Artikel im Internet unter http://www.hidemail.de/blog/how-can-i-extract-just-the-unique-elements-of-an-array-.shtml.
Sonntag, 11.11.2007, 18:30:47 Uhr

How can I extract just the unique elements of an array?


Years ago i wrote a subroutine to find the unique elements of an Array.

The code was the following

sub del_double{
my %all;
grep {$all{$_}=0} @_;
return (keys %all);
}



How does it work?
Well, first the hash %all is defined.
With the grep-function I set the keys of %all with all of the elements of my array. Double keys are not possible, so i have the unique elements in (keys %all).

This funtion ist very fast, and it found it's place in my private perl-library.


A faster method is the following:

sub del_double
{
my %all;
$all{$_}=0 for @_;
return (keys %all);
}



It works with a for-loop, not with grep. And this is the reason, why it is faster than the first method. The mighty grep-function is very powerful with arrays... but slower than a for or a foreach.


Since I began this blog, I've been looking for the very fastest subroutine to solve this problem.

Now, after searching the internet, I found the (I think) fastes way to find the unique elements of an array.

The code is

sub del_double{
my %all=();
@all{@_}=1;
return (keys %all);
}


Why is this so fast?
Differnt to my method, this method gives the array DIRECTLY to the hash %all. It needs no grep, for or foreach, and I think, this is the reason, why it is so fast.

But be careful: The chronological order of the returned array is not the same as the given array.


Artikel im Internet unter http://www.hidemail.de/blog/how-can-i-extract-just-the-unique-elements-of-an-array-.shtml.