|
1
|
|
|
<?php |
|
2
|
|
|
namespace AOE\Crawler\Task; |
|
3
|
|
|
|
|
4
|
|
|
/*************************************************************** |
|
5
|
|
|
* Copyright notice |
|
6
|
|
|
* |
|
7
|
|
|
* (c) 2016 AOE GmbH <[email protected]> |
|
8
|
|
|
* |
|
9
|
|
|
* All rights reserved |
|
10
|
|
|
* |
|
11
|
|
|
* This script is part of the TYPO3 project. The TYPO3 project is |
|
12
|
|
|
* free software; you can redistribute it and/or modify |
|
13
|
|
|
* it under the terms of the GNU General Public License as published by |
|
14
|
|
|
* the Free Software Foundation; either version 3 of the License, or |
|
15
|
|
|
* (at your option) any later version. |
|
16
|
|
|
* |
|
17
|
|
|
* The GNU General Public License can be found at |
|
18
|
|
|
* http://www.gnu.org/copyleft/gpl.html. |
|
19
|
|
|
* |
|
20
|
|
|
* This script is distributed in the hope that it will be useful, |
|
21
|
|
|
* but WITHOUT ANY WARRANTY; without even the implied warranty of |
|
22
|
|
|
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the |
|
23
|
|
|
* GNU General Public License for more details. |
|
24
|
|
|
* |
|
25
|
|
|
* This copyright notice MUST APPEAR in all copies of the script! |
|
26
|
|
|
***************************************************************/ |
|
27
|
|
|
|
|
28
|
|
|
use AOE\Crawler\Controller\CrawlerController; |
|
29
|
|
|
use TYPO3\CMS\Core\Utility\GeneralUtility; |
|
30
|
|
|
use TYPO3\CMS\Scheduler\Task\AbstractTask; |
|
31
|
|
|
|
|
32
|
|
|
/** |
|
33
|
|
|
* Class CrawlerQueueTask |
|
34
|
|
|
* |
|
35
|
|
|
* @package AOE\Crawler\Task |
|
36
|
|
|
* @codeCoverageIgnore |
|
37
|
|
|
*/ |
|
38
|
|
|
class CrawlerQueueTask extends AbstractTask |
|
39
|
|
|
{ |
|
40
|
|
|
/** |
|
41
|
|
|
* Define the current mode to process the crawler |
|
42
|
|
|
* |
|
43
|
|
|
* @var string |
|
44
|
|
|
*/ |
|
45
|
|
|
const MODE = 'queue'; |
|
46
|
|
|
|
|
47
|
|
|
/** |
|
48
|
|
|
* Depth to run the crawler |
|
49
|
|
|
* |
|
50
|
|
|
* @var integer |
|
51
|
|
|
*/ |
|
52
|
|
|
public $depth; |
|
53
|
|
|
|
|
54
|
|
|
/** |
|
55
|
|
|
* Startpage for the crawler |
|
56
|
|
|
* |
|
57
|
|
|
* @var integer |
|
58
|
|
|
*/ |
|
59
|
|
|
public $startPage; |
|
60
|
|
|
|
|
61
|
|
|
/** |
|
62
|
|
|
* Configuration to run (comma separated list) |
|
63
|
|
|
* |
|
64
|
|
|
* @var string |
|
65
|
|
|
*/ |
|
66
|
|
|
public $configuration; |
|
67
|
|
|
|
|
68
|
|
|
/** |
|
69
|
|
|
* Function executed from the Scheduler. |
|
70
|
|
|
* |
|
71
|
|
|
* @return bool |
|
72
|
|
|
*/ |
|
73
|
|
|
public function execute() |
|
74
|
|
|
{ |
|
75
|
|
|
$this->setCliArguments(); |
|
76
|
|
|
|
|
77
|
|
|
/** @var CrawlerController $crawlerObject */ |
|
78
|
|
|
$crawlerObject = GeneralUtility::makeInstance(CrawlerController::class); |
|
79
|
|
|
$crawlerObject->CLI_main_im(); |
|
80
|
|
|
return true; |
|
81
|
|
|
} |
|
82
|
|
|
|
|
83
|
|
|
/** |
|
84
|
|
|
* Simulate cli call with setting the required options to the $_SERVER['argv'] |
|
85
|
|
|
* |
|
86
|
|
|
* @return void |
|
87
|
|
|
*/ |
|
88
|
|
|
protected function setCliArguments() |
|
89
|
|
|
{ |
|
90
|
|
|
// Make it backwards compatible. |
|
91
|
|
|
if (is_null($this->startPage)) { |
|
92
|
|
|
$this->startPage = 0; |
|
93
|
|
|
} |
|
94
|
|
|
|
|
95
|
|
|
$_SERVER['argv'] = [$_SERVER['argv'][0], $this->startPage,'-ss', '-d', $this->depth, '-o', self::MODE, '-conf', implode(',', $this->configuration)]; |
|
96
|
|
|
} |
|
97
|
|
|
|
|
98
|
|
|
/** |
|
99
|
|
|
* Retrieve some details about current scheduler task |
|
100
|
|
|
* to make the list view more useful. |
|
101
|
|
|
* |
|
102
|
|
|
* @return string |
|
103
|
|
|
*/ |
|
104
|
|
|
public function getAdditionalInformation() |
|
105
|
|
|
{ |
|
106
|
|
|
// Make it backwards compatible. |
|
107
|
|
|
if (is_null($this->startPage)) { |
|
108
|
|
|
$this->startPage = 0; |
|
109
|
|
|
} |
|
110
|
|
|
|
|
111
|
|
|
return implode(',', $this->configuration) . ' (depth: ' . $this->depth . ', startPage:' . $this->startPage . ')'; |
|
112
|
|
|
} |
|
113
|
|
|
} |
|
114
|
|
|
|
AOEpeople /
crawler
