On Sat, Jun 02, 2001 at 06:13:04PM +1000, iain truskett wrote:
> Rob talked about
>
> foreach $value(sort {$a <=> $b} (values(%hash)) {
> print $value;
> }
>
> vs.
>
> foreach $value(sort byNum (values(%hash)) {
> print $value;
> }
>
> sub byNum {
> $a <=> $b;
> }
>
>
> Am I correct in thinking that the anonymous subroutine is vaguely more
> efficient?
>
I'm really not sure. It probably all depends. I'm not guru enough to
know all the ins and outs of efficiency. I would guess, though, that if
you have a fairly complex sort routine that you want to use a number of
times, it might be more efficient to compile it just once.
Mainly, I think it's a readability thing. Using {$a <=> $b} inline is
easier for maintenance than naming a routine, then having to go find it
and see what it does. On the other hand, if you had a sort routine
something like the one below, that would be a bit of a pain to code
inline as an anonymous sub. And you could debug it as a named sub at
just the one place in the code.
sub byIP
{
@a = split(/\./, $a);
@b = split(/\./, $b);
($a[0] <=> $b[0]) ||
($a[1] <=> $b[1]) ||
($a[2] <=> $b[2]) ||
($a[3] <=> $b[3]);
}
cheers
rob c
