|
1
|
|
|
<?php |
|
2
|
|
|
|
|
3
|
|
|
declare(strict_types=1); |
|
4
|
|
|
|
|
5
|
|
|
namespace AOE\Crawler; |
|
6
|
|
|
|
|
7
|
|
|
/* |
|
8
|
|
|
* (c) 2020 AOE GmbH <[email protected]> |
|
9
|
|
|
* |
|
10
|
|
|
* This file is part of the TYPO3 Crawler Extension. |
|
11
|
|
|
* |
|
12
|
|
|
* It is free software; you can redistribute it and/or modify it under |
|
13
|
|
|
* the terms of the GNU General Public License, either version 2 |
|
14
|
|
|
* of the License, or any later version. |
|
15
|
|
|
* |
|
16
|
|
|
* For the full copyright and license information, please read the |
|
17
|
|
|
* LICENSE.txt file that was distributed with this source code. |
|
18
|
|
|
* |
|
19
|
|
|
* The TYPO3 project - inspiring people to share! |
|
20
|
|
|
*/ |
|
21
|
|
|
|
|
22
|
|
|
use AOE\Crawler\Controller\CrawlerController; |
|
23
|
|
|
use AOE\Crawler\Converter\JsonCompatibilityConverter; |
|
24
|
|
|
use AOE\Crawler\CrawlStrategy\CallbackExecutionStrategy; |
|
25
|
|
|
use AOE\Crawler\CrawlStrategy\CrawlStrategy; |
|
26
|
|
|
use AOE\Crawler\CrawlStrategy\CrawlStrategyFactory; |
|
27
|
|
|
use AOE\Crawler\Utility\SignalSlotUtility; |
|
28
|
|
|
use TYPO3\CMS\Core\Http\Uri; |
|
29
|
|
|
use TYPO3\CMS\Core\SingletonInterface; |
|
30
|
|
|
use TYPO3\CMS\Core\Utility\GeneralUtility; |
|
31
|
|
|
|
|
32
|
|
|
/** |
|
33
|
|
|
* Fetches a URL based on the selected strategy or via a callback. |
|
34
|
|
|
*/ |
|
35
|
|
|
class QueueExecutor implements SingletonInterface |
|
36
|
|
|
{ |
|
37
|
|
|
/** |
|
38
|
|
|
* @var CrawlStrategy |
|
39
|
|
|
*/ |
|
40
|
|
|
protected $crawlStrategy; |
|
41
|
|
|
|
|
42
|
46 |
|
public function __construct(CrawlStrategyFactory $crawlStrategyFactory) |
|
43
|
|
|
{ |
|
44
|
46 |
|
$this->crawlStrategy = $crawlStrategyFactory->create(); |
|
45
|
46 |
|
} |
|
46
|
|
|
|
|
47
|
|
|
/** |
|
48
|
|
|
* Takes a queue record and fetches the contents of the URL. |
|
49
|
|
|
* In the future, updating the queue item & additional signal/slot/events should also happen in here. |
|
50
|
|
|
* |
|
51
|
|
|
* @return array|bool|mixed|string |
|
52
|
|
|
*/ |
|
53
|
7 |
|
public function executeQueueItem(array $queueItem, CrawlerController $crawlerController) |
|
54
|
|
|
{ |
|
55
|
7 |
|
$parameters = ''; |
|
56
|
7 |
|
if (isset($queueItem['parameters'])) { |
|
57
|
|
|
// Decode parameters: |
|
58
|
|
|
/** @var JsonCompatibilityConverter $jsonCompatibleConverter */ |
|
59
|
6 |
|
$jsonCompatibleConverter = GeneralUtility::makeInstance(JsonCompatibilityConverter::class); |
|
60
|
6 |
|
$parameters = $jsonCompatibleConverter->convert($queueItem['parameters']); |
|
61
|
|
|
} |
|
62
|
|
|
|
|
63
|
7 |
|
if (! is_array($parameters) || empty($parameters)) { |
|
64
|
6 |
|
return 'ERROR'; |
|
65
|
|
|
} |
|
66
|
1 |
|
if ($parameters['_CALLBACKOBJ']) { |
|
67
|
1 |
|
$className = $parameters['_CALLBACKOBJ']; |
|
68
|
1 |
|
unset($parameters['_CALLBACKOBJ']); |
|
69
|
1 |
|
$result = GeneralUtility::makeInstance(CallbackExecutionStrategy::class) |
|
70
|
1 |
|
->fetchByCallback($className, $parameters, $crawlerController); |
|
71
|
1 |
|
$result = ['content' => json_encode($result)]; |
|
72
|
|
|
} else { |
|
73
|
|
|
// Regular FE request |
|
74
|
|
|
$crawlerId = $this->generateCrawlerIdFromQueueItem($queueItem); |
|
75
|
|
|
|
|
76
|
|
|
// Get result: |
|
77
|
|
|
$url = new Uri($parameters['url']); |
|
78
|
|
|
$result = $this->crawlStrategy->fetchUrlContents($url, $crawlerId); |
|
79
|
|
|
if ($result !== false) { |
|
80
|
|
|
$result = ['content' => json_encode($result)]; |
|
81
|
|
|
} |
|
82
|
|
|
|
|
83
|
|
|
$signalPayload = ['url' => $parameters['url'], 'result' => $result]; |
|
84
|
|
|
SignalSlotUtility::emitSignal( |
|
|
|
|
|
|
85
|
|
|
self::class, |
|
86
|
|
|
SignalSlotUtility::SIGNAL_URL_CRAWLED, |
|
87
|
|
|
$signalPayload |
|
88
|
|
|
); |
|
89
|
|
|
} |
|
90
|
1 |
|
return $result; |
|
91
|
|
|
} |
|
92
|
|
|
|
|
93
|
|
|
protected function generateCrawlerIdFromQueueItem(array $queueItem): string |
|
94
|
|
|
{ |
|
95
|
|
|
return $queueItem['qid'] . ':' . md5($queueItem['qid'] . '|' . $queueItem['set_id'] . '|' . $GLOBALS['TYPO3_CONF_VARS']['SYS']['encryptionKey']); |
|
96
|
|
|
} |
|
97
|
|
|
} |
|
98
|
|
|
|
AOEpeople /
crawler
Loading history...