git.roojs.org Git - Pman.Cms/blob - DataObjects/CmsRefer.php

   1 <?php
   2 /**
   3  * Table Definition for referrers
   4  */
   5 class_exists('DB_DataObject') ? '' : require_once 'DB/DataObject.php';
   6
   7 class Pman_Cms_DataObjects_CmsRefer extends DB_DataObject
   8 {
   9
  10     ###START_AUTOCODE
  11     /* the code below is auto generated do not remove the above tag */
  12
  13     var $__table = 'CmsRefer';                       // table name
  14     var $blog_id;                        // int(11)  not_null primary_key auto_increment
  15     var $day;                             // date(10)  not_null primary_key
  16     var $seen;                           // int(11)  not_null
  17     var $scheme;                          // string(5)
  18     var $host;                            // string(128)  not_null primary_key
  19     var $port;                            // string(5)
  20     var $path;                            // string(255)
  21     var $checked;
  22     var $checked_success;
  23     var $query;                           // string(255)
  24     var $title;
  25
  26     /* the code above is auto generated do not remove the tag below */
  27     ###END_AUTOCODE
  28
  29     function checkValid() {
  30         // does it need checking?
  31         // links are valid for 30 days..
  32          // echo "CHECK? <PRE>" . print_r($this, true) . '</PRE>';
  33
  34         if (($this->checked_success > 0) && strtotime($this->checked) > (time() - (30 * 24 * 60 * 60))) {
  35
  36             return $this->checked_success;
  37         }
  38         // if we failed to check, wait another day..
  39         if ($this->checked_success < 0 && $this->checked && strtotime($this->checked) > (time() - (24 * 60 * 60))) {
  40             //echo "holding off checking";exit;
  41             return $this->checked_success;
  42         }
  43         // we delete ourselves if 10 failures..
  44         if ($this->checked_success < -10) {
  45             $this->delete();
  46             //echo "DELETEING?"; exit;
  47             return 0;
  48         }
  49        //echo "starting checking";exit;
  50         $original_success = $this->checked_success;
  51         // DB_DataObject::debugLevel(1);
  52         // update the fact we have checked it before we potentially hang..
  53         require_once 'DB/DataObject/Cast.php';
  54
  55         if (!$this->clean()) {
  56             $this->delete();
  57             return 0;
  58         }
  59
  60
  61         if (strlen($this->query)) {
  62
  63             parse_str ( $this->query,$q );
  64             if (isset($q['q'])) {
  65
  66
  67
  68
  69                 $this->host == 'google.com';
  70                 $this->query = 'q='.  $q['q']; // not if urlencoding is needed.
  71                 $this->title =  urldecode($q['q']);
  72                 $this->checked_success = 1;
  73                 $this->checked = DB_DataObject_Cast::sql('NOW()');
  74                 $up = clone($this);
  75                 $up->update();
  76                 return 2;
  77             }
  78         }
  79
  80         $this->checked_success--;
  81         $this->checked = DB_DataObject_Cast::sql('NOW()');
  82         $up = clone($this);
  83         $up->update();
  84
  85
  86         // this could hang!!???
  87         //echo "checking ".$this->url();
  88
  89         $data = @file_get_contents($this->url());
  90         //echo htmlspecialchars($data);
  91         // look for our url in it!
  92         // fixme also needs to look for page urls..
  93         if ((false === strpos($data,"http://www.akbkhome.com/blog.php/View/{$this->blog_id}/")) &&
  94             (false === strpos($data,"http://www.roojs.com/index.php/View/{$this->blog_id}/")) &&
  95             (false === strpos($data,"http://roojs.com/index.php/View/{$this->blog_id}/")) &&
  96            (false === strpos($data,"http://blog.akbkhome.com/archives/{$this->blog_id}_")))
  97
  98         {
  99             //echo 'link not on page..';
 100             return 0;
 101         }
 102         if (preg_match('#<title>([^<]+)</title>#i',$data,$matches)) {
 103             $this->title = $matches[1];
 104         }
 105
 106
 107         $this->checked_success = 2;
 108         $up = clone($this);
 109         $up->update();
 110
 111         if ($original_success < 1) {
 112             //DB_DataObject::debugLevel(1);
 113             $e = DB_DataObject::factory('cms_page');
 114             $e->get($this->blog_id);
 115             $auth = $e->getAuthorObject();
 116             $this->entryObject = $e;
 117            // $auth->sendComment($this,'referer.txt');
 118         }
 119
 120         return 1;
 121
 122     }
 123
 124
 125     function clean()        // return false if invalid... ???/
 126     {
 127         //DB_DataObject::debugLevel(1);
 128
 129         parse_str ( $this->query,$q );
 130
 131         if (isset($q['PHPSESSID'])) {
 132             unset($q['PHPSESSID']);
 133         }
 134
 135         $ip = gethostbyname($this->host);
 136         switch(true) {
 137
 138
 139             // BIG IGNORE LIST
 140             // cause the blog is called smooking.... !
 141             case preg_match('#/smoking/#',$this->query):
 142             // live journal friends links - rather messy aggreted views
 143             case preg_match('#/friends/#',$this->path):
 144             // flexy teens crap!
 145             case (preg_match('#teen#i',$this->query) &&  preg_match('#flexy#i',$this->query)):
 146             case (preg_match('#teen#i',$this->title) &&  preg_match('#flexy#i',$this->title)):
 147
 148             case preg_match('#teens\.search\.aol\.com$#',$this->host);
 149             //this site is not about smoking..
 150             case preg_match('#smoking#i',$this->query):
 151
 152             // stargeek is a questionable search engine op site.
 153             case preg_match('#stargeek\.com$#',$this->host):
 154
 155             // ignore google tranlations.
 156             case preg_match('#translate\.google\.#',$this->host):
 157
 158             case preg_match('#mywebsearch#',$this->host):
 159             case preg_match('#ie-seven\.com$#',$this->host):
 160             case preg_match('#iest\.ru$#',$this->host):
 161             case preg_match('#ibissecurity\.com$#',$this->host):
 162             case preg_match('#iwebtool\.com$#',$this->host):
 163             case preg_match('#qoils\.com$#',$this->host):
 164             case preg_match('#xml4\.com$#',$this->host):
 165             case preg_match('#firmantivirus\.com$#',$this->host):
 166             case preg_match('#hostingvoice.com$#',$this->host):
 167             case preg_match('#framecart.com$#',$this->host):
 168             case preg_match('#o-list\.com$#',$this->host):
 169             case preg_match('#stopsmoking911\.com$#',$this->host):
 170             case preg_match('#hv-deals\.info$#',$this->host):
 171            case preg_match('#freeforums\.org$#',$this->host):
 172             // nasty link spammer!
 173             case preg_match('#insurance#',$this->host):
 174             //case preg_match('#debtconsolidation#',$this->host):
 175             //case preg_match('#distancelearning#',$this->host):
 176             // workfromhomefre group of IP's
 177             case preg_match('#^207\.36\.200\..*$#',$ip):
 178             //case preg_match('#uxsk.com$#',$this->host): // ns@thatsdata.com
 179             //case preg_match('#vqsn.com$#',$this->host): // ns@thatsdata.com
 180             //case preg_match('#zfjo.com$#',$this->host): //  ns@targe5t.com
 181             //case preg_match('#ybvm.com$#',$this->host): //  ns@targe5t.com
 182             // the thatsdata group of IP addresses..
 183             case preg_match('#^64\.70\.250\..*$#',$ip):
 184             // the targe5t.com group of IP's
 185             case preg_match('#^69\.64\.76\..*$#',$ip); // web sites are on this.
 186             case preg_match('#^69\.64\.78\..*$#',$ip); // ns is on this.
 187             // the themasterns.com group of IPs
 188             case preg_match('#^70\.85\.234\..*$#',$ip); // websites are on this.
 189             case preg_match('#^70\.87\.36\..*$#',$ip); // ns is on this.
 190             case preg_match('#^70\.85\.235\..*$#',$ip); // websites are on this.
 191
 192             // questionable..
 193
 194             return false;
 195
 196             // lexpov a pornsite that spamed us and got caught up in google. etc.
 197             case preg_match('#/lexpov/#',$this->query):
 198             case preg_match('#/lexpov/#',$this->host):
 199             case preg_match('#/optonline/#',$this->host): // stupid search engine that keeps refering..
 200             case preg_match('#/lexpov/#',$this->path):
 201             case preg_match('#/lexpov/#',$this->title):
 202                 return  false;
 203
 204             // replace msn search with google.
 205             case isset($q['as_q']):
 206                 $q['q'] = $q['as_q'];
 207
 208             case isset($q['p']): // yahoo + fallthrough
 209                 if (isset($q['p'])) {
 210                     $q['q'] = $q['p'];
 211                 }
 212
 213
 214             case isset($q['q']):
 215
 216                 // gaming serach engine with direct queries..
 217                 if (preg_match('/roojs/',$q['q'])) {
 218                     return false;
 219                 }
 220
 221
 222                 $this->host = 'google.com';
 223                 $this->query = 'q='.$q['q'];
 224                 $this->path = '/search';
 225
 226                 return true;
 227
 228             // PHP's standard  session id
 229             /*
 230             case (false !== strpos($this->query,'PHPSESSID=')):
 231                 $query = explode('&',$this->query);
 232                 $this->query = '';
 233                 foreach($query as $k=>$v) {
 234
 235                     if (false !== strpos($k,'PHPSESSID=')) {
 236                         continue; //
 237                     }
 238
 239                     $this->query .= strlen($this->query) ? '&' : '';
 240                     $this->query .= $k;
 241                     break;
 242                 }
 243                 break true;
 244             */
 245             case ($this->host == 'www.linux.org.ru'):
 246                 $this->path = preg_replace('#/profile/[a-z0-9]+/#i', '/', $this->path);
 247                 return true;
 248
 249
 250             // php manuals
 251             case preg_match('#\.php\.net$#',$this->host):
 252                 $this->host = 'php.net';
 253                 return true;
 254
 255             // root these ones.. = common referrers
 256             case preg_match('#weblabor\.hu$#',$this->host):
 257                 $this->query = '';
 258                 $this->path = '';
 259                 $this->host = 'weblabor.hu';
 260                 return true;
 261
 262             case preg_match('#planet-php\.net$$#',$this->host):
 263                 $this->query = '';
 264                 $this->path = '';
 265                 $this->host = 'www.planet-php.net';
 266                 return true;
 267
 268             case preg_match('#artima\.com$#',$this->host):
 269                 $this->query = '';
 270                 $this->path = '';
 271                 $this->host = 'www.artima.com';
 272                 return true;
 273
 274             // general
 275
 276             default:
 277                 // prefix www where relivant..
 278                 if (!preg_match('#^www\.#',$this->host)) {
 279                     $q = DB_DataObject::factory('CmsRefer');
 280                     $q->host = 'www.'.$this->host;
 281                     if ($q->find(true)) {
 282                         $this->host = 'www.' . $this->host;
 283                     }
 284                 } else {
 285                     // remove www. if we already have it..
 286
 287                     $q = DB_DataObject::factory('CmsRefer');
 288                     $q->host = substr($this->host,4);
 289                     if ($q->find(true)) {
 290                         $this->host = substr($this->host,4);
 291                     }
 292                 }
 293
 294
 295                 // praenanz.de has trackbacks- as a add on to the referrers..
 296                 $q->path = preg_replace('#/trackbacks-$#','', $q->path);
 297                 break;
 298
 299         }
 300         return true;
 301
 302     }
 303
 304
 305     function url() {
 306         return "http://{$this->host}{$this->path}" . ($this->query ? "?{$this->query}" : '');
 307     }
 308
 309
 310
 311
 312 }