/maude-help