It's a minor point, but as you know every string going into the sort starts with 'A', it is redundant, and much slower, to use substr within the sort block to exclude it.

return sort grep{ substr( $_ , 0, 1 ) eq 'A' } @array;
[download]

Will produce the same result more quickly.

You're probably right. I was going to point out the same thing, but then I noticed that the OP specifically did a numerical sort on the front-truncated string, which made me think that the original array might be something on the lines of:

qw ( A100 B57 X22 A9 A12 A3 C2 )

But who knows? ;-)

dave

Looking back at the OP, you're right, he did originally use <=>, but then changed to cmp in the later post to which I responded..

---

Examine what is said, not who speaks.

"Efficiency is intelligent laziness." -David Dunham
"Think for yourself!" - Abigail
Hooray!

my @array = &filter(qw(Art bob joe andy willy Andrew john Archie));
print "$_\n" for(@array);

sub filter(){
    return sort {substr($a,1) cmp substr($b,1)} grep {substr($_,0,1) e
+q 'A'} @_;
}
[download]