MetaGer.php 34.9 KB
Newer Older
1
2
3
4
5
6
<?php
namespace App;

use Illuminate\Http\Request;
use Jenssegers\Agent\Agent;
use App;
7
8
use Storage;
use Log;
9
use Config;
10
use Redis;
11
12
13
14
use App\lib\TextLanguageDetect\TextLanguageDetect;
use App\lib\TextLanguageDetect\LanguageDetect\TextLanguageDetectException;
use Illuminate\Pagination\LengthAwarePaginator;
use Illuminate\Support\Collection;
15
#use \Illuminate\Pagination\Paginator;
16
17
18
19

class MetaGer
{
	# Einstellungen für die Suche
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
	protected $fokus;
	protected $eingabe;
	protected $q;
	protected $category;
	protected $time;
	protected $page;
	protected $lang;
	protected $cache = "";
	protected $site;
	protected $hostBlacklist = [];
	protected $domainBlacklist = [];
	protected $stopWords = [];
    protected $phrases = [];
	protected $engines = [];
	protected $results = [];
    protected $ads = [];
	protected $warnings = [];
    protected $errors = [];
    protected $addedHosts = [];
39
	# Daten über die Abfrage
40
41
42
	protected $ip;
	protected $language;
	protected $agent;
43
	# Konfigurationseinstellungen:
44
45
46
47
48
49
50
51
52
	protected $sumaFile;
	protected $mobile;
	protected $resultCount;
	protected $sprueche;
    protected $domainsBlacklisted = [];
    protected $urlsBlacklisted = [];
    protected $url;
    protected $languageDetect;

53
54
	function __construct()
	{
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
        $this->starttime = microtime(true);   
        if( file_exists(config_path() . "/blacklistDomains.txt") && file_exists(config_path() . "/blacklistUrl.txt") )
        {
            # Blacklists einlesen:
            $tmp = file_get_contents(config_path() . "/blacklistDomains.txt");
            $this->domainsBlacklisted = explode("\n", $tmp);
            $tmp = file_get_contents(config_path() . "/blacklistUrl.txt");
            $this->urlsBlacklisted = explode("\n", $tmp);
        }else
        {
            Log::warning("Achtung: Eine, oder mehrere Blacklist Dateien, konnten nicht geöffnet werden");
        }

        $this->languageDetect = new TextLanguageDetect();
        $this->languageDetect->setNameMode("2");
70
71
	}

72
73
74
75
76
77
78
79
80
81
82
83
84
    public function getHashCode ()
    {
        $string = url()->full();
        return md5($string);
    }

    public function rankAll ()
    {
        foreach( $this->engines as $engine )
        {
            $engine->rank($this);
        }
    }
85

86
87
88
	public function createView()
	{
		$viewResults = [];
89

90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
        # Wir extrahieren alle notwendigen Variablen und geben Sie an unseren View:
        foreach($this->results as $result)
        {
            $viewResults[] = get_object_vars($result);
        }

        # Wir müssen natürlich noch den Log für die durchgeführte Suche schreiben:
        $this->createLogs();

        if( $this->fokus === "bilder" )
        {
            switch ($this->out) 
            {
                case 'results':
                    return view('metager3bilderresults')
                        ->with('results', $viewResults)
                        ->with('eingabe', $this->eingabe)
                        ->with('mobile', $this->mobile)
                        ->with('warnings', $this->warnings)
                        ->with('errors', $this->errors)
                        ->with('metager', $this)
111
                        ->with('browser', (new Agent())->browser());
112
113
114
115
116
117
118
119
                default:
                    return view('metager3bilder')
                        ->with('results', $viewResults)
                        ->with('eingabe', $this->eingabe)
                        ->with('mobile', $this->mobile)
                        ->with('warnings', $this->warnings)
                        ->with('errors', $this->errors)
                        ->with('metager', $this)
120
                        ->with('browser', (new Agent())->browser());
121
122
123
124
125
126
127
128
129
130
131
132
            }
        }

        switch ($this->out) {
            case 'results':
                return view('metager3results')
                    ->with('results', $viewResults)
                    ->with('eingabe', $this->eingabe)
                    ->with('mobile', $this->mobile)
                    ->with('warnings', $this->warnings)
                    ->with('errors', $this->errors)
                    ->with('metager', $this)
133
                    ->with('browser', (new Agent())->browser());
134
135
136
137
138
139
140
141
142
143
                break;
            case 'results-with-style':
                return view('metager3')
                    ->with('results', $viewResults)
                    ->with('eingabe', $this->eingabe)
                    ->with('mobile', $this->mobile)
                    ->with('warnings', $this->warnings)
                    ->with('errors', $this->errors)
                    ->with('metager', $this)
                    ->with('suspendheader', "yes")
144
                    ->with('browser', (new Agent())->browser());
145
146
147
148
149
150
151
152
                break;
            default:
                return view('metager3')
                    ->with('eingabe', $this->eingabe)
                    ->with('mobile', $this->mobile)
                    ->with('warnings', $this->warnings)
                    ->with('errors', $this->errors)
                    ->with('metager', $this)
153
                    ->with('browser', (new Agent())->browser());
154
155
                break;
        }
156
157
	}

158
159
160
161
162
163
164
165
166
167
168
169
170
    private function createLogs()
    {
        $redis = Redis::connection('redisLogs');
        try
        {
            $logEntry = "";
            $logEntry .= "[" . date(DATE_RFC822, mktime(date("H"),date("i"), date("s"), date("m"), date("d"), date("Y"))) . "]";
            $logEntry .= " pid=" . getmypid();
            $logEntry .= " ref=" . $this->request->header('Referer');
            $useragent = $this->request->header('User-Agent');
            $useragent = str_replace("(", " ", $useragent);
            $useragent = str_replace(")", " ", $useragent);
            $useragent = str_replace(" ", "", $useragent);
171
            $logEntry .= " time=" . round((microtime(true)-$this->starttime), 2) . " serv=" . $this->fokus;
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
            $logEntry .= " search=" . $this->eingabe;
            $redis->rpush('logs.search', $logEntry);
        }catch( \Exception $e)
        {
            return;
        }
    }

    public function removeInvalids ()
    {
        $results = [];
        foreach($this->results as $result)
        {
            if($result->isValid($this))
                $results[] = $result;
        }
        #$this->results = $results;
    }
190

191
192
193
194
	public function combineResults ()
	{
		foreach($this->engines as $engine)
		{
195
196
197
198
199
200
201
202
203
            foreach($engine->results as $result)
            {
                if($result->valid)
                    $this->results[] = $result;
            }
            foreach($engine->ads as $ad)
            {
                $this->ads[] = $ad;
            }
204
		}
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
        uasort($this->results, function($a, $b){
            if($a->getRank() == $b->getRank())
                return 0;
            return ($a->getRank() < $b->getRank()) ? 1 : -1;
        });
        # Validate Results
        $newResults = [];
        foreach($this->results as $result)
        {
            if($result->isValid($this))
                $newResults[] = $result;
        }
        $this->results = $newResults;

        $counter = 0;
        $firstRank = 0;
        foreach($this->results as $result)
        {
            if($counter === 0)
                $firstRank = $result->rank;
            $counter++;
            $result->number = $counter;
            $confidence = 0;
            if($firstRank > 0)
                $confidence = $result->rank/$firstRank;
            else
                $confidence = 0;
            if($confidence > 0.65)
                $result->color = "#FF4000";
            elseif($confidence > 0.4)
                $result->color = "#FF0080";
            elseif($confidence > 0.2)
                $result->color = "#C000C0";
            else
                $result->color = "#000000";
        }

        //Get current page form url e.g. &page=6
        $currentPage = LengthAwarePaginator::resolveCurrentPage();
        $offset= $currentPage-1;

        //Create a new Laravel collection from the array data
        $collection = new Collection($this->results);

        //Define how many items we want to be visible in each page
        $perPage = $this->resultCount;

        //Slice the collection to get the items to display in current page
        $currentPageSearchResults = $collection->slice($offset * $perPage, $perPage)->all();

        # Für diese 20 Links folgt nun unsere Boost-Implementation.
        $currentPageSearchResults = $this->parseBoost($currentPageSearchResults);

        # Für diese 20 Links folgt nun unsere Adgoal- Implementation.
        $currentPageSearchResults = $this->parseAdgoal($currentPageSearchResults);

        //Create our paginator and pass it to the view
        $paginatedSearchResults= new LengthAwarePaginator($currentPageSearchResults, count($collection), $perPage);
        $paginatedSearchResults->setPath('/meta/meta.ger3');
        foreach($this->request->all() as $key => $value)
        {
            if( $key === "out" )
                continue;
            $paginatedSearchResults->addQuery($key, $value);
        }

        $this->results = $paginatedSearchResults;

        $this->validated = false;
        if( isset($this->password) )
        {
            # Wir bieten einen bezahlten API-Zugriff an, bei dem dementsprechend die Werbung ausgeblendet wurde:
            # Aktuell ist es nur die Uni-Mainz. Deshalb überprüfen wir auch nur diese.
            $password = getenv('mainz');
            $eingabe = $this->eingabe;
            $password = md5($eingabe . $password);
            if( $this->password === $password )
            {
                $this->ads = [];
                $this->validated = true;
            }
        }
287
288
	}

289
290
    public function parseBoost($results)
    {
Phil Höfer's avatar
Phil Höfer committed
291
	foreach($results as $result)
292
        {
Phil Höfer's avatar
Phil Höfer committed
293
294
295
		if(preg_match('/^(http[s]?\:\/\/)?(www.)?amazon\.de/',$result->anzeigeLink))
		{
			if(preg_match('/\?/',$result->anzeigeLink))
296
	                {
Phil Höfer's avatar
Phil Höfer committed
297
				$result->link .= '&tag=boostmg01-21';
298
        	        } else
Phil Höfer's avatar
Phil Höfer committed
299
300
301
302
303
304
305
306
			{
				$result->link .= '?tag=boostmg01-21';
			}
			$result->partnershop = true;

		}
	}	
	return $results;
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
    }
    public function parseAdgoal($results)
    {
        $publicKey = getenv('adgoal_public');
        $privateKey = getenv('adgoal_private');
        if( $publicKey === FALSE )
        {
            return $results;
        }
        $tldList = "";
        try{
            foreach($results as $result)
            {
                $link = $result->anzeigeLink;
                if(strpos($link, "http") !== 0)
                {
                    $link = "http://" . $link;
                }
                $tldList .= parse_url($link, PHP_URL_HOST) . ",";
                $result->tld = parse_url($link, PHP_URL_HOST);
            }
            $tldList = rtrim($tldList, ",");

            # Hashwert
            $hash = md5("meta" . $publicKey . $tldList . "GER");

            # Query 
            $query = urlencode($this->q);

            $link = "https://api.smartredirect.de/api_v2/CheckForAffiliateUniversalsearchMetager.php?p=" . $publicKey . "&k=" . $hash . "&tld=" . $tldList . "&q=" . $query; 
            $answer = json_decode(file_get_contents($link));


            # Nun müssen wir nur noch die Links für die Advertiser ändern:
            foreach($answer as $el)
            {
                $hoster = $el[0];
                $hash = $el[1];

                foreach($results as $result)
                {
                    if( $hoster === $result->tld )
                    {
                        # Hier ist ein Advertiser:
                        # Das Logo hinzufügen:
352
353
354
355
                        if( $result->image !== "" )
                            $result->logo = "https://img.smartredirect.de/logos_v2/60x30/" . $hash . ".gif";
                        else
                            $result->image = "https://img.smartredirect.de/logos_v2/120x60/" . $hash . ".gif";
356
357
358
359
360
                        # Den Link hinzufügen:
                        $publicKey = $publicKey;
                        $targetUrl = $result->anzeigeLink;
                        if(strpos($targetUrl, "http") !== 0)
                            $targetUrl = "http://" . $targetUrl;
361
362
                        $gateHash = md5($targetUrl . $privateKey);
                        $newLink = "https://api.smartredirect.de/api_v2/ClickGate.php?p=" . $publicKey . "&k=" . $gateHash . "&url=" . urlencode($targetUrl) . "&q=" . $query;
363
364
365
366
367
368
369
370
371
372
373
374
                        $result->link = $newLink;
                        $result->partnershop = true;
                    }
                }
            }
        }catch(\ErrorException $e)
        {
            return $results;
        }

        return $results;
    }
375

376
377
	public function createSearchEngines (Request $request)
	{
378

379
380
        if( !$request->has("eingabe") )
            return;
381

382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
		# Überprüfe, welche Sumas eingeschaltet sind
        $xml = simplexml_load_file($this->sumaFile);
        $enabledSearchengines = [];
        $overtureEnabled = FALSE;
        $countSumas = 0;
        $sumas = $xml->xpath("suma");
        if($this->fokus === "angepasst")
        {
            foreach($sumas as $suma)
            {
                if($request->has($suma["name"]) 
                	|| ( $this->fokus !== "bilder" 
                		&& ($suma["name"]->__toString() === "qualigo" 
                			|| $suma["name"]->__toString() === "similar_product_ads" 
                			|| ( !$overtureEnabled && $suma["name"]->__toString() === "overtureAds" )
                			)
                		)
                	){

                	if(!(isset($suma['disabled']) && $suma['disabled']->__toString() === "1"))
                    {
                        if($suma["name"]->__toString() === "overture" || $suma["name"]->__toString() === "overtureAds")
                        {
                            $overtureEnabled = TRUE;
                        }
                        if( $suma["name"]->__toString() !== "qualigo" && $suma["name"]->__toString() !== "similar_product_ads" && $suma["name"]->__toString() !== "overtureAds" )
                            $countSumas += 1;
                        $enabledSearchengines[] = $suma;
                    }
                }
            }
        }else{
            foreach($sumas as $suma){
                $types = explode(",",$suma["type"]);
                if(in_array($this->fokus, $types) 
                	|| ( $this->fokus !== "bilder" 
                		&& ($suma["name"]->__toString() === "qualigo" 
                			|| $suma["name"]->__toString() === "similar_product_ads" 
                			|| ( !$overtureEnabled && $suma["name"]->__toString() === "overtureAds" )
                			)
                		)
                	){
                    if(!(isset($suma['disabled']) && $suma['disabled']->__toString() === "1"))
                    {
                        if($suma["name"]->__toString() === "overture" || $suma["name"]->__toString() === "overtureAds")
                        {
                            $overtureEnabled = TRUE;
                        }
                        if( $suma["name"]->__toString() !== "qualigo" && $suma["name"]->__toString() !== "similar_product_ads" && $suma["name"]->__toString() !== "overtureAds" )
                            $countSumas += 1;
                        $enabledSearchengines[] = $suma;
                    }
                }
            }
        }

        # Sonderregelung für alle Suchmaschinen, die zu den Minisuchern gehören. Diese können alle gemeinsam über einen Link abgefragt werden
        $subcollections = [];
        $tmp = [];
        foreach($enabledSearchengines as $engine )
        {
            if( isset($engine['minismCollection']) )
                $subcollections[] = $engine['minismCollection']->__toString();
            else
                $tmp[] = $engine;
        }
        $enabledSearchengines = $tmp;
        if( sizeof($subcollections) > 0)
        {
            $count = sizeof($subcollections) * 10;
            $minisucherEngine = $xml->xpath('suma[@name="minism"]')[0];  
            $subcollections = urlencode("(" . implode(" OR ", $subcollections) . ")");
            $minisucherEngine["formData"] = str_replace("<<SUBCOLLECTIONS>>", $subcollections, $minisucherEngine["formData"]);
            $minisucherEngine["formData"] = str_replace("<<COUNT>>", $count, $minisucherEngine["formData"]);
            $enabledSearchengines[] = $minisucherEngine;
        }

        #die(var_dump($enabledSearchengines));

        if( $countSumas <= 0 )
        {
            $this->errors[] = "Achtung: Sie haben in ihren Einstellungen keine Suchmaschine ausgewählt.";
        }
465
		$engines = [];
466

467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
        $siteSearchFailed = false;
        if( strlen($this->site) > 0 )
        {
            # Wenn eine Sitesearch durchgeführt werden soll, überprüfen wir ob eine der Suchmaschinen überhaupt eine Sitesearch unterstützt:
            $enginesWithSite = 0;
            foreach($enabledSearchengines as $engine)
            {
                if( isset($engine['hasSiteSearch']) && $engine['hasSiteSearch']->__toString() === "1" )
                {
                    $enginesWithSite++;
                }
            }
            if( $enginesWithSite === 0 )
            {
                $this->errors[] = "Sie wollten eine Sitesearch auf " . $this->site . " durchführen. Leider unterstützen die eingestellten Suchmaschinen diese nicht. Sie können <a href=\"" . $this->generateSearchLink("web", false) . "\">hier</a> die Sitesearch im Web-Fokus durchführen. Es werden ihnen Ergebnisse ohne Sitesearch angezeigt.";
                $siteSearchFailed = true;
            }else
            {
                $this->warnings[] = "Sie führen eine Sitesearch durch. Es werden nur Ergebnisse von der Seite: <a href=\"http://" . $this->site . "\" target=\"_blank\">\"" . $this->site . "\"</a> angezeigt.";
            }

        }

        $typeslist = [];
        $counter = 0;
492

493
		foreach($enabledSearchengines as $engine){
494

495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
            if( !$siteSearchFailed && strlen($this->site) > 0 && ( !isset($engine['hasSiteSearch']) || $engine['hasSiteSearch']->__toString() === "0")  )
            {
                
                continue;
            }
            # Wenn diese Suchmaschine gar nicht eingeschaltet sein soll
            $path = "App\Models\parserSkripte\\" . ucfirst($engine["package"]->__toString());

            if( !file_exists(app_path() . "/Models/parserSkripte/" . ucfirst($engine["package"]->__toString()) . ".php"))
            {
                Log::error("Konnte " . $engine["name"] . " nicht abfragen, da kein Parser existiert");
                continue;
            }

            $time = microtime();

            try
            {
                $tmp = new $path($engine, $this);
            } catch( \ErrorException $e)
            {
                Log::error("Konnte " . $engine["name"] . " nicht abfragen." . var_dump($e));
                continue;
            }

            if($tmp->enabled && isset($this->debug))
            {
                $this->warnings[] = $tmp->service . "   Connection_Time: " . $tmp->connection_time . "    Write_Time: " . $tmp->write_time . " Insgesamt:" . ((microtime()-$time)/1000);
            }

            if($tmp->isEnabled())
            {
                $engines[] = $tmp;
                $this->sockets[$tmp->name] = $tmp->fp;
            }
530

531
		}
532

533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
        # Jetzt werden noch alle Kategorien der Settings durchgegangen und die jeweils enthaltenen namen der Suchmaschinen gespeichert.
        $foki = [];
        foreach($sumas as $suma)
        {
            if( (!isset($suma['disabled']) || $suma['disabled'] === "") && ( !isset($suma['userSelectable']) || $suma['userSelectable']->__toString() === "1") )
            {
                if( isset($suma['type']) )
                {
                    $f = explode(",", $suma['type']->__toString());
                    foreach($f as $tmp)
                    {
                        $name = $suma['name']->__toString();
                        $foki[$tmp][$suma['name']->__toString()] = $name;
                    }
                }else
                {
                    $name = $suma['name']->__toString();
                    $foki["andere"][$suma['name']->__toString()] = $name;
                }
            }
        }

        # Es werden auch die Namen der aktuell aktiven Suchmaschinen abgespeichert.
        $realEngNames = [];
        foreach($enabledSearchengines as $realEng) {
            $nam = $realEng["name"]->__toString();
            if($nam !== "qualigo" && $nam !== "overtureAds") {
                $realEngNames[] = $nam;
            }
        }
        # Anschließend werden diese beiden Listen verglichen (jeweils eine der Fokuslisten für jeden Fokus), um herauszufinden ob sie vielleicht identisch sind. Ist dies der Fall, so hat der Nutzer anscheinend Suchmaschinen eines kompletten Fokus eingestellt. Der Fokus wird dementsprechend angepasst.
        foreach($foki as $fok => $engs) {
            $isFokus = true;
            $fokiEngNames = [];
            foreach($engs as $eng) {
                $fokiEngNames[] = $eng;
            }
            foreach($fokiEngNames as $fen) {
                if(!in_array($fen, $realEngNames)) {
                    $isFokus = false;
                }
            }
            foreach($realEngNames as $ren) {
                if(!in_array($ren, $fokiEngNames)) {
                    $isFokus = false;
                }
            }
            if($isFokus) {
                $this->fokus = $fok;
            }
        }

        # Nun passiert ein elementarer Schritt.
        # Wir warten auf die Antwort der Suchmaschinen, da wir vorher nicht weiter machen können.
        # aber natürlich nicht ewig.
        # Die Verbindung steht zu diesem Zeitpunkt und auch unsere Request wurde schon gesendet.
        # Wir geben der Suchmaschine nun bis zu 500ms Zeit zu antworten.
590
591
592
593
594
595
596
597
598
599
600
601
602
603

        # Wir zählen die Suchmaschinen, die durch den Cache beantwortet wurden:
        $enginesToLoad = 0;
        $canBreak = false;
        foreach($engines as $engine)
        {
            if( $engine->cached )
            {
                $enginesToLoad--;
                if( $overtureEnabled && ( $engine->name === "overture" || $engine->name === "overtureAds" ) )
                    $canBreak = true;
            }
        }
        $enginesToLoad += count($engines);
604
605
606
607
608
609
        $loadedEngines = 0;
        $timeStart = microtime(true);
        while( true )
        {
            $time = (microtime(true) - $timeStart) * 1000;
            $loadedEngines = intval(Redis::hlen('search.' . $this->getHashCode()));
610
611
            if( $overtureEnabled && (Redis::hexists('search.' . $this->getHashCode(), 'overture') || Redis::hexists('search.' . $this->getHashCode(), 'overtureAds')))
                $canBreak = true;
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628

            # Abbruchbedingung
            if($time < 500)
            {
                if(($enginesToLoad === 0 || $loadedEngines >= $enginesToLoad) && $canBreak)
                    break;
            }elseif( $time >= 500 && $time < $this->time)
            {
                if( ($enginesToLoad === 0 || ($loadedEngines / ($enginesToLoad * 1.0)) >= 0.8) && $canBreak )
                    break;
            }else
            {
                break;
            }
            usleep(50000);
        }

629
        #exit;
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
        foreach($engines as $engine)
        {
            if(!$engine->loaded)
            {
                try{
                    $engine->retrieveResults();
                } catch(\ErrorException $e)
                {
                    Log::error($e);
                    
                }
            }
        }
        
        # und verwerfen den Rest:
        foreach( $engines as $engine )
        {
            if( !$engine->loaded )
                $engine->shutdown();
        }

        $this->engines = $engines;
652
653
654
655
656
657
658
659
660
661
662
663
664
665
	}

	public function parseFormData (Request $request)
	{
		if($request->input('encoding', '') !== "utf8")
		{
			# In früheren Versionen, als es den Encoding Parameter noch nicht gab, wurden die Daten in ISO-8859-1 übertragen
			$input = $request->all();
			foreach($input as $key => $value)
			{
				$input[$key] = mb_convert_encoding("$value", "UTF-8", "ISO-8859-1");
			}
			$request->replace($input);
		}
666
        $this->url = $request->url();
667
		# Zunächst überprüfen wir die eingegebenen Einstellungen:
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
        # FOKUS
        $this->fokus = trans('fokiNames.'
        	. $request->input('focus', 'web'));
        if(strpos($this->fokus,"."))
        {
            $this->fokus = trans('fokiNames.web');
        }

        # SUMA-FILE
        if(App::isLocale("en")){
            $this->sumaFile = config_path() . "/sumas.xml";
        }else{
            $this->sumaFile = config_path() . "/sumas.xml";
        }
        if(!file_exists($this->sumaFile))
        {
            die("Suma-File konnte nicht gefunden werden");
        }

        # Sucheingabe:
        $this->eingabe = trim($request->input('eingabe', ''));
        if(strlen($this->eingabe) === 0)
        {
            $this->warnings[] = 'Achtung: Sie haben keinen Suchbegriff eingegeben. Sie können ihre Suchbegriffe oben eingeben und es erneut versuchen.';
        }
        $this->q = $this->eingabe;

        # IP:
        $this->ip = $request->ip();

        # Language:
        if( isset($_SERVER['HTTP_LANGUAGE']) )
        {
            $this->language = $_SERVER['HTTP_LANGUAGE'];
        }else
        {
            $this->language = "";
        }
        # Category
        $this->category = $request->input('category', '');
        # Request Times:
        $this->time = $request->input('time', 1000);
       
        # Page
        $this->page = $request->input('page', 1);
        # Lang
        $this->lang = $request->input('lang', 'all');
        if ( $this->lang !== "de" && $this->lang !== "en" && $this->lang !== "all" )
        {
        	$this->lang = "all";
        }
        $this->agent = new Agent();
        $this->mobile = $this->agent->isMobile();

        #Sprüche
        $this->sprueche = $request->input('sprueche', 'off');
        if($this->sprueche === "off" )
            $this->sprueche = true;
        else
            $this->sprueche = false;
        # Ergebnisse pro Seite:
        $this->resultCount = $request->input('resultCount', '20');

        # Manchmal müssen wir Parameter anpassen um den Sucheinstellungen gerecht zu werden:
        if( $request->has('dart') )
        {
        	$this->time = 10000;
        	$this->warnings[] = "Hinweis: Sie haben Dart-Europe aktiviert. Die Suche kann deshalb länger dauern und die maximale Suchzeit wurde auf 10 Sekunden hochgesetzt.";
        }
        if( $this->time <= 500 || $this->time > 20000 )
        {
        	$this->time = 1000;
        }
        if( $request->has('minism') && ( $request->has('fportal') || $request->has('harvest') ) )
        {
        	$input = $request->all();
        	$newInput = [];
        	foreach($input as $key => $value)
        	{
        		if( $key !== "fportal" && $key !== "harvest" )
        		{
        			$newInput[$key] = $value;
        		}
        	}
        	$request->replace($newInput);
        }
        if( App::isLocale("en") )
        {
        	$this->sprueche = "off";
        }
        if($this->resultCount <= 0 || $this->resultCount > 200 )
        {
        	$this->resultCount = 1000;
        }
        if( $request->has('onenewspageAll') || $request->has('onenewspageGermanyAll') )
        {
        	$this->time = 5000;
        	$this->cache = "cache";
        }
        if( $request->has('tab'))
        {
            if($request->input('tab') === "off")
            {
                $this->tab = "_blank";
            }else
            {
                $this->tab = "_self";
            }
        }else
        {
            $this->tab = "_blank";
        }
        if( $request->has('password') )
            $this->password = $request->input('password');
        if( $request->has('quicktips') )
            $this->quicktips = false;
        else
            $this->quicktips = true;

        $this->out = $request->input('out', "html");
        if($this->out !== "html" && $this->out !== "json" && $this->out !== "results" && $this->out !== "results-with-style")
            $this->out = "html";
        $this->request = $request;
791
792
793
794
795
796
797
798
799
800
	}

	public function checkSpecialSearches (Request $request)
	{
		# Site Search:
		if(preg_match("/(.*)\bsite:(\S+)(.*)/si", $this->q, $match))
		{
			$this->site = $match[2];
			$this->q = $match[1] . $match[3];
		}
801
802
803
804
805
806
        if( $request->has('site') )
        {
            $this->site = $request->input('site');
        }
		# Wenn die Suchanfrage um das Schlüsselwort "-host:*" ergänzt ist, sollen bestimmte Hosts nicht eingeblendet werden
		# Wir prüfen, ob das hier der Fall ist:
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
		while(preg_match("/(.*)(^|\s)-host:(\S+)(.*)/si", $this->q, $match))
		{
			$this->hostBlacklist[] = $match[3];
			$this->q = $match[1] . $match[4];
		}
		if( sizeof($this->hostBlacklist) > 0 )
		{
			$hostString = "";
			foreach($this->hostBlacklist as $host)
			{
				$hostString .= $host . ", ";
			}
			$hostString = rtrim($hostString, ", ");
			$this->warnings[] = "Ergebnisse von folgenden Hosts werden nicht angezeigt: \"" . $hostString . "\"";
		}
822
823
		# Wenn die Suchanfrage um das Schlüsselwort "-domain:*" ergänzt ist, sollen bestimmte Domains nicht eingeblendet werden
		# Wir prüfen, ob das hier der Fall ist:
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
		while(preg_match("/(.*)(^|\s)-domain:(\S+)(.*)/si", $this->q, $match))
		{
			$this->domainBlacklist[] = $match[3];
			$this->q = $match[1] . $match[4];
		}
		if( sizeof($this->domainBlacklist) > 0 )
		{
			$domainString = "";
			foreach($this->domainBlacklist as $domain)
			{
				$domainString .= $domain . ", ";
			}
			$domainString = rtrim($domainString, ", ");
			$this->warnings[] = "Ergebnisse von folgenden Domains werden nicht angezeigt: \"" . $domainString . "\"";
		}
		
		# Alle mit "-" gepräfixten Worte sollen aus der Suche ausgeschlossen werden.
841
		# Wir prüfen, ob das hier der Fall ist:
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
		while(preg_match("/(.*)(^|\s)-(\S+)(.*)/si", $this->q, $match))
		{
			$this->stopWords[] = $match[3];
			$this->q = $match[1] . $match[4];
		}
		if( sizeof($this->stopWords) > 0 )
		{
			$stopwordsString = "";
			foreach($this->stopWords as $stopword)
			{
				$stopwordsString .= $stopword . ", ";
			}
			$stopwordsString = rtrim($stopwordsString, ", ");
			$this->warnings[] = "Sie machen eine Ausschlusssuche. Ergebnisse mit folgenden Wörtern werden nicht angezeigt: \"" . $stopwordsString . "\"";
		}

858
859
860
		# Meldung über eine Phrasensuche
        $p = "";
        $tmp = $this->q;
861
		while(preg_match("/(.*)\"(.+)\"(.*)/si", $tmp, $match)){
862
863
            $tmp = $match[1] . $match[3];
            $this->phrases[] = strtolower($match[2]);
864
		}
865
866
867
868
869
870
871
        foreach($this->phrases as $phrase)
        {
            $p .= "\"$phrase\", ";
        }
        $p = rtrim($p, ", ");
        if(sizeof($this->phrases) > 0)
            $this->warnings[] = "Sie führen eine Phrasensuche durch: $p";
872
873
	}

874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
1001
1002
1003
1004
1005
1006
1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
1019
1020
1021
1022
1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039
1040
1041
1042
1043
1044
1045
1046
1047
1048
1049
1050
1051
1052
1053
1054
1055
1056
1057
1058
1059
1060
1061
1062
1063
1064
1065
1066
1067
1068
1069
1070
1071
1072
1073
1074
1075
1076
1077
1078
    public function getFokus ()
    {
        return $this->fokus;
    }

    public function getIp ()
    {
        return $this->ip;
    }

    public function getEingabe ()
    {
        return $this->eingabe;
    }

    public function getQ ()
    {
            return $this->q;
    }

    public function getUrl ()
    {
        return $this->url;
    }
    public function getTime ()
    {
        return $this->time;
    }

    public function getLanguage ()
    {
        return $this->language;
    }

    public function getLang ()
    {
        return $this->lang;
    }

    public function getSprueche ()
    {
        return $this->sprueche;
    }

    public function getCategory ()
    {
        return $this->category;
    }

    public function getPhrases ()
    {
        return $this->phrases;
    }

    public function getSumaFile ()
    {
        return $this->sumaFile;
    }

    public function getUserHostBlacklist ()
    {
        return $this->hostBlacklist;
    }

    public function getUserDomainBlacklist ()
    {
        return $this->domainBlacklist;
    }

    public function getDomainBlacklist ()
    {
        return $this->domainsBlacklisted;
    }

    public function getUrlBlacklist ()
    {
        return $this->urlsBlacklisted;
    }
    public function getLanguageDetect ()
    {
        return $this->languageDetect;
    }
    public function getStopWords ()
    {
        return $this->stopWords;
    }
    public function getHostCount($host)
    {
        if(isset($this->addedHosts[$host]))
        {
            return $this->addedHosts[$host];
        }else
        {
            return 0;
        }
    }
    public function addHostCount($host)
    {
        $hash = md5($host);
        if(isset($this->addedHosts[$hash]))
        {
            $this->addedHosts[$hash] += 1;
        }else
        {
            $this->addedHosts[$hash] = 1;
        }
    }
    public function getSite()
    {
        return $this->site;
    }
    public function addLink($link)
    {
        if(strpos($link, "http://") === 0)
            $link = substr($link, 7);
        if(strpos($link, "https://") === 0)
            $link = substr($link, 8);
        if(strpos($link, "www.") === 0)
            $link = substr($link, 4);
        $link = trim($link, "/");
        $hash = md5($link);
        if(isset($this->addedLinks[$hash]))
        {
            return false;
        }else
        {
            $this->addedLinks[$hash] = 1;

            return true;
        }
    }

    public function generateSearchLink($fokus, $results = true)
    {
        $requestData = $this->request->except('page');
        $requestData['focus'] = $fokus;
        if($results)
            $requestData['out'] = "results";
        else
            $requestData['out'] = "";
        $link = action('MetaGerSearch@search', $requestData);
        return $link;
    }

    public function generateQuicktipLink()
    {
        $link = action('MetaGerSearch@quicktips');

        return $link;
    }

    public function generateSiteSearchLink($host)
    {
        $host = urlencode($host);
        $requestData = $this->request->except(['page','out']);
        $requestData['eingabe'] .= " site:$host";
        $requestData['focus'] = "web";
        $link = action('MetaGerSearch@search', $requestData);
        return $link;
    }

    public function generateRemovedHostLink ($host)
    {
        $host = urlencode($host);
        $requestData = $this->request->except(['page','out']);
        $requestData['eingabe'] .= " -host:$host";
        $link = action('MetaGerSearch@search', $requestData);
        return $link;
    }

    public function generateRemovedDomainLink ($domain)
    {
        $domain = urlencode($domain);
        $requestData = $this->request->except(['page','out']);
        $requestData['eingabe'] .= " -domain:$domain";
        $link = action('MetaGerSearch@search', $requestData);
        return $link;
    }

    public function getTab ()
    {
        return $this->tab;
    }
    public function getResults ()
    {
        return $this->results;
    }
    public function popAd()
    {
        if(count($this->ads) > 0)
            return get_object_vars(array_shift($this->ads));
        else
            return null;
    }
    public function getImageProxyLink($link)
    {
        $requestData = [];
        $requestData["url"] = $link;
        $link = action('Pictureproxy@get', $requestData);
        return $link;
    }
    public function showQuicktips ()
    {
        return $this->quicktips;
    }
Phil Höfer's avatar
Phil Höfer committed
1079
}