| 1 |
<?php |
| 2 |
/** |
| 3 |
* Create a web friendly URL slug from a string. |
| 4 |
* |
| 5 |
* Although supported, transliteration is discouraged because |
| 6 |
* 1) most web browsers support UTF-8 characters in URLs |
| 7 |
* 2) transliteration causes a loss of information |
| 8 |
* |
| 9 |
* @author Sean Murphy <sean@iamseanmurphy.com> |
| 10 |
* @copyright Copyright 2012 Sean Murphy. All rights reserved. |
| 11 |
* @license http://creativecommons.org/publicdomain/zero/1.0/ |
| 12 |
* |
| 13 |
* @param string $str |
| 14 |
* @param array $options |
| 15 |
* @return string |
| 16 |
*/ |
| 17 |
function mb_url_slug($str, $options = array()) { |
| 18 |
if (! function_exists('mb_convert_encoding')) |
| 19 |
return $str; |
| 20 |
|
| 21 |
// Make sure string is in UTF-8 and strip invalid UTF-8 characters |
| 22 |
$str = mb_convert_encoding((string)$str, 'UTF-8', mb_list_encodings()); |
| 23 |
|
| 24 |
$defaults = array( |
| 25 |
'delimiter' => '-', |
| 26 |
'limit' => null, |
| 27 |
'lowercase' => true, |
| 28 |
'replacements' => array(), |
| 29 |
'transliterate' => false, |
| 30 |
); |
| 31 |
|
| 32 |
// Merge options |
| 33 |
$options = array_merge($defaults, $options); |
| 34 |
|
| 35 |
$char_map = array( |
| 36 |
// Latin |
| 37 |
'À' => 'A', 'Á' => 'A', 'Â' => 'A', 'Ã' => 'A', 'Ä' => 'A', '� |
| 38 |
' => 'A', 'Æ' => 'AE', 'Ç' => 'C', |
| 39 |
'È' => 'E', 'É' => 'E', 'Ê' => 'E', 'Ë' => 'E', 'Ì' => 'I', 'Í' => 'I', 'Î' => 'I', 'Ï' => 'I', |
| 40 |
'Ð' => 'D', 'Ñ' => 'N', 'Ò' => 'O', 'Ó' => 'O', 'Ô' => 'O', 'Õ' => 'O', 'Ö' => 'O', 'Ő' => 'O', |
| 41 |
'Ø' => 'O', 'Ù' => 'U', 'Ú' => 'U', 'Û' => 'U', 'Ü' => 'U', 'Ű' => 'U', 'Ý' => 'Y', 'Þ' => 'TH', |
| 42 |
'ß' => 'ss', |
| 43 |
'à' => 'a', 'á' => 'a', 'â' => 'a', 'ã' => 'a', 'ä' => 'a', 'å' => 'a', 'æ' => 'ae', 'ç' => 'c', |
| 44 |
'è' => 'e', 'é' => 'e', 'ê' => 'e', 'ë' => 'e', 'ì' => 'i', 'í' => 'i', 'î' => 'i', 'ï' => 'i', |
| 45 |
'ð' => 'd', 'ñ' => 'n', 'ò' => 'o', 'ó' => 'o', 'ô' => 'o', 'õ' => 'o', 'ö' => 'o', 'ő' => 'o', |
| 46 |
'ø' => 'o', 'ù' => 'u', 'ú' => 'u', 'û' => 'u', 'ü' => 'u', 'ű' => 'u', 'ý' => 'y', 'þ' => 'th', |
| 47 |
'ÿ' => 'y', |
| 48 |
|
| 49 |
// Latin symbols |
| 50 |
'©' => '(c)', |
| 51 |
|
| 52 |
// Greek |
| 53 |
'Α' => 'A', 'Β' => 'B', 'Γ' => 'G', 'Δ' => 'D', 'Ε' => 'E', 'Ζ' => 'Z', 'Η' => 'H', 'Θ' => '8', |
| 54 |
'Ι' => 'I', 'Κ' => 'K', 'Λ' => 'L', 'Μ' => 'M', 'Ν' => 'N', 'Ξ' => '3', 'Ο' => 'O', 'Π' => 'P', |
| 55 |
'Ρ' => 'R', 'Σ' => 'S', 'Τ' => 'T', 'Υ' => 'Y', 'Φ' => 'F', 'Χ' => 'X', 'Ψ' => 'PS', 'Ω' => 'W', |
| 56 |
'Ά' => 'A', 'Έ' => 'E', 'Ί' => 'I', 'Ό' => 'O', 'Ύ' => 'Y', 'Ή' => 'H', 'Ώ' => 'W', 'Ϊ' => 'I', |
| 57 |
'Ϋ' => 'Y', |
| 58 |
'α' => 'a', 'β' => 'b', 'γ' => 'g', 'δ' => 'd', 'ε' => 'e', 'ζ' => 'z', 'η' => 'h', 'θ' => '8', |
| 59 |
'ι' => 'i', 'κ' => 'k', 'λ' => 'l', 'μ' => 'm', 'ν' => 'n', 'ξ' => '3', 'ο' => 'o', 'π' => 'p', |
| 60 |
'ρ' => 'r', 'σ' => 's', 'τ' => 't', '� |
| 61 |
' => 'y', 'φ' => 'f', 'χ' => 'x', 'ψ' => 'ps', 'ω' => 'w', |
| 62 |
'ά' => 'a', 'έ' => 'e', 'ί' => 'i', 'ό' => 'o', 'ύ' => 'y', 'ή' => 'h', 'ώ' => 'w', 'ς' => 's', |
| 63 |
'ϊ' => 'i', 'ΰ' => 'y', 'ϋ' => 'y', 'ΐ' => 'i', |
| 64 |
|
| 65 |
// Turkish |
| 66 |
'Ş' => 'S', 'İ' => 'I', 'Ç' => 'C', 'Ü' => 'U', 'Ö' => 'O', 'Ğ' => 'G', |
| 67 |
'ş' => 's', 'ı' => 'i', 'ç' => 'c', 'ü' => 'u', 'ö' => 'o', 'ğ' => 'g', |
| 68 |
|
| 69 |
// Russian |
| 70 |
'А' => 'A', 'Б' => 'B', 'В' => 'V', 'Г' => 'G', 'Д' => 'D', 'Е' => 'E', 'Ё' => 'Yo', 'Ж' => 'Zh', |
| 71 |
'З' => 'Z', 'И' => 'I', 'Й' => 'J', 'К' => 'K', 'Л' => 'L', 'М' => 'M', 'Н' => 'N', 'О' => 'O', |
| 72 |
'П' => 'P', 'Р' => 'R', 'С' => 'S', 'Т' => 'T', 'У' => 'U', 'Ф' => 'F', 'Х' => 'H', 'Ц' => 'C', |
| 73 |
'Ч' => 'Ch', 'Ш' => 'Sh', 'Щ' => 'Sh', 'Ъ' => '', 'Ы' => 'Y', 'Ь' => '', 'Э' => 'E', 'Ю' => 'Yu', |
| 74 |
'Я' => 'Ya', |
| 75 |
'а' => 'a', 'б' => 'b', 'в' => 'v', 'г' => 'g', 'д' => 'd', 'е' => 'e', 'ё' => 'yo', 'ж' => 'zh', |
| 76 |
'з' => 'z', 'и' => 'i', 'й' => 'j', 'к' => 'k', 'л' => 'l', 'м' => 'm', 'н' => 'n', 'о' => 'o', |
| 77 |
'п' => 'p', 'р' => 'r', 'с' => 's', 'т' => 't', 'у' => 'u', 'ф' => 'f', '� |
| 78 |
' => 'h', 'ц' => 'c', |
| 79 |
'ч' => 'ch', 'ш' => 'sh', 'щ' => 'sh', 'ъ' => '', 'ы' => 'y', 'ь' => '', 'э' => 'e', 'ю' => 'yu', |
| 80 |
'я' => 'ya', |
| 81 |
|
| 82 |
// Ukrainian |
| 83 |
'Є' => 'Ye', 'І' => 'I', 'Ї' => 'Yi', 'Ґ' => 'G', |
| 84 |
'є' => 'ye', 'і' => 'i', 'ї' => 'yi', 'ґ' => 'g', |
| 85 |
|
| 86 |
// Czech |
| 87 |
'Č' => 'C', 'Ď' => 'D', 'Ě' => 'E', 'Ň' => 'N', 'Ř' => 'R', 'Š' => 'S', 'Ť' => 'T', 'Ů' => 'U', |
| 88 |
'Ž' => 'Z', |
| 89 |
'č' => 'c', 'ď' => 'd', 'ě' => 'e', 'ň' => 'n', 'ř' => 'r', 'š' => 's', 'ť' => 't', 'ů' => 'u', |
| 90 |
'ž' => 'z', |
| 91 |
|
| 92 |
// Polish |
| 93 |
'Ą' => 'A', 'Ć' => 'C', 'Ę' => 'e', 'Ł' => 'L', 'Ń' => 'N', 'Ó' => 'o', 'Ś' => 'S', 'Ź' => 'Z', |
| 94 |
'Ż' => 'Z', |
| 95 |
'� |
| 96 |
' => 'a', 'ć' => 'c', 'ę' => 'e', 'ł' => 'l', 'ń' => 'n', 'ó' => 'o', 'ś' => 's', 'ź' => 'z', |
| 97 |
'ż' => 'z', |
| 98 |
|
| 99 |
// Latvian |
| 100 |
'Ā' => 'A', 'Č' => 'C', 'Ē' => 'E', 'Ģ' => 'G', 'Ī' => 'i', 'Ķ' => 'k', 'Ļ' => 'L', '� |
| 101 |
' => 'N', |
| 102 |
'Š' => 'S', 'Ū' => 'u', 'Ž' => 'Z', |
| 103 |
'ā' => 'a', 'č' => 'c', 'ē' => 'e', 'ģ' => 'g', 'ī' => 'i', 'ķ' => 'k', 'ļ' => 'l', 'ņ' => 'n', |
| 104 |
'š' => 's', 'ū' => 'u', 'ž' => 'z' |
| 105 |
); |
| 106 |
|
| 107 |
// Make custom replacements |
| 108 |
$str = preg_replace(array_keys($options['replacements']), $options['replacements'], $str); |
| 109 |
|
| 110 |
// Transliterate characters to ASCII |
| 111 |
if ($options['transliterate']) { |
| 112 |
$str = str_replace(array_keys($char_map), $char_map, $str); |
| 113 |
} |
| 114 |
|
| 115 |
// Replace non-alphanumeric characters with our delimiter |
| 116 |
$str = preg_replace('/[^\p{L}\p{Nd}]+/u', $options['delimiter'], $str); |
| 117 |
|
| 118 |
// Remove duplicate delimiters |
| 119 |
$str = preg_replace('/(' . preg_quote($options['delimiter'], '/') . '){2,}/', '$1', $str); |
| 120 |
|
| 121 |
// Truncate slug to max. characters |
| 122 |
$str = mb_substr($str, 0, ($options['limit'] ? $options['limit'] : mb_strlen($str, 'UTF-8')), 'UTF-8'); |
| 123 |
|
| 124 |
// Remove delimiter from ends |
| 125 |
$str = trim($str, $options['delimiter']); |
| 126 |
|
| 127 |
return $options['lowercase'] ? mb_strtolower($str, 'UTF-8') : $str; |
| 128 |
} |
| 129 |
?> |
| 130 |
|
