在數據采集與頁面分析中,常需要抓取給定url頁面的內容,或者第二、第三層次深度頁面內容。
這裡是一個測試例子的實現,僅供參考。
代碼如下:
/*
匹配給定頁面鏈接
return:array match[link,content,all]
*/
function match_links($host, $document) {
$pattern = '/<a(.*?)href="(.*?)"(.*?)>(.*?)</a>/i';
preg_match_all($pattern, $document, $m);
return $m;
preg_match_all("'<s*as.*?hrefs*=s*(["'])?(?(1)(.*?)1|([^s>]+))[^>]*>?(.*?)</a>'isx",$document,$links);
while(list($key,$val) = each($links[2])) {
if(!empty($val))
if(preg_match("/http/",$val)){
$match['link'][] = $val;
}
else {
$match['link'][] = $host . $val;
}
}
while(list($key,$val) = each($links[3])) {
if(!empty($val))
if(preg_match("/http/",$val)){
$match['link'][] = $val;
}
else {
$match['link'][] = $host . $val;
}
}
while(list($key,$val) = each($links[4])) {
if(!empty($val))
$match['content'][] = $val;
}
while(list($key,$val) = each($links[0])) {
if(!empty($val))
$match['all'][] = $val;
}
return $match['link'];
}
/*
從給定url中獲取頁面文本內容
*/
function get_content_from_url($url) {
$str = @file_get_contents($url);
if(mb_check_encoding($str, "GBK"))
$str = iconv("GBK","UTF-8", $str);
$str = strip_tags($str); // 過濾html標簽
/*
$str = preg_replace( "@<script(.*?)</script>@is", "", $str );
$str = preg_replace( "@<iframe(.*?)</iframe>@is", "", $str );
$str = preg_replace( "@<style(.*?)</style>@is", "", $str );
$str = preg_replace( "@<(.*?)>@is", "", $str );
*/
//過濾非漢字字符
preg_match_all('/[x{4e00}-x{9fff}]+/u', $str, $matches);
$str = join(',', $matches[0]);
if(!$str)
return NULL;
return $str;
}
function get_content($url,$depth) {
if(!$url || $depth < 1)
return false;
while($depth > 1){
$str = @file_get_contents($url);
if(!$str)
return false;
$parseurl = parse_url($url);
if($parseurl['host'])
$host = $parseurl[scheme] . "://" . $parseurl['host'];
$arrlink = match_links($host,$str);
$arr_url = array_unique($arrlink);
$depth--;
foreach($arr_url as $url){
$content .= get_content($url, $depth); //遞歸調用
}
}
$content .= get_content_from_url($url);
return $content;
}