| | |
| | | * @param string Input charset name |
| | | * @return The validated charset name |
| | | */ |
| | | function rcube_parse_charset($charset) |
| | | function rcube_parse_charset($input) |
| | | { |
| | | $charset = strtoupper($charset); |
| | | static $charsets = array(); |
| | | $charset = strtoupper($input); |
| | | |
| | | if (isset($charsets[$input])) |
| | | return $charsets[$input]; |
| | | |
| | | $charset = preg_replace(array( |
| | | '/^[^0-9A-Z]+/', // e.g. _ISO-8859-JP$SIO |
| | |
| | | $str = preg_replace(array('/[^A-Z0-9]/', '/^X+/'), '', $charset); |
| | | |
| | | if (isset($aliases[$str])) |
| | | return $aliases[$str]; |
| | | |
| | | if (preg_match('/U[A-Z][A-Z](7|8|16|32)(BE|LE)*/', $str, $m)) |
| | | return 'UTF-' . $m[1] . $m[2]; |
| | | |
| | | if (preg_match('/ISO8859([0-9]{0,2})/', $str, $m)) { |
| | | $result = $aliases[$str]; |
| | | // UTF |
| | | else if (preg_match('/U[A-Z][A-Z](7|8|16|32)(BE|LE)*/', $str, $m)) |
| | | $result = 'UTF-' . $m[1] . $m[2]; |
| | | // ISO-8859 |
| | | else if (preg_match('/ISO8859([0-9]{0,2})/', $str, $m)) { |
| | | $iso = 'ISO-8859-' . ($m[1] ? $m[1] : 1); |
| | | # some clients sends windows-1252 text as latin1, |
| | | # it is safe to use windows-1252 for all latin1 |
| | | return $iso == 'ISO-8859-1' ? 'WINDOWS-1252' : $iso; |
| | | // some clients sends windows-1252 text as latin1, |
| | | // it is safe to use windows-1252 for all latin1 |
| | | $result = $iso == 'ISO-8859-1' ? 'WINDOWS-1252' : $iso; |
| | | } |
| | | |
| | | // handle broken charset names e.g. WINDOWS-1250HTTP-EQUIVCONTENT-TYPE |
| | | if (preg_match('/(WIN|WINDOWS)([0-9]+)/', $str, $m)) { |
| | | return 'WINDOWS-' . $m[2]; |
| | | else if (preg_match('/(WIN|WINDOWS)([0-9]+)/', $str, $m)) { |
| | | $result = 'WINDOWS-' . $m[2]; |
| | | } |
| | | else { |
| | | $result = $charset; |
| | | } |
| | | |
| | | return $charset; |
| | | $charsets[$input] = $result; |
| | | |
| | | return $result; |
| | | } |
| | | |
| | | |