substr截取中文字符出现乱码的解决方案

作者：佚名   时间：2012-04-11   分享到：

这里提供一个函数可较好地解决substr遇到中文字符的问题。中文字符按2个长度单位来计算，使得中英文混用环境下字符串截取结果最后的显示长度接近；舍弃最后一个不完整字符，保证不会出现显示上的乱码；且兼容了中文字符常用的utf-8编码和GB2312编码，有很好的通用性。
 function getstr($string, $length, $encoding  = 'utf-8') {
    $string = trim($string);
 
    if($length && strlen($string) > $length) {
        //截断字符
        $wordscut = '';
        if(strtolower($encoding) == 'utf-8') {
            //utf8编码
            $n = 0;
            $tn = 0;
            $noc = 0;
            while ($n < strlen($string)) {
                $t = ord($string[$n]);
                if($t == 9 || $t == 10 || (32 <= $t && $t <= 126)) {
                    $tn = 1;
                    $n++;
                    $noc++;
                } elseif(194 <= $t && $t <= 223) {
                    $tn = 2;
                    $n += 2;
                    $noc += 2;
                } elseif(224 <= $t && $t < 239) {
                    $tn = 3;
                    $n += 3;
                    $noc += 2;
                } elseif(240 <= $t && $t <= 247) {
                    $tn = 4;
                    $n += 4;
                    $noc += 2;
                } elseif(248 <= $t && $t <= 251) {
                    $tn = 5;
                    $n += 5;
                    $noc += 2;
                } elseif($t == 252 || $t == 253) {
                    $tn = 6;
                    $n += 6;
                    $noc += 2;
                } else {
                    $n++;
                }
                if ($noc >= $length) {
                    break;
                }
            }
            if ($noc > $length) {
                $n -= $tn;
            }
            $wordscut = substr($string, 0, $n);
        } else {
            for($i = 0; $i < $length - 1; $i++) {
                if(ord($string[$i]) > 127) {
                    $wordscut .= $string[$i].$string[$i + 1];
                    $i++;
                } else {
                    $wordscut .= $string[$i];
                }
            }
        }
        $string = $wordscut;
    }
    return trim($string);
}
 
// 示例
echo getstr("0一二三四五六七",1).'
';  // 0
echo getstr("0一二三四五六七",2).'
';  // 0
echo getstr("0一二三四五六七",3).'
';  // 0一
echo getstr("0一二三四五六七",4).'
';  // 0一
echo getstr("0一二三四五六七",5).'
';  // 0一二
echo getstr("0一a二b三四五六七",1).'
';    // 0
echo getstr("0一a二b三四五六七",2).'
';    // 0
echo getstr("0一a二b三四五六七",3).'
';    // 0一
echo getstr("0一a二b三四五六七",4).'
';    // 0一a
echo getstr("0一a二b三四五六七",5).'
';    // 0一a