HILFE! Filenames für Windows oder Linux(?) kastrieren!

[Felix Riesterer]

Lieber Tom,

ich habe mir mit Hilfe des SELFHTML-Forumsarchivs folgende Klasse gebaut. Vielleicht nützt sie Dir ja etwas?

<?php  
/**  
 * UTF-8 Ensurer  
 *  
 * This class is for making sure some string data is  
 * encoded in UTF-8.  
 *  
 * Felix Riesterer (http://felix-riesterer.de)  
 */  
  
  
class UTF8_Ensurer {  
	public  function ensure ($s) {  
		if (is_string($s) && !$this->is_utf8($s)) {  
			$s = utf8_encode($s);  
		}  
  
		return $s;  
	}  
  
	public  function iso ($s) {  
		if (is_string($s) && $this->is_utf8($s)) {  
			$s = utf8_decode($s);  
		}  
  
		return $s;  
	}  
  
	/* check for UTF-8 encoding: the following set of functions  
	 * has been taken (and adapted) from SELFHTML-Forum archive  
	 * http://forum.de.selfhtml.org/archiv/2005/10/t116805/#m747567  
	 */  
	public  function is_utf8 ($s) {  
		$t = $this;  
  
		$len = strlen($s);  
		$i = 0;  
  
		while ($i < $len) {  
  
			$c = ord($s{$i++});  
  
			if ($t->valid_1byte($c)) { // continue  
				continue;  
  
			} elseif ($t->valid_2byte($c)) { // check 1 byte  
  
				if ($i >= $len || !$t->valid_nextbyte(ord($s{$i++})))  
					return false;  
  
			} elseif ($t->valid_3byte($c)) { // check 2 bytes  
  
				if ($i >= $len || !$t->valid_nextbyte(ord($s{$i++})))  
					return false;  
  
				if ($i >= $len || !$t->valid_nextbyte(ord($s{$i++})))  
					return false;  
  
			} elseif ($t->valid_4byte($c)) { // check 3 bytes  
  
				if ($i >= $len || !$t->valid_nextbyte(ord($s{$i++})))  
					return false;  
  
				if ($i >= $len || !$t->valid_nextbyte(ord($s{$i++})))  
					return false;  
  
				if ($i >= $len || !$t->valid_nextbyte(ord($s{$i++})))  
					return false;  
  
			} else {  
  
				return false; // 10xxxxxx occuring alone  
			} // goto next char  
		}  
  
		return true; // done  
	}  
  
	private function valid_1byte ($c) {  
		if (!is_int($c))  
			return false;  
		return ($c & 0x80)==0x00;  
	}  
  
	private function valid_2byte ($c) {  
		if (!is_int($c))  
			return false;  
		return ($c & 0xE0)==0xC0;  
	}  
  
	private function valid_3byte ($c) {  
		if (!is_int($c))  
			return false;  
		return ($c & 0xF0)==0xE0;  
	}  
  
	private function valid_4byte ($c) {  
		if (!is_int($c))  
			return false;  
		return ($c & 0xF8)==0xF0;  
	}  
  
	private function valid_nextbyte ($c) {  
		if (!is_int($c))  
			return false;  
		return ($c & 0xC0)==0x80;  
	}  
}  
  
?>

Dazu noch dieses Schmankerl aus eigener Produktion:

	public  function normalize_utf8_to_lower_case ($s) {  
		// replacements for lower-case ASCII-characters  
		$r = array(  
			'a'  => 'AaÀÁÂÃÅàáâãåĀāĂ㥹ǍǎǺǻẠạẢảẤấẦầẨẩẪẫẬậẮắẰằẲẳẴẵẶặΆ',  
			'ae' => 'ÄäæÆǼǽ',  
			'b'  => 'bB',  
			'c'  => 'CcÇçĆćĈĉĊċČč',  
			'd'  => 'DdÐĎďĐđ',  
			'e'  => 'EeÈÉÊËèéêëĒēĔĕĖėĘęĚěẸẹẺẻẼẽẾếỀềỂểỄễỆệΈ',  
			'f'  => 'Ffƒ',  
			'g'  => 'GgĜĝĞğĠġĢģ',  
			'h'  => 'HhĤĥĦħΉ',  
			'i'  => 'IiÌÍÎÏìíîïĨĩĪīĬĭĮįİıǏǐỈỉỊịΊΐ',  
			'ij' => 'IJij',  
			'j'  => 'JjĴĵ',  
			'k'  => 'KkĶķĸ',  
			'l'  => 'LlĹĺĻļĽľĿŀŁł',  
			'm'  => 'Mm',  
			'n'  => 'NnÑñŃńŅņŇňʼn',  
			'ng' => 'Ŋŋ',  
			'o'  => 'OoÒÓÔÕØòóôõøŌōŎŏŐőƠơǑǒǾǿΌỌọỎỏỐốỒồỔổỖỗỘộỚớỜờỞởỠỡỢợ',  
			'oe' => 'ÖöŒœ',  
			'p'  => 'Pp',  
			'q'  => 'Qq',  
			'r'  => 'RrŔŕŖŗŘř',  
			's'  => 'SsŚśŜŝŞşŠš',  
			'ss' => 'ßß',  
			't'  => 'TtŢţŤťŦŧ',  
			'u'  => 'UuÙÚÛŨũŪūŬŭŮůŰűŲųƯưǓǔǕǖǗǘǙǚǛǜùúûỤụỦủỨứỪừỬửỮữỰự',  
			'ue' => 'Üü',  
			'v'  => 'Vv',  
			'w'  => 'WwŴŵẀẁẂẃẄẅ',  
			'x'  => 'Xx',  
			'y'  => 'YyÝýÿŶŷŸΎỲỳỴỵỶỷỸỹ',  
			'z'  => 'ZzŹźŻżŽž'  
		);  
  
		// replace  
		foreach ($r as $c => $variants) {  
			$s = preg_replace("~[$variants]~su", $c, $s);  
		}  
  
		return $s;  
	}

Wenn es Dein Problem nicht löst, dann brauchst Du vielleicht eine neue Betrachtungsweise darauf?

Liebe Grüße,

Felix Riesterer.