1
|
|
|
<?php |
2
|
|
|
|
3
|
|
|
declare(strict_types=1); |
4
|
|
|
|
5
|
|
|
namespace AOE\Crawler\CrawlStrategy; |
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\Configuration\ExtensionConfigurationProvider; |
23
|
|
|
use AOE\Crawler\Utility\PhpBinaryUtility; |
24
|
|
|
use Psr\Http\Message\UriInterface; |
25
|
|
|
use Psr\Log\LoggerAwareInterface; |
26
|
|
|
use Psr\Log\LoggerAwareTrait; |
27
|
|
|
use TYPO3\CMS\Core\Core\Environment; |
28
|
|
|
use TYPO3\CMS\Core\Utility\CommandUtility; |
29
|
|
|
use TYPO3\CMS\Core\Utility\ExtensionManagementUtility; |
30
|
|
|
use TYPO3\CMS\Core\Utility\GeneralUtility; |
31
|
|
|
|
32
|
|
|
/** |
33
|
|
|
* Executes another process via shell_exec() to include cli/bootstrap.php which in turn |
34
|
|
|
* includes the index.php for frontend. |
35
|
|
|
*/ |
36
|
|
|
class SubProcessExecutionStrategy implements LoggerAwareInterface, CrawlStrategy |
37
|
|
|
{ |
38
|
|
|
use LoggerAwareTrait; |
39
|
|
|
|
40
|
|
|
/** |
41
|
|
|
* @var array |
42
|
|
|
*/ |
43
|
|
|
protected $extensionSettings; |
44
|
|
|
|
45
|
1 |
|
public function __construct(?ExtensionConfigurationProvider $configurationProvider = null) |
46
|
|
|
{ |
47
|
1 |
|
$configurationProvider = $configurationProvider ?? GeneralUtility::makeInstance(ExtensionConfigurationProvider::class); |
48
|
1 |
|
$settings = $configurationProvider->getExtensionConfiguration(); |
49
|
1 |
|
$this->extensionSettings = is_array($settings) ? $settings : []; |
50
|
1 |
|
} |
51
|
|
|
|
52
|
|
|
/** |
53
|
|
|
* Fetches a URL by calling a shell script. |
54
|
|
|
* |
55
|
|
|
* @return array|bool|mixed |
56
|
|
|
*/ |
57
|
|
|
public function fetchUrlContents(UriInterface $url, string $crawlerId) |
58
|
|
|
{ |
59
|
|
|
$url = (string) $url; |
60
|
|
|
$parsedUrl = parse_url($url); |
61
|
|
|
|
62
|
|
|
if ($parsedUrl === false) { |
63
|
|
|
$this->logger->debug( |
|
|
|
|
64
|
|
|
sprintf('Could not parse_url() for string "%s"', $url), |
65
|
|
|
['crawlerId' => $crawlerId] |
66
|
|
|
); |
67
|
|
|
return false; |
68
|
|
|
} |
69
|
|
|
|
70
|
|
|
if (! in_array($parsedUrl['scheme'], ['', 'http', 'https'], true)) { |
71
|
|
|
$this->logger->debug( |
72
|
|
|
sprintf('Scheme does not match for url "%s"', $url), |
73
|
|
|
['crawlerId' => $crawlerId] |
74
|
|
|
); |
75
|
|
|
return false; |
76
|
|
|
} |
77
|
|
|
|
78
|
|
|
if (! is_array($parsedUrl)) { |
|
|
|
|
79
|
|
|
return []; |
80
|
|
|
} |
81
|
|
|
|
82
|
|
|
$requestHeaders = $this->buildRequestHeaders($parsedUrl, $crawlerId); |
83
|
|
|
|
84
|
|
|
$commandParts = [ |
85
|
|
|
ExtensionManagementUtility::extPath('crawler') . 'cli/bootstrap.php', |
86
|
|
|
$this->getFrontendBasePath(), |
87
|
|
|
$url, |
88
|
|
|
base64_encode(serialize($requestHeaders)), |
89
|
|
|
]; |
90
|
|
|
$commandParts = CommandUtility::escapeShellArguments($commandParts); |
91
|
|
|
$cmd = escapeshellcmd(PhpBinaryUtility::getPhpBinary()); |
92
|
|
|
$cmd .= ' ' . implode(' ', $commandParts); |
93
|
|
|
|
94
|
|
|
$startTime = microtime(true); |
95
|
|
|
$content = $this->executeShellCommand($cmd); |
96
|
|
|
$this->logger->info($url . ' ' . (microtime(true) - $startTime)); |
97
|
|
|
|
98
|
|
|
if ($content === null) { |
|
|
|
|
99
|
|
|
return false; |
100
|
|
|
} |
101
|
|
|
return unserialize($content); |
102
|
|
|
} |
103
|
|
|
|
104
|
|
|
private function buildRequestHeaders(array $url, string $crawlerId): array |
105
|
|
|
{ |
106
|
|
|
$reqHeaders = []; |
107
|
|
|
$reqHeaders[] = 'GET ' . $url['path'] . ($url['query'] ? '?' . $url['query'] : '') . ' HTTP/1.0'; |
108
|
|
|
$reqHeaders[] = 'Host: ' . $url['host']; |
109
|
|
|
$reqHeaders[] = 'Connection: close'; |
110
|
|
|
if (isset($url['user'], $url['pass']) && $url['user'] !== '' && $url['pass'] !== '') { |
111
|
|
|
$reqHeaders[] = 'Authorization: Basic ' . base64_encode($url['user'] . ':' . $url['pass']); |
112
|
|
|
} |
113
|
|
|
$reqHeaders[] = 'X-T3crawler: ' . $crawlerId; |
114
|
|
|
$reqHeaders[] = 'User-Agent: TYPO3 crawler'; |
115
|
|
|
return $reqHeaders; |
116
|
|
|
} |
117
|
|
|
|
118
|
|
|
/** |
119
|
|
|
* Executes a shell command and returns the outputted result. |
120
|
|
|
* |
121
|
|
|
* @param string $command Shell command to be executed |
122
|
|
|
* @return string|null Outputted result of the command execution |
123
|
|
|
*/ |
124
|
|
|
private function executeShellCommand($command) |
125
|
|
|
{ |
126
|
|
|
return shell_exec($command); |
127
|
|
|
} |
128
|
|
|
|
129
|
|
|
/** |
130
|
|
|
* Gets the base path of the website frontend. |
131
|
|
|
* (e.g. if you call http://mydomain.com/cms/index.php in |
132
|
|
|
* the browser the base path is "/cms/") |
133
|
|
|
* |
134
|
|
|
* @return string Base path of the website frontend |
135
|
|
|
*/ |
136
|
|
|
private function getFrontendBasePath() |
137
|
|
|
{ |
138
|
|
|
$frontendBasePath = '/'; |
139
|
|
|
|
140
|
|
|
// Get the path from the extension settings: |
141
|
|
|
if (isset($this->extensionSettings['frontendBasePath']) && $this->extensionSettings['frontendBasePath']) { |
142
|
|
|
$frontendBasePath = $this->extensionSettings['frontendBasePath']; |
143
|
|
|
// If empty, try to use config.absRefPrefix: |
144
|
|
|
} elseif (isset($GLOBALS['TSFE']->absRefPrefix) && ! empty($GLOBALS['TSFE']->absRefPrefix)) { |
145
|
|
|
$frontendBasePath = $GLOBALS['TSFE']->absRefPrefix; |
146
|
|
|
// If not in CLI mode the base path can be determined from $_SERVER environment: |
147
|
|
|
} elseif (! Environment::isCli()) { |
148
|
|
|
$frontendBasePath = GeneralUtility::getIndpEnv('TYPO3_SITE_PATH'); |
149
|
|
|
} |
150
|
|
|
|
151
|
|
|
// Base path must be '/<pathSegements>/': |
152
|
|
|
if ($frontendBasePath !== '/') { |
153
|
|
|
$frontendBasePath = '/' . ltrim($frontendBasePath, '/'); |
154
|
|
|
$frontendBasePath = rtrim($frontendBasePath, '/') . '/'; |
155
|
|
|
} |
156
|
|
|
|
157
|
|
|
return $frontendBasePath; |
158
|
|
|
} |
159
|
|
|
} |
160
|
|
|
|
This check looks for calls to methods that do not seem to exist on a given type. It looks for the method on the type itself as well as in inherited classes or implemented interfaces.
This is most likely a typographical error or the method has been renamed.