[mercury-users] Re: can one_solution/2 and unsorted_solutions/2 be det?

[Fergus Henderson]

Lee Naish wrote:
> 
> Fergus wrote:
> 
> >In fact Mercury does allow the programmer to fix both the search rule
> >and the computation rule.  The Mercury language reference manual documents
> >that implementations must support a strict sequential (left-to-right)
> >operational semantics.  The current Mercury implementation supports
> >this with the `--strict-sequential' option.  You can also get more
> >fine-grained control with the `--no-reorder-conj', `--no-reorder-disj',
> >and `--fully-strict' options.
> 
> Last time I was involved in discussion of these, `--no-reorder-conj' did
> not guarantee no reordering (it just prevented unnecessary reordering).
> I suggested the meaning should be changed.  Was this done?

No.

I agree that this would be a good idea, it is just not a high priority
at the moment.  The existing `--no-reorder-conj' option is still useful, even
if the name is perhaps a bit misleading; it gives a fixed computation rule.

> Does `--strict-sequential' really guarantee no reordering (ie, will the
> compiler produce an error message if left to right ordering is not "well
> moded")?

No.

-- 
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.