On 9/10/07, Jeremy Kister <[EMAIL PROTECTED]> wrote:
> I am trying to optimize some sorting code I have. The data structure is
> as follows:
>
> my %hash = (x => [ 'a','b','c' ],
> y => [ 'd','e' ],
> z => [ 'f' ],
> );
>
>
> The result I expect is simply the highest number of elements. In this
> case, the result should be "3" because x has three elements, which is
> more than y and z have.
>
>
> I whipped up two versions of the sort code:
>
> foreach my $key (keys %hash){
> if(@{$hash{$key}} > $highest){
> $most = @{$hash{$key}};
> }
> }
>
>
> #and
>
> my $most = @{$hash{(sort { @{$hash{$b}} <=> @{$hash{$a}} } keys %hash)[0]}};
>
>
>
> I ran each 500,000 times. I expected the second version finish sooner,
> but it is noticeably slower than the first.
>
> Can someone show me a better way to get this done ?
snip
The problem is that you aren't sorting an array or a list, you are
sorting a transform ed list. The Perl loop is faster because it
performs the transform only once for each element. If fact, the
transform costs significantly more than the comparison.
--
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
http://learn.perl.org/
