From 0ae8cbf5c0b1a198b963490985b7738392ebcb97 Mon Sep 17 00:00:00 2001 From: ahriman Date: Mon, 3 Dec 2018 19:22:25 -0500 Subject: installed dokuwiki, added to navbar, updated news --- wiki/inc/SafeFN.class.php | 158 ++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 158 insertions(+) create mode 100644 wiki/inc/SafeFN.class.php (limited to 'wiki/inc/SafeFN.class.php') diff --git a/wiki/inc/SafeFN.class.php b/wiki/inc/SafeFN.class.php new file mode 100644 index 0000000..b9e4a2b --- /dev/null +++ b/wiki/inc/SafeFN.class.php @@ -0,0 +1,158 @@ + + * @date 2010-04-02 + */ +class SafeFN { + + // 'safe' characters are a superset of $plain, $pre_indicator and $post_indicator + private static $plain = '-./[_0123456789abcdefghijklmnopqrstuvwxyz'; // these characters aren't converted + private static $pre_indicator = '%'; + private static $post_indicator = ']'; + + /** + * Convert an UTF-8 string to a safe ASCII String + * + * conversion process + * - if codepoint is a plain or post_indicator character, + * - if previous character was "converted", append post_indicator to output, clear "converted" flag + * - append ascii byte for character to output + * (continue to next character) + * + * - if codepoint is a pre_indicator character, + * - append ascii byte for character to output, set "converted" flag + * (continue to next character) + * + * (all remaining characters) + * - reduce codepoint value for non-printable ASCII characters (0x00 - 0x1f). Space becomes our zero. + * - convert reduced value to base36 (0-9a-z) + * - append $pre_indicator characater followed by base36 string to output, set converted flag + * (continue to next character) + * + * @param string $filename a utf8 string, should only include printable characters - not 0x00-0x1f + * @return string an encoded representation of $filename using only 'safe' ASCII characters + * + * @author Christopher Smith + */ + public static function encode($filename) { + return self::unicode_to_safe(utf8_to_unicode($filename)); + } + + /** + * decoding process + * - split the string into substrings at any occurrence of pre or post indicator characters + * - check the first character of the substring + * - if its not a pre_indicator character + * - if previous character was converted, skip over post_indicator character + * - copy codepoint values of remaining characters to the output array + * - clear any converted flag + * (continue to next substring) + * + * _ else (its a pre_indicator character) + * - if string length is 1, copy the post_indicator character to the output array + * (continue to next substring) + * + * - else (string length > 1) + * - skip the pre-indicator character and convert remaining string from base36 to base10 + * - increase codepoint value for non-printable ASCII characters (add 0x20) + * - append codepoint to output array + * (continue to next substring) + * + * @param string $filename a 'safe' encoded ASCII string, + * @return string decoded utf8 representation of $filename + * + * @author Christopher Smith + */ + public static function decode($filename) { + return unicode_to_utf8(self::safe_to_unicode(strtolower($filename))); + } + + public static function validate_printable_utf8($printable_utf8) { + return !preg_match('#[\x01-\x1f]#',$printable_utf8); + } + + public static function validate_safe($safe) { + return !preg_match('#[^'.self::$plain.self::$post_indicator.self::$pre_indicator.']#',$safe); + } + + /** + * convert an array of unicode codepoints into 'safe_filename' format + * + * @param array int $unicode an array of unicode codepoints + * @return string the unicode represented in 'safe_filename' format + * + * @author Christopher Smith + */ + private static function unicode_to_safe($unicode) { + + $safe = ''; + $converted = false; + + foreach ($unicode as $codepoint) { + if ($codepoint < 127 && (strpos(self::$plain.self::$post_indicator,chr($codepoint))!==false)) { + if ($converted) { + $safe .= self::$post_indicator; + $converted = false; + } + $safe .= chr($codepoint); + + } else if ($codepoint == ord(self::$pre_indicator)) { + $safe .= self::$pre_indicator; + $converted = true; + } else { + $safe .= self::$pre_indicator.base_convert((string)($codepoint-32),10,36); + $converted = true; + } + } + if($converted) $safe .= self::$post_indicator; + return $safe; + } + + /** + * convert a 'safe_filename' string into an array of unicode codepoints + * + * @param string $safe a filename in 'safe_filename' format + * @return array int an array of unicode codepoints + * + * @author Christopher Smith + */ + private static function safe_to_unicode($safe) { + + $unicode = array(); + $split = preg_split('#(?=['.self::$post_indicator.self::$pre_indicator.'])#',$safe,-1,PREG_SPLIT_NO_EMPTY); + + $converted = false; + foreach ($split as $sub) { + $len = strlen($sub); + if ($sub[0] != self::$pre_indicator) { + // plain (unconverted) characters, optionally starting with a post_indicator + // set initial value to skip any post_indicator + for ($i=($converted?1:0); $i < $len; $i++) { + $unicode[] = ord($sub[$i]); + } + $converted = false; + } else if ($len==1) { + // a pre_indicator character in the real data + $unicode[] = ord($sub); + $converted = true; + } else { + // a single codepoint in base36, adjusted for initial 32 non-printable chars + $unicode[] = 32 + (int)base_convert(substr($sub,1),36,10); + $converted = true; + } + } + + return $unicode; + } + +} -- cgit 1.4.1-2-gfad0