php正则匹配截取json

Question

请教一个正则截取问题

$script='<script type="text/javascript">
cgiData={
pageIdx : 0,
pageCount : 1,
pageSize : 100,
groupsList : ({"groups":[{"id":0,"name":"未分组","cnt":16},{"id":1,"name":"黑名单","cnt":0},{"id":2,"name":"星标组","cnt":0}]}).groups,
        friendsList : ({"contacts":[{"id":6,"nick_name":"yan","remark_name":"","group_id":0},{"id":8,"nick_name":"mm","remark_name":"","group_id":0},{"id":4,"nick_name":"gg","remark_name":"","group_id":0},{"id":2,"nick_name":"sa","remark_name":"","group_id":0}]}).contacts,
            currentGroupId : 0,
        type : 0,
    userRole : 1 * 1,
    verifyMsgCount :  * 1
};
seajs.use("user/index");
</script>';

怎么把contacts中间的[{"id":6,"nickname":"yan","remarkname":"","groupid":0},{"id":8,"nickname":"mm","remarkname":"","groupid":0},{"id":4,"nickname":"gg","remarkname":"","groupid":0},{"id":2,"nickname":"sa","remarkname":"","groupid":0}]这段截取出来？

Answer 1

(?<="contacts"\:\[){.*}(?=\].*\.contacts)

注：PHP里好像需要把"转义一下变成\"
顺便推荐一个正则参考用的网址：http://deerchao.net/tutorials/regex/regex.htm

Answer 2

PHP的DEMO：

preg_match_all("/\[{.*?\}]/is", $script, $matches);
$result = json_decode($matches[0][1]);

这个正则不是最佳方案，不过结果是你想要的就是。

Answer 3

我觉得没必要把中间的截取出来，

• 很简单的就可以把前后两个script剥离
• json_encode剩下的字符串
• 访问你需要的值

------------------ update ------------------

preg_match_all("/(?<=contacts\":)\s*\[\s*\{(.*?)\]/", $script, $matches);
var_dump($matches);

结果:

array(1) {
  [0]=>
  array(1) {
    [0]=>
    string(226) "[{"id":6,"nick_name":"yan","remark_name":"","group_id":0},{"id":8,"nick_name":"mm","remark_name":"","group_id":0},{"id":4,"nick_name":"gg","remark_name":"","group_id":0},{"id":2,"nick_name":"sa","remark_name":"","group_id":0}]"
  }
}

关于那个正则表达式你可以在这个网站中优化：
debuggex