[m-rev.] for review: arrays and the debugger

[Zoltan Somogyi]

On 19-Jun-2001, Fergus Henderson <fjh at cs.mu.OZ.AU> wrote:
> > Are you arguing against the addition of a limited_deconstruct predicate to
> > std_util or not?
> > If yes, what is your argument?
> 
> I've explained them at length already, but to summarize, I don't
> think the efficiency benefits are worth complicating the interface and
> implementation by adding this procedure.

But you have admitted that the efficiency difference is not a few percent,
but enough to affect feasibility. The interface complication is non-existent.
Users of those predicates *must* read warnings in their documentation about
the circumstances in which using deconstruct by itself is not a good idea;
the only question is whether they are given a standard alternative
(limited_deconstruct) or whether they are told to make their own. The
complication of the implementation is minimal. It can be as little as a
three-line Mercury predicate that performs an arity test before calling
deconstruct, and it must exist *somewhere*; better once in the library
than many times in user programs.

In other words, I don't find your arguments convincing.

I will wait two more days for a review of the details of the diff, from
anyone. If I don't get one, I will commit it.

Zoltan.
--------------------------------------------------------------------------
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
--------------------------------------------------------------------------