[m-rev.] Re: construct__get_functor bug

[Fergus Henderson]

[From mercury-developers]

On 19-Dec-2002, Peter Ross <pro at missioncriticalit.com> wrote:
> I would like comments [...]
...
> library/construct.m:
> 	Add a new version of get_functor which takes a term of the
> 	type we are trying to get the functor for.  This allows one to
> 	handle functors with existentially quantified arguments.

I don't really understand how this is supposed to work,
i.e. what the purpose of the term is.

What happens if you call get_functor with a functor number
that doesn't match the functor in the term?

Also, what happens if you call get_functor with a term
whose type doesn't match the type_desc passed in?
It is probably best to just use the type_info `TypeInfo_for_T',
which gets implicitly passed to the C code, rather than
passing a type_desc in.

> runtime/mercury_construct.c:
> 	Add missing MR_TYPECTOR_REP_UNIV cases to the switch.

I think it would be better to add cases for that which call
MR_fatal_error("MR_TYPECTOR_REP_UNIV unexpected").
These cases should not occur anymore, since univ is not
treated specially anymore.

> +MR_bool
> +MR_may_contain_existq_type_infos(const MR_Word data,
> +		const MR_TypeCtorInfo tci, MR_DuFunctorDesc **functor_desc,
> +		MR_Word **arg_vector)
> +{
> +    MR_DuTypeLayout du_type_layout;
> +    switch(MR_type_ctor_rep(tci)) {
> +	case MR_TYPECTOR_REP_DU:
> +	case MR_TYPECTOR_REP_DU_USEREQ:
> +	case MR_TYPECTOR_REP_UNIV:
> +	{
> +	    const MR_DuPtagLayout   *ptag_layout;
> +	    const MR_DuExistInfo    *exist_info;
> +	    int                     ptag;
> +	    MR_Word                 sectag;
> +
> +	    ptag = MR_tag(data);
> +	    du_type_layout = MR_type_ctor_layout(tci).MR_layout_du;
> +	    ptag_layout = &du_type_layout[ptag];
> +
> +	    switch (ptag_layout->MR_sectag_locn) {
> +		case MR_SECTAG_NONE:
> +		    *functor_desc = ptag_layout->MR_sectag_alternatives[0];
> +		    *arg_vector = (MR_Word *) MR_body(data, ptag);
> +		    break;
> +		case MR_SECTAG_LOCAL:
> +		    sectag = MR_unmkbody(data);
> +		    *functor_desc =
> +			ptag_layout->MR_sectag_alternatives[sectag];
> +		    *arg_vector = NULL;
> +		    break;
> +		case MR_SECTAG_REMOTE:
> +		    sectag = MR_field(ptag, data, 0);
> +		    *functor_desc =
> +			ptag_layout->MR_sectag_alternatives[sectag];
> +		    *arg_vector = (MR_Word *) MR_body(data, ptag) + 1;
> +		    break;
> +		case MR_SECTAG_VARIABLE:
> +		    /* XXX how do I handle this case */
> +		default:
> +		    MR_fatal_error("MR_may_contain_existq_type_infos"
> +			 ": invalid sectag_locn");
> +		}
> +
> +	    	return MR_TRUE;
> +	    }
> +	    default:
> +		return MR_FALSE;
> +    }

Misleading indentation: the "default:" and "return MR_FALSE" lines
should both be unintended four spaces.

-- 
Fergus Henderson <fjh at cs.mu.oz.au>  |  "I have always known that the pursuit
The University of Melbourne         |  of excellence is a lethal habit"
WWW: <http://www.cs.mu.oz.au/~fjh>  |     -- the last words of T. S. Garp.
--------------------------------------------------------------------------
mercury-reviews mailing list
post:  mercury-reviews at cs.mu.oz.au
administrative address: owner-mercury-reviews at cs.mu.oz.au
unsubscribe: Address: mercury-reviews-request at cs.mu.oz.au Message: unsubscribe
subscribe:   Address: mercury-reviews-request at cs.mu.oz.au Message: subscribe
--------------------------------------------------------------------------