|
1
|
|
|
<?php |
|
2
|
|
|
|
|
3
|
|
|
namespace shweshi\OpenGraph; |
|
4
|
|
|
|
|
5
|
|
|
use DOMDocument; |
|
6
|
|
|
use Illuminate\Support\Facades\Log; |
|
7
|
|
|
use shweshi\OpenGraph\Exceptions\FetchException; |
|
8
|
|
|
|
|
9
|
|
|
class OpenGraph |
|
10
|
|
|
{ |
|
11
|
|
|
|
|
12
|
|
|
private const string DEFAULT_USER_AGENT = 'Curl'; |
|
|
|
|
|
|
13
|
|
|
|
|
14
|
|
|
private string $userAgent = self::DEFAULT_USER_AGENT; |
|
15
|
|
|
|
|
16
|
|
|
/** |
|
17
|
|
|
* @throws FetchException |
|
18
|
|
|
*/ |
|
19
|
|
|
public function fetch($url, $allMeta = null, $lang = null, $options = LIBXML_NOWARNING | LIBXML_NOERROR): array |
|
20
|
|
|
{ |
|
21
|
|
|
$html = $this->curl_get_contents($url, $lang, $this->userAgent); |
|
22
|
|
|
/** |
|
23
|
|
|
* parsing starts here:. |
|
24
|
|
|
*/ |
|
25
|
|
|
$doc = new DOMDocument(); |
|
26
|
|
|
|
|
27
|
|
|
$libxml_previous_state = libxml_use_internal_errors(true); |
|
28
|
|
|
$doc->loadHTML('<?xml encoding="utf-8" ?>'.$html, $options); |
|
29
|
|
|
//catch possible errors due to empty or malformed HTML |
|
30
|
|
|
if ($options > 0 && ($options & (LIBXML_NOWARNING | LIBXML_NOERROR)) == 0) { |
|
31
|
|
|
Log::warning(libxml_get_errors()); |
|
32
|
|
|
} |
|
33
|
|
|
libxml_clear_errors(); |
|
34
|
|
|
// restore previous state |
|
35
|
|
|
libxml_use_internal_errors($libxml_previous_state); |
|
36
|
|
|
|
|
37
|
|
|
$tags = $doc->getElementsByTagName('meta'); |
|
38
|
|
|
$metadata = []; |
|
39
|
|
|
foreach ($tags as $tag) { |
|
40
|
|
|
$metaProperty = ($tag->hasAttribute('property')) ? $tag->getAttribute('property') : $tag->getAttribute('name'); |
|
41
|
|
|
if (!$allMeta && $metaProperty && str_starts_with($tag->getAttribute('property'), 'og:')) { |
|
42
|
|
|
$key = strtr(substr($metaProperty, 3), '-', '_'); |
|
43
|
|
|
$value = $this->get_meta_value($tag); |
|
44
|
|
|
} |
|
45
|
|
|
if ($allMeta && $metaProperty) { |
|
46
|
|
|
$key = (str_starts_with($metaProperty, 'og:')) ? strtr(substr($metaProperty, 3), '-', '_') : $metaProperty; |
|
47
|
|
|
$value = $this->get_meta_value($tag); |
|
48
|
|
|
} |
|
49
|
|
|
if (!empty($key)) { |
|
50
|
|
|
$metadata[$key] = $value; |
|
51
|
|
|
} |
|
52
|
|
|
/* |
|
53
|
|
|
* Verify image url |
|
54
|
|
|
*/ |
|
55
|
|
|
if (isset($metadata['image'])) { |
|
56
|
|
|
$isValidImageUrl = $this->verify_image_url($metadata['image']); |
|
57
|
|
|
if (!$isValidImageUrl) { |
|
58
|
|
|
$metadata['image'] = ''; |
|
59
|
|
|
} |
|
60
|
|
|
} |
|
61
|
|
|
} |
|
62
|
|
|
|
|
63
|
|
|
return $metadata; |
|
64
|
|
|
} |
|
65
|
|
|
|
|
66
|
|
|
/** |
|
67
|
|
|
* @throws FetchException |
|
68
|
|
|
*/ |
|
69
|
|
|
protected function curl_get_contents($url, $lang, $userAgent): bool|string |
|
70
|
|
|
{ |
|
71
|
|
|
$headers = [ |
|
72
|
|
|
'Accept: text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8', |
|
73
|
|
|
'Cache-Control: no-cache', |
|
74
|
|
|
'User-Agent: '.$userAgent, |
|
75
|
|
|
]; |
|
76
|
|
|
|
|
77
|
|
|
if ($lang) { |
|
78
|
|
|
$headers[] = 'Accept-Language: '.$lang; |
|
79
|
|
|
} |
|
80
|
|
|
|
|
81
|
|
|
$curl = curl_init(); |
|
82
|
|
|
|
|
83
|
|
|
curl_setopt_array($curl, [ |
|
84
|
|
|
CURLOPT_URL => $url, |
|
85
|
|
|
CURLOPT_FAILONERROR => false, |
|
86
|
|
|
CURLOPT_FOLLOWLOCATION => true, |
|
87
|
|
|
CURLOPT_RETURNTRANSFER => true, |
|
88
|
|
|
CURLOPT_SSL_VERIFYHOST => 2, |
|
89
|
|
|
CURLOPT_SSL_VERIFYPEER => true, |
|
90
|
|
|
CURLOPT_ENCODING => 'UTF-8', |
|
91
|
|
|
CURLOPT_MAXREDIRS => 10, |
|
92
|
|
|
CURLOPT_TIMEOUT => 30, |
|
93
|
|
|
CURLOPT_HTTP_VERSION => CURL_HTTP_VERSION_1_1, |
|
94
|
|
|
CURLOPT_CUSTOMREQUEST => 'GET', |
|
95
|
|
|
CURLOPT_HTTPHEADER => $headers, |
|
96
|
|
|
]); |
|
97
|
|
|
|
|
98
|
|
|
$response = curl_exec($curl); |
|
99
|
|
|
if (curl_errno(/** @scrutinizer ignore-type */ $curl) !== 0) { |
|
100
|
|
|
throw new FetchException(curl_error(/** @scrutinizer ignore-type */ $curl), curl_errno($curl), null, curl_getinfo(/** @scrutinizer ignore-type */ $curl)); |
|
101
|
|
|
} |
|
102
|
|
|
|
|
103
|
|
|
curl_close(/** @scrutinizer ignore-type */ $curl); |
|
104
|
|
|
|
|
105
|
|
|
return $response; |
|
106
|
|
|
} |
|
107
|
|
|
|
|
108
|
|
|
protected function verify_image_url($url) |
|
109
|
|
|
{ |
|
110
|
|
|
$path = parse_url($url, PHP_URL_PATH); |
|
111
|
|
|
$encoded_path = array_map('urlencode', explode('/', $path)); |
|
112
|
|
|
$url = str_replace($path, implode('/', $encoded_path), $url); |
|
113
|
|
|
if (!filter_var($url, FILTER_VALIDATE_URL)) { |
|
114
|
|
|
return false; |
|
115
|
|
|
} |
|
116
|
|
|
|
|
117
|
|
|
try { |
|
118
|
|
|
$contextHeaders = ['ssl' => [ |
|
119
|
|
|
'verify_peer' => false, |
|
120
|
|
|
'verify_peer_name' => false, |
|
121
|
|
|
]]; |
|
122
|
|
|
|
|
123
|
|
|
if ($this->userAgent !== self::DEFAULT_USER_AGENT) { |
|
124
|
|
|
$contextHeaders['http'] = [ |
|
125
|
|
|
'user_agent' => $this->userAgent, |
|
126
|
|
|
]; |
|
127
|
|
|
} |
|
128
|
|
|
|
|
129
|
|
|
$streamContext = stream_context_create($contextHeaders); |
|
130
|
|
|
$headers = get_headers($url, true, $streamContext); |
|
131
|
|
|
|
|
132
|
|
|
return (bool) stripos($headers[0], '200 OK'); |
|
133
|
|
|
} catch (\Exception $e) { |
|
134
|
|
|
Log::error($e->getMessage()); |
|
135
|
|
|
return false; |
|
136
|
|
|
} |
|
137
|
|
|
} |
|
138
|
|
|
|
|
139
|
|
|
protected function get_meta_value($tag) |
|
140
|
|
|
{ |
|
141
|
|
|
if (!empty($tag->getAttribute('content'))) { |
|
142
|
|
|
$value = $tag->getAttribute('content'); |
|
143
|
|
|
} elseif (!empty($tag->getAttribute('value'))) { |
|
144
|
|
|
$value = $tag->getAttribute('value'); |
|
145
|
|
|
} else { |
|
146
|
|
|
$value = ''; |
|
147
|
|
|
} |
|
148
|
|
|
|
|
149
|
|
|
return $value; |
|
150
|
|
|
} |
|
151
|
|
|
|
|
152
|
|
|
public function userAgent(string $userAgent): self |
|
153
|
|
|
{ |
|
154
|
|
|
$this->userAgent = $userAgent; |
|
155
|
|
|
|
|
156
|
|
|
return $this; |
|
157
|
|
|
} |
|
158
|
|
|
} |
|
159
|
|
|
|
shweshi /
OpenGraph
Loading history...