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