高效的中文字符串截取函数
作者:徐祖宁
在php中使用传统的字符串截取函数substr处理含有中文字符的字符串时会出现汉字被切断的现象。当可以使用php扩展库时，我们可以用mb_substr代替。但是该扩展库在连接时有一定的困难——linux下需重新编译php，有时并不能可做到，更何况其冗余函数较多。
在网络上可以看到很多实现此功能的函数。但算法多是循环判断，当字符串较大时效率极低。
为此这里介绍两个高效的函数：c_substr、m_substr。他们的用法完全与substr和mb_substr相同。不同之处在于：c_substr按字节计算，即一个汉字的长度为2；m_substr按字计算，即一个汉字的长度为1。可根据需要选用。
代码:
function c_substr($str,$start=0) {
$ch = chr(127);
$p = array("/[\x81-\xfe]([\x81-\xfe]|[\x40-\xfe])/","/[\x01-\x77]/");
$r = array("","");
if(func_num_args() > 2)
$end = func_get_arg(2);
else
$end = strlen($str);
if($start < 0)
$start += $end;
if($start > 0) {
$s = substr($str,0,$start);
if($s[strlen($s)-1] > $ch) {
$s = preg_replace($p,$r,$s);
$start += strlen($s);
}
}
$s = substr($str,$start,$end);
$end = strlen($s);
if($s[$end-1] > $ch) {
$s = preg_replace($p,$r,$s);
$end += strlen($s);
}
return substr($str,$start,$end);
}
function m_substr($str,$start) {
preg_match_all("/[\x80-\xff]?./",$str,$ar);
if(func_num_args() >= 3) {
$end = func_get_arg(2);
return join("",array_slice($ar[0],$start,$end));
}else
return join("",array_slice($ar[0],$start));
}
$ch = chr(127);
$p = array("/[\x81-\xfe]([\x81-\xfe]|[\x40-\xfe])/","/[\x01-\x77]/");
$r = array("","");
if(func_num_args() > 2)
$end = func_get_arg(2);
else
$end = strlen($str);
if($start < 0)
$start += $end;
if($start > 0) {
$s = substr($str,0,$start);
if($s[strlen($s)-1] > $ch) {
$s = preg_replace($p,$r,$s);
$start += strlen($s);
}
}
$s = substr($str,$start,$end);
$end = strlen($s);
if($s[$end-1] > $ch) {
$s = preg_replace($p,$r,$s);
$end += strlen($s);
}
return substr($str,$start,$end);
}
function m_substr($str,$start) {
preg_match_all("/[\x80-\xff]?./",$str,$ar);
if(func_num_args() >= 3) {
$end = func_get_arg(2);
return join("",array_slice($ar[0],$start,$end));
}else
return join("",array_slice($ar[0],$start));
}
Windix Feng 也改写了一个UTF-8适用的，也一起记录。留待以后用再看看。
function utf8_substr2($str,$start) {
/*
UTF-8 version of substr(), for people who can't use mb_substr() like me.
Length is not the count of Bytes, but the count of UTF-8 Characters
Author: Windix Feng
Bug report to: windix(AT)263.net, http://www.douzi.org/blog
- History -
1.0 2004-02-01 Initial Version
2.0 2004-02-01 Use PREG instead of STRCMP and cycles, SPEED UP!
*/
preg_match_all("/[\x01-\x7f]|[\xc2-\xdf][\x80-\xbf]|\xe0[\xa0-\xbf][\x80-\xbf]|[\xe1-\xef][\x80-\xbf][\x80-\xbf]|\xf0[\x90-\xbf][\x80-\xbf][\x80-\xbf]|[\xf1-\xf7][\x80-\xbf][\x80-\xbf][\x80-\xbf]/", $str, $ar);
if(func_num_args() >= 3) {
$end = func_get_arg(2);
return join("",array_slice($ar[0],$start,$end));
} else {
return join("",array_slice($ar[0],$start));
}
}
/*
UTF-8 version of substr(), for people who can't use mb_substr() like me.
Length is not the count of Bytes, but the count of UTF-8 Characters
Author: Windix Feng
Bug report to: windix(AT)263.net, http://www.douzi.org/blog
- History -
1.0 2004-02-01 Initial Version
2.0 2004-02-01 Use PREG instead of STRCMP and cycles, SPEED UP!
*/
preg_match_all("/[\x01-\x7f]|[\xc2-\xdf][\x80-\xbf]|\xe0[\xa0-\xbf][\x80-\xbf]|[\xe1-\xef][\x80-\xbf][\x80-\xbf]|\xf0[\x90-\xbf][\x80-\xbf][\x80-\xbf]|[\xf1-\xf7][\x80-\xbf][\x80-\xbf][\x80-\xbf]/", $str, $ar);
if(func_num_args() >= 3) {
$end = func_get_arg(2);
return join("",array_slice($ar[0],$start,$end));
} else {
return join("",array_slice($ar[0],$start));
}
}
2004年04月23日 | 分类归档: Tech
评论