projects / weblabels.fsf.org.git / blame
| Commit | Line | Data |
|---|---|---|
| 7f254ad8 AE |
1 | <?php\r |
| 2 | \r | |
| 3 | /**\r | |
| 4 | * Class that handles operations involving percent-encoding in URIs.\r | |
| 5 | *\r | |
| 6 | * @warning\r | |
| 7 | * Be careful when reusing instances of PercentEncoder. The object\r | |
| 8 | * you use for normalize() SHOULD NOT be used for encode(), or\r | |
| 9 | * vice-versa.\r | |
| 10 | */\r | |
| 11 | class HTMLPurifier_PercentEncoder\r | |
| 12 | {\r | |
| 13 | \r | |
| 14 | /**\r | |
| 15 | * Reserved characters to preserve when using encode().\r | |
| 16 | */\r | |
| 17 | protected $preserve = array();\r | |
| 18 | \r | |
| 19 | /**\r | |
| 20 | * String of characters that should be preserved while using encode().\r | |
| 21 | */\r | |
| 22 | public function __construct($preserve = false) {\r | |
| 23 | // unreserved letters, ought to const-ify\r | |
| 24 | for ($i = 48; $i <= 57; $i++) $this->preserve[$i] = true; // digits\r | |
| 25 | for ($i = 65; $i <= 90; $i++) $this->preserve[$i] = true; // upper-case\r | |
| 26 | for ($i = 97; $i <= 122; $i++) $this->preserve[$i] = true; // lower-case\r | |
| 27 | $this->preserve[45] = true; // Dash -\r | |
| 28 | $this->preserve[46] = true; // Period .\r | |
| 29 | $this->preserve[95] = true; // Underscore _\r | |
| 30 | $this->preserve[126]= true; // Tilde ~\r | |
| 31 | \r | |
| 32 | // extra letters not to escape\r | |
| 33 | if ($preserve !== false) {\r | |
| 34 | for ($i = 0, $c = strlen($preserve); $i < $c; $i++) {\r | |
| 35 | $this->preserve[ord($preserve[$i])] = true;\r | |
| 36 | }\r | |
| 37 | }\r | |
| 38 | }\r | |
| 39 | \r | |
| 40 | /**\r | |
| 41 | * Our replacement for urlencode, it encodes all non-reserved characters,\r | |
| 42 | * as well as any extra characters that were instructed to be preserved.\r | |
| 43 | * @note\r | |
| 44 | * Assumes that the string has already been normalized, making any\r | |
| 45 | * and all percent escape sequences valid. Percents will not be\r | |
| 46 | * re-escaped, regardless of their status in $preserve\r | |
| 47 | * @param $string String to be encoded\r | |
| 48 | * @return Encoded string.\r | |
| 49 | */\r | |
| 50 | public function encode($string) {\r | |
| 51 | $ret = '';\r | |
| 52 | for ($i = 0, $c = strlen($string); $i < $c; $i++) {\r | |
| 53 | if ($string[$i] !== '%' && !isset($this->preserve[$int = ord($string[$i])]) ) {\r | |
| 54 | $ret .= '%' . sprintf('%02X', $int);\r | |
| 55 | } else {\r | |
| 56 | $ret .= $string[$i];\r | |
| 57 | }\r | |
| 58 | }\r | |
| 59 | return $ret;\r | |
| 60 | }\r | |
| 61 | \r | |
| 62 | /**\r | |
| 63 | * Fix up percent-encoding by decoding unreserved characters and normalizing.\r | |
| 64 | * @warning This function is affected by $preserve, even though the\r | |
| 65 | * usual desired behavior is for this not to preserve those\r | |
| 66 | * characters. Be careful when reusing instances of PercentEncoder!\r | |
| 67 | * @param $string String to normalize\r | |
| 68 | */\r | |
| 69 | public function normalize($string) {\r | |
| 70 | if ($string == '') return '';\r | |
| 71 | $parts = explode('%', $string);\r | |
| 72 | $ret = array_shift($parts);\r | |
| 73 | foreach ($parts as $part) {\r | |
| 74 | $length = strlen($part);\r | |
| 75 | if ($length < 2) {\r | |
| 76 | $ret .= '%25' . $part;\r | |
| 77 | continue;\r | |
| 78 | }\r | |
| 79 | $encoding = substr($part, 0, 2);\r | |
| 80 | $text = substr($part, 2);\r | |
| 81 | if (!ctype_xdigit($encoding)) {\r | |
| 82 | $ret .= '%25' . $part;\r | |
| 83 | continue;\r | |
| 84 | }\r | |
| 85 | $int = hexdec($encoding);\r | |
| 86 | if (isset($this->preserve[$int])) {\r | |
| 87 | $ret .= chr($int) . $text;\r | |
| 88 | continue;\r | |
| 89 | }\r | |
| 90 | $encoding = strtoupper($encoding);\r | |
| 91 | $ret .= '%' . $encoding . $text;\r | |
| 92 | }\r | |
| 93 | return $ret;\r | |
| 94 | }\r | |
| 95 | \r | |
| 96 | }\r | |
| 97 | \r | |
| 98 | // vim: et sw=4 sts=4\r |