|
1
|
|
|
<?php |
|
2
|
|
|
/** |
|
3
|
|
|
* This file is part of dispositif/wikibot application |
|
4
|
|
|
* 2019 : Philippe M. <[email protected]> |
|
5
|
|
|
* For the full copyright and MIT license information, please view the LICENSE file. |
|
6
|
|
|
*/ |
|
7
|
|
|
|
|
8
|
|
|
declare(strict_types=1); |
|
9
|
|
|
|
|
10
|
|
|
namespace App\Application; |
|
11
|
|
|
|
|
12
|
|
|
use App\Infrastructure\TagParser; |
|
13
|
|
|
use Exception; |
|
14
|
|
|
use GuzzleHttp\Client; |
|
15
|
|
|
|
|
16
|
|
|
/** |
|
17
|
|
|
* todo Move Infra? |
|
18
|
|
|
* Class PublisherAction |
|
19
|
|
|
* |
|
20
|
|
|
* @package App\Application |
|
21
|
|
|
*/ |
|
22
|
|
|
class PublisherAction |
|
23
|
|
|
{ |
|
24
|
|
|
private $url; |
|
25
|
|
|
|
|
26
|
|
|
public function __construct(string $url) |
|
27
|
|
|
{ |
|
28
|
|
|
$this->url = $url; |
|
29
|
|
|
} |
|
30
|
|
|
|
|
31
|
|
|
public function getUrl(): ?string |
|
32
|
|
|
{ |
|
33
|
|
|
return $this->url; |
|
34
|
|
|
} |
|
35
|
|
|
|
|
36
|
|
|
/** |
|
37
|
|
|
* import source from URL with Guzzle. |
|
38
|
|
|
* |
|
39
|
|
|
* @return string|null |
|
40
|
|
|
* @throws Exception |
|
41
|
|
|
*/ |
|
42
|
|
|
public function getHTMLSource(): string |
|
43
|
|
|
{ |
|
44
|
|
|
$client = new Client( |
|
45
|
|
|
[ |
|
46
|
|
|
'timeout' => 5, |
|
47
|
|
|
'headers' => ['User-Agent' => getenv('USER_AGENT')], |
|
48
|
|
|
] |
|
49
|
|
|
); |
|
50
|
|
|
$response = $client->get($this->url); |
|
51
|
|
|
|
|
52
|
|
|
if (200 !== $response->getStatusCode()) { |
|
53
|
|
|
throw new Exception('response error '.$response->getStatusCode().' '.$response->getReasonPhrase()); |
|
54
|
|
|
} |
|
55
|
|
|
|
|
56
|
|
|
return $response->getBody()->getContents(); |
|
57
|
|
|
} |
|
58
|
|
|
|
|
59
|
|
|
/** |
|
60
|
|
|
* @param string $html |
|
61
|
|
|
* |
|
62
|
|
|
* @return array |
|
63
|
|
|
* @throws Exception |
|
64
|
|
|
*/ |
|
65
|
|
|
public function extractWebData(string $html): array |
|
66
|
|
|
{ |
|
67
|
|
|
$ld = $this->extractLdJson($html); |
|
68
|
|
|
$meta = $this->getMetaTags($html); |
|
69
|
|
|
|
|
70
|
|
|
return ['JSON-LD' => $ld, 'meta' => $meta]; |
|
71
|
|
|
} |
|
72
|
|
|
|
|
73
|
|
|
/** |
|
74
|
|
|
* extract LD-JSON metadata from <script type="application/ld+json">. |
|
75
|
|
|
* |
|
76
|
|
|
* @param string $html |
|
77
|
|
|
* |
|
78
|
|
|
* @return array |
|
79
|
|
|
* @throws Exception |
|
80
|
|
|
*/ |
|
81
|
|
|
private function extractLdJson(string $html): array |
|
82
|
|
|
{ |
|
83
|
|
|
$parser = new TagParser(); |
|
84
|
|
|
$results = $parser->importHtml($html)->xpathResults( |
|
85
|
|
|
'//script[@type="application/ld+json"]' |
|
86
|
|
|
); |
|
87
|
|
|
|
|
88
|
|
|
foreach ($results as $result) { |
|
89
|
|
|
$json = trim($result); |
|
90
|
|
|
// filtrage empty value (todo?) |
|
91
|
|
|
if (0 === strlen($json)) { |
|
92
|
|
|
continue; |
|
93
|
|
|
} |
|
94
|
|
|
$data = json_decode($json, true); |
|
95
|
|
|
if (!is_array($data)) { |
|
96
|
|
|
return []; |
|
97
|
|
|
} |
|
98
|
|
|
|
|
99
|
|
|
// filtrage : @type => BreadcrumbList (lemonde) |
|
100
|
|
|
// TODO : c'est quoi ça ? |
|
101
|
|
|
if (isset($data['@type']) && 'BreadcrumbList' === $data['@type']) { |
|
102
|
|
|
continue; |
|
103
|
|
|
} |
|
104
|
|
|
|
|
105
|
|
|
return $data; |
|
106
|
|
|
} |
|
107
|
|
|
|
|
108
|
|
|
return []; |
|
109
|
|
|
} |
|
110
|
|
|
|
|
111
|
|
|
/** |
|
112
|
|
|
* todo move/refac/delete? |
|
113
|
|
|
* |
|
114
|
|
|
* @param string $str |
|
115
|
|
|
* |
|
116
|
|
|
* @return array |
|
117
|
|
|
*/ |
|
118
|
|
|
private function getMetaTags(string $str): array |
|
119
|
|
|
{ |
|
120
|
|
|
$pattern = ' |
|
121
|
|
|
~<\s*meta\s |
|
122
|
|
|
# using lookahead to capture type to $1 |
|
123
|
|
|
(?=[^>]*? |
|
124
|
|
|
\b(?:name|property|http-equiv)\s*=\s* |
|
125
|
|
|
(?|"\s*([^"]*?)\s*"|\'\s*([^\']*?)\s*\'| |
|
126
|
|
|
([^"\'>]*?)(?=\s*/?\s*>|\s\w+\s*=)) |
|
127
|
|
|
) |
|
128
|
|
|
# capture content to $2 |
|
129
|
|
|
[^>]*?\bcontent\s*=\s* |
|
130
|
|
|
(?|"\s*([^"]*?)\s*"|\'\s*([^\']*?)\s*\'| |
|
131
|
|
|
([^"\'>]*?)(?=\s*/?\s*>|\s\w+\s*=)) |
|
132
|
|
|
[^>]*> |
|
133
|
|
|
~ix'; |
|
134
|
|
|
|
|
135
|
|
|
if (preg_match_all($pattern, $str, $out)) { |
|
136
|
|
|
return array_combine($out[1], $out[2]); |
|
|
|
|
|
|
137
|
|
|
} |
|
138
|
|
|
|
|
139
|
|
|
return []; |
|
140
|
|
|
} |
|
141
|
|
|
} |
|
142
|
|
|
|
Dispositif /
Wikibot
Loading history...