php采集页面指定标签里面的内容

/*
* 参数说明: $tag_id:所要获取的元素Tag Id $url:所要获取页面的Url $tag:所要获取的标签 $data
*/
function getWebTag($tag_id, $url = false, $tag = 'div', $data = false) {
    if ($url !== false) {
        $data = file_get_contents ( $url );
    }
    $charset_pos = stripos ( $data, 'charset' );
    if ($charset_pos) {
        if (stripos ( $data, 'utf-8', $charset_pos )) {
            $data = iconv ( 'utf-8', 'utf-8', $data );
        } else if (stripos ( $data, 'gb2312', $charset_pos )) {
            $data = iconv ( 'gb2312', 'utf-8', $data );
        } else if (stripos ( $data, 'gbk', $charset_pos )) {
            $data = iconv ( 'gbk', 'utf-8', $data );
        }
    }
    preg_match_all ( '/<' . $tag . '/i', $data, $pre_matches, PREG_OFFSET_CAPTURE ); // 获取所有div前缀
    preg_match_all ( '/<\/' . $tag . '/i', $data, $suf_matches, PREG_OFFSET_CAPTURE ); // 获取所有div后缀
    $hit = strpos ( $data, $tag_id );
    if ($hit == - 1)
        return false; // 未命中
    $divs = array (); // 合并所有div
    foreach ( $pre_matches [0] as $index => $pre_div ) {
        $divs [( int ) $pre_div [1]] = 'p';
        $divs [( int ) $suf_matches [0] [$index] [1]] = 's';
    }
    // 对div进行排序
    $sort = array_keys ( $divs );
    asort ( $sort );
    $count = count ( $pre_matches [0] );
    foreach ( $pre_matches [0] as $index => $pre_div ) {
        // <div $hit <div+1 时div被命中
        if (($pre_matches [0] [$index] [1] < $hit) && ($hit < $pre_matches [0] [$index + 1] [1])) {
            $deeper = 0;
            // 弹出被命中div前的div
            while ( array_shift ( $sort ) != $pre_matches [0] [$index] [1] && ($count --) )
                continue;
                // 对剩余div进行匹配,若下一个为前缀,则向下一层,$deeper加1,
                // 否则后退一层,$deeper减1,$deeper为0则命中匹配,计算div长度
            foreach ( $sort as $key ) {
                if ($divs [$key] == 'p')
                    $deeper ++;
                else if ($deeper == 0) {
                    $length = $key - $pre_matches [0] [$index] [1];
                    break;
                } else {
                    $deeper --;
                }
            }
            $hitDivString = substr ( $data, $pre_matches [0] [$index] [1], $length ) . '</' . $tag . '>';
            break;
        }
    }
    return $hitDivString;
}
 
echo getWebTag ( 'id="content"', 'http://www.cdlyh.com/article/read/id/12345.html', 'div' );

 

上一篇:Code::Blocks/Sublime中文乱码问题


下一篇:如何解决表单提交的中文乱码问题