1
2
3
4
5
6
7
8
9
10
| // Filter English punctuation
$word = preg_replace("/[[:punct:]]/i", " ", $word);
// Filter Chinese punctuation
mb_regex_encoding('utf-8');
$char = "。、!?:;﹑•"…''""〝〞∕¦‖— 〈〉﹞﹝「」‹›〖〗】【»«』『〕〔》《﹐¸﹕︰﹔!¡?¿﹖﹌﹏﹋'´ˊˋ―﹫︳︴¯_ ̄﹢﹦﹤‐˜﹟﹩﹠﹪﹡﹨﹍﹉﹎﹊ˇ︵︶︷︸︹︿﹀︺︽︾ˉ﹁﹂﹃﹄︻︼()";
$word = mb_ereg_replace("[".$char."]", " ", $word, "UTF-8");
// Filter consecutive spaces
$word = preg_replace("/\s+/", " ", $word);
|
mb_ereg_replace is used for Chinese character replacement. When using regex, you don’t need to add / / delimiters.