Last active
August 29, 2015 14:09
-
-
Save korobochkin/5db48e4777a85823e08e to your computer and use it in GitHub Desktop.
WordPress additional filters for sanitize bad filenames with Cyrillic symbols, iOS smiles and other bad symbols.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| <?php | |
| function selena_network_sanitize_file_name ($name, $original_name) { | |
| $chars_table = array ( | |
| // Cyrillic alphabet | |
| 'А' => 'a', 'Б' => 'b', 'В' => 'v', 'Г' => 'g', 'Д' => 'd', | |
| 'а' => 'a', 'б' => 'b', 'в' => 'v', 'г' => 'g', 'д' => 'd', | |
| 'Е' => 'e', 'Ж' => 'zh', 'З' => 'z', 'И' => 'i', 'Й' => 'j', | |
| 'е' => 'e', 'ж' => 'zh', 'з' => 'z', 'и' => 'i', 'й' => 'j', | |
| 'К' => 'k', 'Л' => 'l', 'М' => 'm', 'Н' => 'n', 'О' => 'o', | |
| 'к' => 'k', 'л' => 'l', 'м' => 'm', 'н' => 'n', 'о' => 'o', | |
| 'П' => 'p', 'Р' => 'r', 'С' => 's', 'Т' => 't', 'У' => 'u', | |
| 'п' => 'p', 'р' => 'r', 'с' => 's', 'т' => 't', 'у' => 'u', | |
| 'Ф' => 'f', 'Х' => 'h', 'Ц' => 'c', 'Ч' => 'ch', 'Ш' => 'sh', | |
| 'ф' => 'f', 'х' => 'h', 'ц' => 'c', 'ч' => 'ch', 'ш' => 'sh', | |
| 'Щ' => 'shch', 'Ь' => '', 'Ю' => 'ju', 'Я' => 'ja', | |
| 'щ' => 'shch', 'ь' => '', 'ю' => 'ju', 'я' => 'ja', | |
| // Ukrainian | |
| 'Ґ' => 'g', 'Є' => 'ye', 'І' => 'i', 'Ї' => 'yi', | |
| 'ґ' => 'g', 'є' => 'ye', 'і' => 'i', 'ї' => 'yi', | |
| // Russian | |
| 'Ё' => 'yo', 'Ы' => 'y', 'Ъ' => '', 'Э' => 'e', | |
| 'ё' => 'yo', 'ы' => 'y', 'ъ' => '', 'э' => 'e', | |
| // Belorussian | |
| 'Ў' => 'u', | |
| 'ў' => 'u', | |
| // German | |
| 'Ä' => 'ae', 'Ö' => 'oe', 'Ü' => 'ue', 'ß' => 'ss', | |
| 'ä' => 'ae', 'ö' => 'oe', 'ü' => 'ue', | |
| // Polish | |
| 'Ą' => 'a', 'Ć' => 'c', 'Ę' => 'e', 'Ł' => 'l', 'Ń' => 'n', | |
| 'ą' => 'a', 'ć' => 'c', 'ę' => 'e', 'ł' => 'l', 'ń' => 'n', | |
| 'Ó' => 'o', 'Ś' => 's', 'Ź' => 'z', 'Ż' => 'z', | |
| 'ó' => 'o', 'ś' => 's', 'ź' => 'z', 'ż' => 'z', | |
| // Hungarian | |
| 'Ő' => 'o', 'Ű' => 'u', | |
| 'ő' => 'o', 'ű' => 'u', | |
| // Czech | |
| 'Ě' => 'e', 'Š' => 's', 'Č' => 'c', 'Ř' => 'r', 'Ž' => 'z', | |
| 'ě' => 'e', 'š' => 's', 'č' => 'c', 'ř' => 'r', 'ž' => 'z', | |
| 'Ý' => 'y', 'Á' => 'a', 'É' => 'e', 'Ď' => 'd', 'Ť' => 't', | |
| 'ý' => 'y', 'á' => 'a', 'é' => 'e', 'ď' => 'd', 'ť' => 't', | |
| 'Ň' => 'n', 'Ú' => 'u', 'Ů' => 'u', | |
| 'ň' => 'n', 'ú' => 'u', 'ů' => 'u', | |
| // Greek alphabet & modern polytonic characters | |
| 'Α' => 'a', 'Β' => 'v', 'Γ' => 'g', 'Δ' => 'd', 'Ε' => 'e', | |
| 'α' => 'a', 'β' => 'v', 'γ' => 'g', 'δ' => 'd', 'ε' => 'e', | |
| 'Ζ' => 'z', 'Η' => 'i', 'Θ' => 'th', 'Ι' => 'i', 'Κ' => 'k', | |
| 'ζ' => 'z', 'η' => 'i', 'θ' => 'th', 'ι' => 'i', 'κ' => 'k', | |
| 'Λ' => 'l', 'Μ' => 'm', 'Ν' => 'n', 'Ξ' => 'x', 'Ο' => 'o', | |
| 'λ' => 'l', 'μ' => 'm', 'ν' => 'n', 'ξ' => 'x', 'ο' => 'o', | |
| 'Π' => 'p', 'Ρ' => 'r', 'Σ' => 's', 'Τ' => 't', 'Υ' => 'y', | |
| 'π' => 'p', 'ρ' => 'r', 'σ' => 's', 'τ' => 't', 'υ' => 'y', | |
| 'Φ' => 'f', 'Χ' => 'ch', 'Ψ' => 'ps', 'Ω' => 'o', 'Ά' => 'a', | |
| 'φ' => 'f', 'χ' => 'ch', 'ψ' => 'ps', 'ω' => 'o', 'ά' => 'a', | |
| 'Έ' => 'e', 'Ή' => 'i', 'Ί' => 'i', 'Ό' => 'o', 'Ύ' => 'y', | |
| 'έ' => 'e', 'ή' => 'i', 'ί' => 'i', 'ό' => 'o', 'ύ' => 'y', | |
| 'Ώ' => 'o', 'Ϊ' => 'i', 'Ϋ' => 'y', | |
| 'ώ' => 'o', 'ς' => 's', 'ΐ' => 'i', 'ϊ' => 'i', 'ϋ' => 'y', 'ΰ' => 'y', | |
| // Extra all (http:www.atm.ox.ac.ukuseriwicharmap.html) | |
| 'À' => 'a', 'Á' => 'a', 'Â' => 'a', 'Ã' => 'a', 'Å' => 'a', | |
| 'à' => 'a', 'á' => 'a', 'â' => 'a', 'ã' => 'a', 'å' => 'a', | |
| 'Æ' => 'ae', 'Ç' => 'c', 'È' => 'e', 'É' => 'e', 'Ê' => 'e', | |
| 'æ' => 'ae', 'ç' => 'c', 'è' => 'e', 'é' => 'e', 'ê' => 'e', | |
| 'Ë' => 'e', 'Ì' => 'i', 'Í' => 'i', 'Î' => 'i', 'Ï' => 'i', | |
| 'ë' => 'e', 'ì' => 'i', 'í' => 'i', 'î' => 'i', 'ï' => 'i', | |
| 'Ð' => 'd', 'Ñ' => 'n', 'Ò' => 'o', 'Ô' => 'o', 'Õ' => 'o', | |
| 'ð' => 'd', 'ñ' => 'n', 'ò' => 'o', 'ô' => 'o', 'õ' => 'o', | |
| '×' => 'x', 'Ø' => 'o', 'Ù' => 'u', 'Ú' => 'u', 'Û' => 'u', | |
| '×' => 'x', 'ø' => 'o', 'ù' => 'u', 'ú' => 'u', 'û' => 'u', | |
| 'Þ' => 'p', 'Ÿ' => 'y', | |
| 'þ' => 'p', 'ÿ' => 'y', | |
| // Other | |
| '№' => '', '“' => '', '”' => '', '«' => '', '»' => '', | |
| '„' => '', '@' => '', '%' => '', '‘' => '', '’' => '', | |
| '`' => '', '´' => '', 'º' => 'o', 'ª' => 'a', | |
| ); | |
| $name = remove_accents ($name); | |
| $name = strtr ($name, $chars_table); | |
| $name = urlencode ($name); | |
| // Перезапись % на ничего после urlencode | |
| $name = str_replace (array ('%'), '', $name ); | |
| /* | |
| * Проверка на длину файла | |
| * UNIX поддерживает 255 символов в имени максимум. | |
| * Мы обрезаем длину имени файла до 130 символов, как и на wordpress.com, | |
| * WordPress добавляет дополнительные символы в имя файла, например | |
| * image-1000x1000.jpg (поэтому мы резервируем половину возможной длины). | |
| */ | |
| $info = pathinfo ($name); | |
| $name = substr ($info ['filename'], 0, 130); | |
| $name .= '.' . $info ['extension']; | |
| return $name; | |
| } | |
| add_filter ('sanitize_file_name', 'selena_network_sanitize_file_name', 99, 2); | |
| ?> |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment