README

mb_levenshtein PHP function

Levenshtein PHP function with UTF-8 support. This function finds similarity distance between two strings.

• Source Code @ GitHub
• Composer available @ Packagist

Functions

• Returns in Levenshtein Distance. (The smaller, the closer)

  mb_levenshtein ( string $str1 , string $str2 ) : int

  mb_levenshtein ( string $str1 , string $str2 , int $cost_ins , int $cost_rep , int $cost_del ) : int

• Returns in Levenshtein Ratio between 0 to 1. (The bigger, the closer)

  mb_levenshtein_ratio ( string $str1 , string $str2 ) : float

  mb_levenshtein_ratio ( string $str1 , string $str2 , int $cost_ins , int $cost_rep , int $cost_del ) : float

Usage

<?php include_once('./mb_levenshtein.php'); $query = 'cafe'; $comps = [ 'coffee', 'café', 'tea', 'sake', ]; echo "Query word: ${query}" . PHP_EOL; foreach ($comps as $comp) { $sim = mb_levenshtein($query, $comp); echo " ${comp}: ${sim}" . PHP_EOL; }

Results:

$ # The smaller, the closer $ php ./sample.php Query word: cafe  coffee: 3  café: 1  tea: 4  sake: 2

Composer

• To use released version:

  composer require keinos/mb_levenshtein

• To use latest version:

  composer require keinos/mb_levenshtein:dev-master

<?php require_once('vendor/autoload.php'); $query = 'cafe'; $comps = [ 'coffee', 'café', 'tea', 'sake', ]; echo "Query word: ${query}" . PHP_EOL; foreach ($comps as $comp) { $sim = mb_levenshtein($query, $comp); echo " ${comp}: ${sim}" . PHP_EOL; }

Results:

$ ls sample.php $ composer require keinos/mb_levenshtein Using version ^1.0 for keinos/mb_levenshtein ./composer.json has been created Loading composer repositories with package information Updating dependencies (including require-dev) Package operations: 1 install, 0 updates, 0 removals  - Installing keinos/mb_levenshtein (1.0.0): Downloading (100%) Writing lock file Generating autoload files $ # The smaller, the closer $ php ./sample.php Query word: cafe  coffee: 3  café: 1  tea: 4  sake: 2 $ ls composer.json composer.lock sample.php vendor $ cat composer.json {  "require": {  "keinos/mb_levenshtein": "^1.0"  } }