Re: tri-condition

[askok@...]

There exists some benefits for a 'native' core level 
implementation of ? : but at least for my needs, this 
should do.

Having the "tf(cond) { a, b }" syntax is to my liking, 
because it clearly separates the values from the 
condition. Performance is here not the main goal; 
readability and "sugarness" is. :)

With the token filter, syntax becomes "? cond { a, b }" 
which I would regard rather nice; field experiments are 
starting, that is, whether I'd actually start using that 
ideom in my own code.

-asko


On Wed, 25 Oct 2006 14:42:24 -0700
 "Taj Khattra" <taj.khattra@gmail.com> wrote:
> > See Exercise 1.4 in SICP.
>
> i think you meant 1.5.
>
> anyways, i wasn't asking why it evaluates both, but 
> pointing out that
> it does evaluate both (which is not how one might expect 
> it to work
> cf. C and Perl's ternary operator).