Browse through the history of space, crawl out of the person viewing your space (typically restricting 20 people, unless you turn on yellow diamonds), and then crawl out of these 20 people's browsing history and crawl down, you can control the crawl depth.
Here just to give the fear of the code snippet, you can further optimize the QQ classification store. Through QQ Mutual browsing relationship, you can draw a friend network through drawing tools. Wait a minute
Code involves pthreads if not clear read: PHP Advanced programming Multi-Threading
Http://netkiller.github.io/journal/thread.php.html
<?PHP/*Homepage:http://netkiller.github.ioauthor:neo <[email protected]>*/if(!extension_loaded(' Pthreads ')) die(' Please install pthreads '); include_once(' Snoopy.class.php '); classCrawlerworkerextendsWorker {protected Static $DBH; Public function__construct () {} Public functionrun () {/*$dbhost = ' db.example.com '; Database server $dbuser = ' example.com '; Database user name $DBPW = ' password '; Database Password $dbname = ' example '; Database name self:: $DBH = new PDO ("mysql:host= $dbhost;p ort=3306;dbname= $dbname", $dbuser, $DBPW, Array (PDO :: Mysql_attr_init_command = ' SET NAMES \ ' utf8\ ', pdo::mysql_attr_compress = True, pdo::attr _persistent = true)); */ } protected functiongetinstance () {returnSelf::$DBH; } } /*The collectable class implements machinery for Pool::collect*/classCrawlerextendsStackable { Public $depth= 3; Private Static $level= 0; Public function__construct ($QQ) { $this->QQ =$QQ; } Public functionrun () {Try { $DBH=$this->worker->getinstance (); $this->recursion (Array($this-QQ)); } Catch(pdoexception$e) { $error=sprintf("%s,%s\n",$mobile,$id ); file_put_contents("Mobile_error.log",$error,file_append); } //printf ("Runtime:%s,%s\n", Date (' y-m-d h:i:s '), $this->worker->getthreadid ()); $lst = $this->qzone ($this->qq); Print_r ($LST); } Public functionRecursion ($qqs){ if(Self::$level<=$this-depth) { Self::$level++; }Else if(Self::$level> 0) { self::$level--; } printf(' Level:%s\n ', Self::$level); //sleep (1); Usleep(Mt_rand(10000,1000000)); if(Self::$level>=$this-depth) { return; } foreach($qqs as $uin) { $lst=$this->qzone ($uin); Print_r($lst); $this->recursion ($lst); } } Public functionQzone$QQ){ $url= ' http://m.qzone.com/mqz_get_visitor?g_tk=1191852101&res_mode=0&res_uin= '.$QQ.‘ &offset=0&count=100&page=1&format=json&t=1401762986882&sid= Dodkvcyv6azjn87cxxq5mao1xgakyjhg18c8aa5e0201%3d%3d '; $snoopy=NewSnoopy; //need an proxy? $snoopy->proxy_host = "My.proxy.host"; $snoopy->proxy_port = "8080"; Set browser and Referer: $snoopy->agent = "mozilla/4.0 (compatible; MSIE 6.0; Windows NT 5.1) "; $snoopy->referer = "http://m.qzone.com/"; //set Some cookies://$snoopy->cookies["SessionID"] = ' 238472834723489 '; $snoopy->cookies["FavoriteColor"] = "blue"; Set an raw-header: $snoopy->rawheaders["Pragma"] = "No-cache"; //set some internal variables: $snoopy->maxredirs = 2; $snoopy->offsiteok =false; $snoopy->expandlinks =false; //set username and password (optional)//$snoopy->user = "Joe"; $snoopy->pass = "Bloe"; Fetch the text of the website www.google.com: if($snoopy->fetchtext ($url)){ //Other Methods:fetch, Fetchform, Fetchlinks, Submittext and submitlinks//Response code://PR int "Response code:". $snoopy->response_code. " <br/>\n "; Print the headers://print "<b>Headers:</b><br/>"; while (list ($key, $val) = each ($snoopy->headers)) {//Print $key. ":". $val. " <br/>\n "; }//Print the texts of the website://print_r (Json_decode ($snoopy->results)); $results=Array(); $tmp= Json_decode ($snoopy-results); if($tmp){ if(Property_exists ($tmp, ' Data ')){ foreach($tmp->data->List as $lst ){ $results[] =$lst-UIn; } } } return($results); } Else { Print"Snoopy:error while fetching document:".$snoopy->error. " \ n "; } }} $pool=NewPool (\crawlerworker::class, []); #foreach (range (100000) as $number) {#$pool->submit (New Crawler ($number));#} $pool->submit (NewCrawler (' 13721218 '));$pool->submit (NewCrawler (' 291379 '));//$pool->submit (new Crawler (' xxx ')),//$pool->submit (new Crawler (' xxx '));//$pool->submit (' New Crawler (' xxx '));//And so on//$pool->submit (New Crawler (' nnn ')); $pool-shutdown ();?>
PHP multithreading pthreads implementation QQ number crawler