In general, if I need to use C-style for loops to traverse arrays, I simply put the @array limit directly in the condition:

for( my $i = 0; $i < @array; ++i ) {
   ...;
}
[download]

I realise that you were probably thinking that as you have two nested loops running to the same limit, it might be more efficient to do the size 'calculation' before both and use a scalar; but the reality is that there is no calculation being done, the size of the array represented by @array is simply a direct reference to an internal integer, so the cost is actually less than referencing the integer value of a scalar variable.

The difference isn't big enough to worry about for most purposes:

cmpthese -1,{
    a=>q[
        my @a = 1 .. 1e3;
        for( my $i = 0; $i < @a; ++$i ){
            for( my$j = 0; $j < @a; ++$j ){ 
                1; 
            } 
        }
    ], 
    b=>q[
        my @a = 1 .. 1e3; 
        my $l = @a; 
        for( my $i = 0; $i < $l; ++$i ){
            for( my $j = 0; $j < $l; ++$j ){ 
                1; 
            } 
        }
    ] 
};;
    Rate    a    b
a 5.81/s   -- -41%
b 9.84/s  69%   --
[0] Perl>
[download]

But the point is that there is little purpose in assigning the size of an array to a scalar variable, just reference it directly when you need it.

Ditto for the length of a perl string.

That is actually much more efficient, thank you. I will use that from now on.