Example #1
0
function UnikData($data = '', $db = 'titre', $size = 10)
{
    $res = '';
    $sup = $_ENV['ksup'];
    $s[] = "origindata:" . $data;
    if (RS == 404) {
        return;
    }
    if (!isgoodurl() or preg_match("~url.data:|image/png;|base64|/(adtech|iframeproxy)~", U) or e(',badurl', 1)) {
        return;
    }
    static $trig;
    if (!$trig) {
        Gt('unikdata:start');
        $trig = 1;
    }
    #Returns only unique Array Data ! For Titles + descriptions (1Mo + 13ms)
    if (strpos(u3, '.jpg')) {
        return;
    }
    $rev = 1342976292;
    Gt('udata');
    $u = $db . ':' . hu3;
    #Maj
    if ($y = param($data, '&')) {
        extract($y);
    }
    #extraction paramètres ...?
    $origin = Rem($data, CLT(surl));
    $data = trim(str_replace(array('»', 'Â', '£', 'N%253BO=D'), ' ', $data), '!?»,. ');
    if (!$_ENV['auth']) {
        $u = Preg_Replace("~(#|\\?|&).*~", '', $u);
    } else {
        $u = Preg_Replace("~[#|\\?|&][^(&|\\?)" . $_ENV['auth'] . "]+~", '', $u);
    }
    #!genius!really nice syntax-"~(#|\?|&)[^(".$_ENV['auth'].")].*~"
    while (strlen($data) < $size && $sup) {
        $next = Array_shift($sup);
        if (!stripos($data, $next)) {
            $data .= ' ' . $next;
        }
    }
    #push ksup inside
    if (strpos(SURL, '?', 0)) {
        Preg_match_all("~=([^&]+)~is", SURL, $t);
        if ($t) {
            $t = $t[1];
            $data .= ' ' . implode(' ', $t);
        }
    }
    #si titre similaire, pusher le query string
    $ret = $data;
    #av("data:$data db:$db ret:$ret".pre($s));
    ###Si pas de résultats : invalider le record en placant un "1" dessus
    if (1 or H == 'a74.fr') {
        $arg = 'no=1,bd=' . sip;
        $prev = array();
        #previously db -> comes to mysql  unikdata:id,url,db,data,score
        #CREATE TABLE `unikdata`(`id` INT(8) NOT NULL AUTO_INCREMENT PRIMARY KEY,`url` VARCHAR(255) NOT NULL,`data` TEXT NOT NULL,`score` INT(5) NOT NULL) ENGINE = MYISAM;#`db` VARCHAR(25) NOT NULL,
        $s[] = "select sql_cache data from " . DB . ".unikdata where url=\"{$u}\"";
        $x = sql5(['cache' => 1, 'sql' => end($s)], "bd=" . sip) or $noexist = 1;
        #if(strpos(U,'webcam.php')){print_r(compact('y','s','u','x'));die;}
        $s[] = "result:" . count($x) . "/" . $x . "/data:" . $data;
        if ($x == $data) {
            return $data;
        } else {
            #on défini le score puis On recherche si des correspondances existent déjà
            if (!Q && e('tprio,forcetitre', 1)) {
                $score[$u] = 0;
            } else {
                $score[$u] = substr_count(u, '/') * 20 + substr_count(u, '?') * 10 + substr_count(u, '&') * 10 + strlen(u);
            }
            /*Score de lui même*/
            $datas[$u] = $data;
            $s[] = "select sql_cache url,score from " . DB . ".unikdata where data=\"{$data}\" and url<>\"{$u}\" and left(url,6) rlike'{$db}:" . H . ".*' order by score asc";
            $y = sql5(['cache' => 1, 'sql' => end($s)], $arg);
            /*
            select * from ben.unikdata where url='titre:a74.fr/z/webcam.php';
            select sql_cache url,score from ben.unikdata where data="Webcams Ski Haute Savoie" and url<>"titre:a74.fr/z/webcam.php" and left(url,6) rlike'titre:a74.fr.*' order by score asc
            insert into ben.unikdata(url,data,score)VALUES("titre:a74.fr/z/webcam.php","Webcams Ski Haute Savoie","32")
            */
            #Si une ou plusieurs Pages have the same results
            if ($y) {
                if (!is_array($y)) {
                    $y = [$y];
                }
                foreach ($y as $t) {
                    $score[$t['url']] = $t['score'];
                    $datas[$t['url']] = $data;
                }
            }
            if (count($score) == 1) {
                if ($noexist) {
                    return $data;
                    $s[] = "insert ignore into " . DB . ".unikdata(url,data,score)VALUES(\"{$u}\",\"{$data}\",\"{$score[$u]}\")";
                    sql5(end($s), $arg);
                } else {
                    return $data;
                    $s[] = "update " . DB . ".unikdata set data=\"{$data}\",score=\"{$score[$u]}\" where url=\"{$u}\"";
                    $x = sql5(end($s), $arg);
                }
                #Non car le score est unique !
                #if($db=='titre')DbM('unikdata',surl."<li>db?$db / ".hu3."->$data<li>noexist?$noexist<li>sql:".pre($s)."<li>score:".pre($score)."<li>datas:".pre($datas)."<li>res:".pre($res)."<li>GT:".pre($_ENV['dbt']),1);
                return $data;
            } else {
                #si plusieurs résultats pour mêmes données, il peut exister des similarités ..
                $s[] = "select sql_cache url,data,score from " . DB . ".unikdata where data rlike(\"{$data}*\") and url rlike'{$db}:" . H . ".*' and url<>\"{$u}\" order by score asc";
                $y = sql5(['cache' => 1, 'sql' => end($s)], $arg);
                #les autres résultats proches, on s'en tape !
                if ($y) {
                    foreach ($y as $t) {
                        $score[$t['url']] = $t['score'];
                        $datas[$t['url']] = $t['data'];
                    }
                }
                Asort($score);
                foreach ($score as $url => $v) {
                    #calcul des doublons et population de la matrice des résultats
                    $n = 0;
                    $t2 = $temp = $datas[$url];
                    while (in_Array(accents(strtolower($t2)), $prev) && $n < 20) {
                        #déjà un doublon de score inférieur - 20 opérations maximum
                        if ($trig != 'chiffres' && Preg_match_all("~([0-9]+)~", u, $m)) {
                            $n++;
                            $i = implode('', $m[1]);
                            $t2 = $temp . " {$i}";
                            $trig = 'chiffres';
                            continue;
                        }
                        $i++;
                        $t2 = $temp . " {$i}";
                        #sinon on incrémente
                    }
                    $res[$url] = $t2;
                    $prev[] = accents(strtolower($t2));
                    #les scores les plus pourris ramassent la merde
                    if ($url == $u) {
                        $ret = $t2;
                    }
                    #calcule le retour pour l'url courante
                }
                foreach ($res as $url => $v) {
                    if ($noexist && $url == $u) {
                        $s[] = "insert into " . DB . ".unikdata(url,data,score)VALUES(\"{$u}\",\"{$v}\",\"{$score[$u]}\")";
                        sql5(end($s), $arg);
                        continue;
                    }
                    $s[] = "update " . DB . ".unikdata set data=\"{$v}\" where url=\"{$url}\"";
                    sql5(end($s), $arg);
                    #Il est possible que l'url courante n'aie pas d'enregistrements !!!
                }
            }
            if ($db == 'titre') {
                DbM('unikdata', surl . "<li>db?{$db} / " . hu3 . "->{$data}<li>noexist?{$noexist}<li>sql:" . pre($s) . "<li>score:" . pre($score) . "<li>datas:" . pre($datas) . "<li>res:" . pre($res) . "<li>GT:" . pre($_ENV['dbt']), 1);
            }
            return $ret;
        }
    } else {
        $u = u3;
        #local db mode
        if ($db == 'titre' && filemtime(DR . $db . '.db') < $rev) {
            Unlink(DR . $db . '.db');
        }
        #Unlink the whole file
        #if(!strpos(u3,'=')&&$_ENV['soloQ'])$u=Preg_replace("~\?(?!({$_ENV['soloQ']})).*~is",'',u3);#déjà calculée dans autoappend
        #Determine unique Array 1
        while (strlen($data) < $size && $_ENV['ksup']) {
            $next = Array_shift($_ENV['ksup']);
            if (!stripos($data, $next)) {
                $data .= ' ' . $next;
            }
        }
        #push ksup inside
        Rem($key, $u, 'index');
        $x = FGC(DR . $db . '.db');
        #if(J9)die(pre($x));
        if ($x[$key] && now > $_ENV['rev'] && !strpos($_ENV['args'], 'forcetitre')) {
            return $x[$key];
        }
        #Si la clé date d'avant la dernière révision on la retourne
        if ($x[$key] != $data) {
            #différente ou non définie
            $score[$key] = substr_count(u, '/') * 20 + substr_count(u, '?') * 10 + substr_count(u, '&') * 10 + strlen(u);
            #Score de lui même
            if (strpos($_ENV['args'], ',tprio')) {
                $score[$key] = 0;
            }
            if (strpos(surl, '?', 0)) {
                Preg_match_all("~=([^&]+)~is", surl, $t);
                if ($t) {
                    $t = $t[1];
                    $data .= ' ' . implode(' ', $t);
                }
                #si titre similaire, ajouter le query string
            }
            #Cas 1 : il y a un chiffre dans l'url et il est différent des titres des autres, s'il y a déjà un chiffre, on remplace ce dernier par 156
            if (is_Array($x)) {
                null;
            } else {
                dbM("x is not array-no f*****g way:{$x[$key]}§" . pre($x));
            }
            #elseif(!$x){dbM("uniktitre:x is null ?? wtf ??");return;}
            $y = Array_map('strtolower', $x);
            $e = print_r(error_get_last(), 1);
            #if(stripos($e,'array given'))dBM("uniktitre:$db:{$x[$key]}\n<br>err:".pre($e)."\n<Br>y:".pre($y)."<Br>x:".pre($x));
            $found = array_keys($y, strtolower($data));
            #Arrive très souvent..;
            if (count($found) && Preg_match_all("~([0-9]+)~", u, $m)) {
                #avec des chiffres à injecter dans le titre
                $i = implode('', $m[1]);
                $data = $origin . " {$i}";
                $found = array_keys($y, strtolower($data));
                #last verif
                if (count($found) < 1) {
                    FAP("tx=1&file=" . DR . $db . '.db', array($u => $data));
                    db('found:numbers injected in title');
                    return $data;
                }
                #Ok si aucun doublon
            }
            if (count($found)) {
                #des doublons de titres ont été trouvés
                foreach ($found as $k2) {
                    #si les scores sont égaux, on ne pourra la comparer à un autre numéro..
                    $score[$k2] = substr_count($k2, '/') * 20 + substr_count($k2, '?') * 10 + substr_count($k2, '&') * 10 + strlen($k2);
                    if ($k2 == 'index') {
                        $score[$k2] = 0;
                    }
                    #raz page 0
                }
                Asort($score);
                $os = $score;
                foreach ($score as $k => $v) {
                    #Remplacer le score par les valeurs titre
                    while (in_array($data, $score) && $n < 10) {
                        $n++;
                        $i++;
                        $data = $origin . " {$i}";
                    }
                    if ($n > 10) {
                        Dbm("{$db} n:{$n}; data:" . pre($data) . "score:" . pre($score));
                    }
                    $score[$k] = $data;
                    GT('whilescore');
                    #crée autant
                }
                #if($score[0]==$score[self])Best=1;swap titles
                #if(count($score)<2)$score=end($score);
                FAP("tx=2,{$key}&file=" . dr . $db . '.db', $score);
                #DbM('unikdata',surl."<li>data:$data<li>x[key]:".print_r($x[$key],1)."<li>found:".print_r($found,1)."<li>score:".print_r($os,1)."<li>res:".print_r($score,1));
                return $score[$key];
            }
            if ($x[$key] && now > $_ENV['rev'] && !strpos($_ENV['args'], 'forcetitre')) {
                return $x[$key];
            }
        }
        #swap those values pour url prioritaire!!!
        if ($x[$key] && now > $_ENV['rev'] && !strpos($_ENV['args'], 'forcetitre')) {
            return $x[$key];
        }
        #ajouter ici les valeurs des query strings !
        if ($data && $key) {
            FAP("tx=3,{$key}&file=" . dr . $db . '.db', $key, $data);
            gt('l' . __LINE__);
            return $data;
        }
        DB(surl . ':notitle found:' . count($found) . 'x[key]' . $x[$key] . 'data:' . $data . 'origin:' . $origin);
    }
}
Example #2
0
function Starter()
{
    if (us == 'ben' || in_array(h, ['cron', 'cli'])) {
        return;
    }
    e(',starter');
    gt('starter');
    $ip = IP;
    $ref = cleanref(REFE);
    #pour le cron
    if (!STARTER or Preg_Match("~url.data:|image/png;|base64|/!|2001/|admin|intranet|zsimu|(bg|Tag)\\.php|(jsr?|css|jpg|png|ico|gif|eot|htc)\$|\\.(swf|js|css|jpg|png|ico|gif|eot|htc)~is", U . REFE, $t) && !stripos(U . REFE, "submit")) {
        e(',starter:return:quick');
        return array('host' => RH, 'ip' => IP, 'dlp' => SU, 'u' => SU, 'r' => $ref, 'ref' => $ref);
    }
    #ne rien logger pour ces cas là .. retour simple de données
    if (is_file(IPF)) {
        e(',isfile:ipf');
        $t = FGC(IPF);
        G2('ipf', 'Read');
        $t['ip'] = IP;
        if (J9) {
            $t['admlog']++;
        }
        foreach ($t as $k => $v) {
            if (is_numeric($k) or $k == 'SESSION') {
                unset($t[$k]);
            }
        }
        #suppresion des clés numériques issues d'un bug sur FAP
        #if(in_Array(trim($t['mots'],'() '),Array('','not provided'))){unset($t['mots']);$t['motseffacés']=1;}
        if (Preg_Match("~[A-Za-z]~is", $t['host'])) {
            null;
        } elseif (RH != IP && RH) {
            $t['host'] = RH;
            GT('host=rh');
        }
        #correction
        #else{$x['host']=gethostbyaddr($ip);GT('host=reverse');}#correction = take to much time
        if (!is_numeric($t['hits'])) {
            $t['hits'] = 1;
        } else {
            $t['hits'] = $t['hits'] + 1;
        }
        #c,d,b,p
    }
    #New data here :)
    #need keyword on js,php,rewriting,#remplacer par des clés nouvelles
    if (!strpos(REFE, H) && strlen($ref) > 5 && !Preg_Match('~' . HOMESITE . '~', REFE) && $t['fmt']['ref'] + 300 < now) {
        $t['ref'] = $ref;
        $t['fmt']['ref'] = now;
    }
    #lastref
    #if(SU!=refe&&strlen($ref)>5&&$t['fmt']['re']+1500<now){$t['fmt']['re']=now;$t['ref']=$ref;}#rem:internal referers
    if (SU != refe && !Preg("~/([!|\\?]|cont)~", SU) && $t['fmt']['dlp'] + 1500 < now) {
        $t['fmt']['dlp'] = now;
        $t['dlp'] = SU;
    }
    #dlp
    #DLP is set with a new keyword only !!!!
    if ($_COOKIE['__utmz'] && $t['fmt']['kw'] + 3600 < NOW) {
        #Si cookie sur le domaine de moins d'une heure, on récupère ce mot clé
        #Si mot clé défini par ggtracker<>$yt[mots], le refe!=$yt[dlp]
        if (!strpos($_COOKIE['__utmz'], 'not provided')) {
            if (Preg_Match("~utmctr=([^;|]+)~", $_COOKIE['__utmz'], $z)) {
                G2('utmz', '>keyw');
                if (trim($z[1])) {
                    $t['mots'] = $z[1];
                    $t['fmt']['kw'] = now;
                }
            }
        }
        GT('s2');
        #30
    }
    #$f5=TMP."hosts/$ip";#Touch($f5,fmkt($f5)+1);GT('s1');
    GT('s4');
    if (!Preg_Match("~/z/|/\\?~", SU) && Preg_Match('~gclid|aclk~i', Q . REFE)) {
        $t['aclk']++;
        $t['gclid']++;
        Fap(LOGS . 'adwords.db', now, array(IP, u, $ref));
        gt('s5');
    }
    if (Preg_Match("~googleads|doubleclick~is", REFE)) {
        #ref
        Preg_Match_ALL("~&(url)=([^&]+)~is", refe, $y);
        Array_shift($y);
        #|okw|kw[0-9]
        foreach ($y[0] as $n => $k) {
            unset($y[0][$n]);
            $y[0][$k] = urldecode($y[1][$n]);
        }
        Array_pop($y);
        $y = $y[0];
        Fap(LOGS . 'adwords.db', 't' . now, $y);
        $t['ads'] = $y;
    }
    #gclid#si cookie utmz, alors .. fills the keywords again
    #if($t){foreach($t as $k=>$v)$x[$k]=$v;}FPC(ipf,$x,4);GT('fPcipf');
    $x = RH;
    if (is_numeric(RH) || !RH) {
        $x = IP;
    }
    GT('host');
    #$x=gethostbyaddr($ip);if(!$x)
    $t['ip'] = ip;
    $t['host'] = $x;
    $t['hits'] = 1;
    FAP(IPF, $t);
    GT('fPcipf2');
    return $t;
}