[m-dev.] Re: declarative debugger tests failing

[Fergus Henderson]

On 16-Jan-2001, Mark Anthony BROWN <dougl at cs.mu.OZ.AU> wrote:
> This is just to let you know what I'm up to.
> 
> I'm still looking into the test cases that are failing.  The problem seems
> to come from some calls to __Compare__ for type list(maybe(univ)), where
> both univs contain values of type list(int).  These calls occur when the
> oracle does a map search for previous answers.  Wrong comparison_results
> are returned, leading to map search failures.
> 
> Here's some of the output from mdb:
> 
>     3463:   1473 21 EXIT __Compare__ for list:list/1-0 (det) list.m:27 (declarative_execution.m:106)
> mdb> p *
>        HeadVar__1               '<'
>        HeadVar__2               .(yes(univ_cons(.(4, .(5, [])))), .(yes(univ_cons/1), ./2))
>        HeadVar__3               .(yes(univ_cons(.(1, .(2, ./2)))), .(yes(univ_cons/1), ./2))
> 
> In this case the result should be '>', since 4 > 1.  In some other cases
> equal terms give a result of '<'.
> 
> This problem only seems to happen in the front end of the debugger -- 
> I haven't yet been able to reproduce it in standalone tests.
> 
> I suspect that Tom's recent change to univs either introduced a bug or
> exposed an already existing bug in the declarative debugger.  I'm looking
> into that now.

I think the former is more likely, since one of the other test
cases, which was calling io__write on an array, was failing due
to det_type_to_univ failing, even though both of the types
involved were `array(int)'.  Comparisons of univs seems to be
broken.

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