How can I find out if two strings are mostly equal (in perl)?

Question

I have a string that I want to compare against an array of strings, and return the array element that most closely matches.

I can write a sliding correlator that counts the number of matching characters at each step and returns the max correlation. But is there a better way?

For example:
control_string = drv_probability_1_max

List:
burst_period_min/max
duty_cycle_min/max
probablility_0_min/max
probablility_1_min/max

Where ideally it returns "probablility_1_min/max"

Answer 1

use String::Approx;

Answer 2

You might be looking for the String::Similarity module.

Answer 3

Take a look at Text::Levenshtein or List::Compare (String::Approx is the same method, but a more complete package)

use strict; use warnings;

use Text::Levenshtein qw(distance);

my $ctl = "drv_probability_1_max";

my @list=qw|
burst_period_min/max
duty_cycle_min/max
probablility_0_min/max
probablility_1_min/max
|;

my @dist=distance($ctl,@list);

print "Levenshtein distances: @dist\n";

my $idmin=0;
$dist[$idmin] < $dist[$_] or $idmin = $_ for 1..$#dist;

print "\"$list[$idmin]\" seems the closest...\n\n\n";

Output:

Levenshtein distances: 16 16 10 9
"probablility_1_min/max" seems the closest...

Read more about Levenshtein Distance or implement a Perl algorithm directly..