1
|
|
|
<?php |
2
|
|
|
|
3
|
|
|
declare(strict_types=1); |
4
|
|
|
|
5
|
|
|
namespace AOE\Crawler\Middleware; |
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 Psr\Http\Message\ResponseInterface; |
23
|
|
|
use Psr\Http\Message\ServerRequestInterface; |
24
|
|
|
use Psr\Http\Server\MiddlewareInterface; |
25
|
|
|
use Psr\Http\Server\RequestHandlerInterface; |
26
|
|
|
use TYPO3\CMS\Core\Context\Context; |
27
|
|
|
use TYPO3\CMS\Core\Http\Response; |
28
|
|
|
use TYPO3\CMS\Core\Utility\GeneralUtility; |
29
|
|
|
|
30
|
|
|
/** |
31
|
|
|
* Evaluates HTTP headers and checks if Crawler should register itself. |
32
|
|
|
* Needs to be run after TSFE initialization AND Frontend User Authentication. |
33
|
|
|
* |
34
|
|
|
* Once done, the queue is fetched, and then the frontend request runs through. |
35
|
|
|
* |
36
|
|
|
* Finally, at the very end, if the crawler is still running, output the data and replace the response. |
37
|
|
|
*/ |
38
|
|
|
class CrawlerInitialization implements MiddlewareInterface |
39
|
|
|
{ |
40
|
|
|
/** |
41
|
|
|
* @var Context |
42
|
|
|
*/ |
43
|
|
|
protected $context; |
44
|
|
|
|
45
|
|
|
public function __construct(?Context $context = null) |
46
|
|
|
{ |
47
|
|
|
$this->context = $context ?? GeneralUtility::makeInstance(Context::class); |
48
|
|
|
} |
49
|
|
|
|
50
|
|
|
/** |
51
|
|
|
* @throws \TYPO3\CMS\Core\Context\Exception\AspectNotFoundException |
52
|
|
|
* @throws \TYPO3\CMS\Core\Error\Http\ServiceUnavailableException |
53
|
|
|
*/ |
54
|
|
|
public function process(ServerRequestInterface $request, RequestHandlerInterface $handler): ResponseInterface |
55
|
|
|
{ |
56
|
|
|
$queueParameters = $request->getAttribute('tx_crawler'); |
57
|
|
|
if ($queueParameters === null) { |
58
|
|
|
return $handler->handle($request); |
59
|
|
|
} |
60
|
|
|
|
61
|
|
|
$GLOBALS['TSFE']->applicationData['forceIndexing'] = true; |
62
|
|
|
$GLOBALS['TSFE']->applicationData['tx_crawler']['running'] = true; |
63
|
|
|
$GLOBALS['TSFE']->applicationData['tx_crawler']['parameters'] = $queueParameters; |
64
|
|
|
$GLOBALS['TSFE']->applicationData['tx_crawler']['log'] = [ |
65
|
|
|
'User Groups: ' . $queueParameters['feUserGroupList'], |
66
|
|
|
]; |
67
|
|
|
|
68
|
|
|
// Execute the frontend request as is |
69
|
|
|
$handler->handle($request); |
70
|
|
|
|
71
|
|
|
$GLOBALS['TSFE']->applicationData['tx_crawler']['vars'] = [ |
72
|
|
|
'id' => $GLOBALS['TSFE']->id, |
73
|
|
|
'gr_list' => implode(',', $this->context->getAspect('frontend.user')->getGroupIds()), |
|
|
|
|
74
|
|
|
'no_cache' => $GLOBALS['TSFE']->no_cache, |
75
|
|
|
]; |
76
|
|
|
|
77
|
|
|
$this->runPollSuccessHooks(); |
78
|
|
|
|
79
|
|
|
// Output log data for crawler (serialized content): |
80
|
|
|
$content = serialize($GLOBALS['TSFE']->applicationData['tx_crawler']); |
81
|
|
|
$response = new Response(); |
82
|
|
|
$response->getBody()->write($content); |
83
|
|
|
|
84
|
|
|
return $response; |
85
|
|
|
} |
86
|
|
|
|
87
|
|
|
/** |
88
|
|
|
* Required because some extensions (staticpub) might never be requested to run due to some Core side effects |
89
|
|
|
* and since this is considered as error the crawler should handle it properly |
90
|
|
|
*/ |
91
|
|
|
protected function runPollSuccessHooks(): void |
92
|
|
|
{ |
93
|
|
|
if (! is_array($GLOBALS['TSFE']->applicationData['tx_crawler']['content']['parameters']['procInstructions'])) { |
94
|
|
|
return; |
95
|
|
|
} |
96
|
|
|
foreach ($GLOBALS['TYPO3_CONF_VARS']['EXTCONF']['crawler']['pollSuccess'] ?? [] as $pollable) { |
97
|
|
|
if (in_array($pollable, $GLOBALS['TSFE']->applicationData['tx_crawler']['content']['parameters']['procInstructions'], true)) { |
98
|
|
|
if (empty($GLOBALS['TSFE']->applicationData['tx_crawler']['success'][$pollable])) { |
99
|
|
|
$GLOBALS['TSFE']->applicationData['tx_crawler']['errorlog'][] = 'Error: Pollable extension (' . $pollable . ') did not complete successfully.'; |
100
|
|
|
} |
101
|
|
|
} |
102
|
|
|
} |
103
|
|
|
} |
104
|
|
|
} |
105
|
|
|
|