--- Jeff Pinyan <[EMAIL PROTECTED]> wrote:
> On May 22, Paul said:
>
> >I know that in many C compilers, the a ? b : c construct with the
> >ternary ?: operator si not stable after the second or third nesting,
> >but I've never seen that sort of problem in tests I've run in Perl.
>
> The only to watch out for is precendence:
>
> $a ? $b = $c : $b = $d;
>
> is like
>
> ($a ? $b = $c : $b) = $d;
>
> which always sets $b to $d.
Yup, tho in that case I'd usually say
$b = $a ? $c : $d;
>
> > sub rate ($) {
> > $_[0] eq 'A' ? .03 :
> > $_[0] eq 'B' ? .05 :
> > $_[0] eq 'C' ? .06 :
> > .08; # the default
> > }
>
> That's fine, but I prefer to use hashes for that sort of thing.
>
> {
> my %rates = qw(
> A .03
> B .05
> C .06
> );
> sub rate {
> exists $rates{$_[0]} ? $rates{$_[0]} : .08
> }
> }
Yep! But the actual function has a little more processing:
sub charge ($) { # receives a boolean; rerate or not
my $band = @_ and $routes{$mf03->route} ? 0 : $mf03->band;
my $mou = $mf03->bmou*($peak{$mf03->tod}||.50) #50% off-peak
discount
if $band; # wasted effort on 'X', but that's rare
return int(100 * sprintf "%.2f",
$band eq '0' ? .12 :
$band eq 'X' ? .12 :
$band eq 'A' ? .04 + $mou * .03 : # setup is .04 per call
$band eq 'B' ? .04 + $mou * .05 :
$band eq 'C' ? .04 + $mou * .06 :
.04 + $mou * .08);
}
Still tossing this salad,
but I think I've got all my veggies in it now. =o)
__________________________________________________
Do You Yahoo!?
Yahoo! Auctions - buy the things you want at great prices
http://auctions.yahoo.com/
