Re: [q-lang-users] "listof" surprises

[Albert G. <Dr....@t-...>]

Hi Marco,

again, it's a feature, not a bug. Please refer to Section 9.2 [Special
Constructors] of the manual for an explanation:
http://q-lang.sourceforge.net/qdoc/qdoc_9.html#SEC44

Quote: "Pattern matching on non-special arguments of a function takes
into account embedded special data constructors and will automagically
evaluate deferred subterms as needed."

This "call-by-need" pattern matching works pretty much like in Haskell.
It was introduced in Q 7.1. You can find a discussion of this feature in
the mailing list archive:

http://sourceforge.net/mailarchive/forum.php?thread_name=4463BF32.2020900%40t-online.de&forum_name=q-lang-users

It makes it much easier to define non-special functions by pattern
matching on lazy data structures. Special functions work as before (they
will not do any automatic evaluations when matching special subterms),
so declaring g to be a special form as well will give you the behaviour
that you want.

Cheers,
Albert

Marco Maggesi wrote:
> Unfortunately, I found another obstacle (bug?).
> I really do not understand this: given the program
> 
>    special expr E;
>    f (expr X) = expr X;
>    g (expr 0) = expr 0;
>    g (expr X) = expr X;
> 
> I get
> 
>    ==> f (expr random)
>    expr random
> 
>    ==> g (expr random)
>    expr 2635609360

-- 
Dr. Albert Gr"af
Dept. of Music-Informatics, University of Mainz, Germany
Email:  Dr....@t-..., ag...@mu...
WWW:    http://www.musikinformatik.uni-mainz.de/ag