in reply to Re^2: How can I improve the efficiency of this very intensive code?
in thread How can I improve the efficiency of this very intensive code?

Update: I just read frodo72's post, which makes the point better than this code.

In that case, if you are just going to use the first one, wouldn't finding the first one be faster than sort?

sub IsStrongMatch
{
    # Return true if id2 is only top ranked match for id1
    my $id1 = shift;
    my $id2 = shift;
    my $rC  = shift;

    for my $i1 ( keys %{$rC} )
    {
        next if $i1 == $id1;
        my $i2;
        for ( keys %{ $rC->{ $i1 } } ) {
            $i2 = $_ if not defined( $i2 );
              # if there's a good way to find a starting $i2
              # before beginning, then take this out, and
              # don't check every time

            $i2 = $_ if
                 $rC->{ $i1 }->{ $_ }
                   <
                 $rC->{ $i1 }->{ $i2 };
        }

        if ( $id2 == $i2 )
        {
            return 0;
        }
    }
    return 1;
}

[download]

    -Bryan

Replies are listed 'Best First'.
Re^4: How can I improve the efficiency of this very intensive code?
by polettix (Vicar) on Aug 07, 2005 at 00:07 UTC
    This is exactly what I was going to suggest, apart from a minor optimisation regarding the initialisation of the first value. The I recalled of the existence of reduce in List::Util, which does the job nicely. But maybe I should have been less cryptic and explain what the solution does exactly - thanks for the example, I'll update my post linking to it :)

    I know it's a corner case that should not happen, but I'd check for $i2 to be defined before comparing it against $id2.

    Flavio
    perl -ple'$_=reverse' <<<ti.xittelop@oivalf

    Don't fool yourself.
[reply]
[d/l]
[select]

In Section Seekers of Perl Wisdom