[m-rev.] for review: deconstruct by functor number

[Ralph Becket]

Simon Taylor, Tuesday,  2 January 2007:
> 
> > Why wouldn't num_functors just return 0 for such types?  I don't
> > understand this behaviour.
> 
> The intention is that if num_functors succeeds, other functions which
> only work for discriminated union types can be used.  There are very
> few situations in which you want to use the result of num_functors
> without testing that it is greater than zero.  I'd also argue that
> it doesn't make sense to talk about the number of functors of type
> `float', it just doesn't have functors at all.
> 
> Also, it would be better to leave the interface alone rather than
> changing the return value because that change is more likely to
> silently break code rather than causing an error.

Taking this line, the documentation for num_functors should include this
justification.  Julien and I spent about ten minutes trying to work this
out from the source code, but couldn't find an explanation.
--------------------------------------------------------------------------
mercury-reviews mailing list
Post messages to:       mercury-reviews at csse.unimelb.edu.au
Administrative Queries: owner-mercury-reviews at csse.unimelb.edu.au
Subscriptions:          mercury-reviews-request at csse.unimelb.edu.au
--------------------------------------------------------------------------