|
一,、編碼范圍 1. GBK (GB2312/GB18030) 2. UTF-8 (Unicode)
二,、代碼例子 //判斷內(nèi)容里有沒(méi)有中文-GBK (PHP)
function check_is_chinese($s){ return preg_match('/[\x80-\xff]./', $s); } //獲取字符串長(zhǎng)度-GBK (PHP) function gb_strlen($str){ $count = 0; for($i=0; $i<strlen($str); $i++){ $s = substr($str, $i, 1); if (preg_match("/[\x80-\xff]/", $s)) ++$i; ++$count; } return $count; } //截取字符串字串-GBK (PHP) function gb_substr($str, $len){ $count = 0; for($i=0; $i<strlen($str); $i++){ if($count == $len) break; if(preg_match("/[\x80-\xff]/", substr($str, $i, 1))) ++$i; ++$count; } return substr($str, 0, $i); } //統(tǒng)計(jì)字符串長(zhǎng)度-UTF8 (PHP) function utf8_strlen($str) { $count = 0; for($i = 0; $i < strlen($str); $i++){ $value = ord($str[$i]); if($value > 127) { $count++; if($value >= 192 && $value <= 223) $i++; elseif($value >= 224 && $value <= 239) $i = $i + 2; elseif($value >= 240 && $value <= 247) $i = $i + 3; else die('Not a UTF-8 compatible string'); } $count++; } return $count; } //截取字符串-UTF8(PHP) function utf8_substr($str,$position,$length){ $start_position = strlen($str); $start_byte = 0; $end_position = strlen($str); $count = 0; for($i = 0; $i < strlen($str); $i++){ if($count >= $position && $start_position > $i){ $start_position = $i; $start_byte = $count; } if(($count-$start_byte)>=$length) { $end_position = $i; break; } $value = ord($str[$i]); if($value > 127){ $count++; if($value >= 192 && $value <= 223) $i++; elseif($value >= 224 && $value <= 239) $i = $i + 2; elseif($value >= 240 && $value <= 247) $i = $i + 3; else die('Not a UTF-8 compatible string'); } $count++; } return(substr($str,$start_position,$end_position-$start_position)); } //字符串長(zhǎng)度統(tǒng)計(jì)-UTF8 [中文3個(gè)字節(jié)，俄文,、韓文占2個(gè)字節(jié),，字母占1個(gè)字節(jié)] (Ruby) def utf8_string_length(str) temp = CGI::unescape(str) i = 0; j = 0; temp.length.times{|t| if temp[t] < 127 i += 1 elseif temp[t] >= 127 and temp[t] < 224 j += 1 if 0 == (j % 2) i += 2 j = 0 end else j += 1 if 0 == (j % 3) i +=2 j = 0 end end } return i } //判斷是否是有韓文-UTF-8 (JavaScript) function checkKoreaChar(str) { for(i=0; i<str.length; i++) { if(((str.charCodeAt(i) > 0x3130 && str.charCodeAt(i) < 0x318F) || (str.charCodeAt(i) >= 0xAC00 && str.charCodeAt(i) <= 0xD7A3))) { return true; } } return false; } //判斷是否有中文字符-GBK (JavaScript)
三,、參考文檔 http://www./ 另： /* |
|
|
