|
1
|
|
|
<?php |
|
2
|
|
|
/* |
|
3
|
|
|
* This file is part of dispositif/wikibot application (@github) |
|
4
|
|
|
* 2019-2023 © Philippe M./Irønie <[email protected]> |
|
5
|
|
|
* For the full copyright and MIT license information, view the license file. |
|
6
|
|
|
*/ |
|
7
|
|
|
|
|
8
|
|
|
declare(strict_types=1); |
|
9
|
|
|
|
|
10
|
|
|
|
|
11
|
|
|
namespace App\Domain\ExternLink; |
|
12
|
|
|
|
|
13
|
|
|
use App\Application\Http\ExternHttpClient; |
|
14
|
|
|
use App\Domain\Utils\TextUtil; |
|
15
|
|
|
use App\Infrastructure\InternetDomainParser; |
|
16
|
|
|
use Exception; |
|
17
|
|
|
use Psr\Log\LoggerInterface; |
|
18
|
|
|
use Psr\Log\NullLogger; |
|
19
|
|
|
|
|
20
|
|
|
/** |
|
21
|
|
|
* Représente une page web d'un Lien Externe (hors wiki) |
|
22
|
|
|
* Class ExternPage |
|
23
|
|
|
* @package App\Domain |
|
24
|
|
|
*/ |
|
25
|
|
|
class ExternPage |
|
26
|
|
|
{ |
|
27
|
|
|
// todo move to config |
|
28
|
|
|
protected const PRETTY_DOMAIN_EXCLUSION |
|
29
|
|
|
= [ |
|
30
|
|
|
'.中国', |
|
31
|
|
|
'.gov', |
|
32
|
|
|
'.free.fr', |
|
33
|
|
|
'.gouv.fr', |
|
34
|
|
|
'.com.cn', |
|
35
|
|
|
'site.google.com', |
|
36
|
|
|
'wordpress.com', |
|
37
|
|
|
'blogspot.com', |
|
38
|
|
|
]; |
|
39
|
|
|
|
|
40
|
|
|
/** |
|
41
|
|
|
* @var string |
|
42
|
|
|
*/ |
|
43
|
|
|
private $url; |
|
44
|
|
|
|
|
45
|
|
|
/** |
|
46
|
|
|
* @var string |
|
47
|
|
|
*/ |
|
48
|
|
|
private $html; |
|
49
|
|
|
|
|
50
|
|
|
/** @var TagParserInterface|null */ |
|
51
|
|
|
private $tagParser; |
|
52
|
|
|
|
|
53
|
|
|
/** @var InternetDomainParserInterface|null */ |
|
54
|
|
|
private $domainParser; |
|
55
|
|
|
|
|
56
|
|
|
/** @var LoggerInterface */ |
|
57
|
|
|
private $log; |
|
58
|
|
|
|
|
59
|
|
|
/** |
|
60
|
|
|
* ExternPage constructor. |
|
61
|
|
|
* @throws Exception |
|
62
|
|
|
*/ |
|
63
|
|
|
public function __construct( |
|
64
|
|
|
string $url, |
|
65
|
|
|
string $html, |
|
66
|
|
|
?TagParserInterface $tagParser = null, |
|
67
|
|
|
?InternetDomainParserInterface $domainParser = null, |
|
68
|
|
|
?LoggerInterface $log = null |
|
69
|
|
|
) |
|
70
|
|
|
{ |
|
71
|
|
|
if (!ExternHttpClient::isHttpURL($url)) { |
|
72
|
|
|
throw new Exception('string is not an URL ' . $url); |
|
73
|
|
|
} |
|
74
|
|
|
$this->url = $url; |
|
75
|
|
|
$this->html = $html; |
|
76
|
|
|
$this->tagParser = $tagParser; |
|
77
|
|
|
$this->domainParser = $domainParser; |
|
78
|
|
|
$this->log = $log ?? new NullLogger(); |
|
79
|
|
|
} |
|
80
|
|
|
|
|
81
|
|
|
public function getUrl(): string |
|
82
|
|
|
{ |
|
83
|
|
|
return $this->url; |
|
84
|
|
|
} |
|
85
|
|
|
|
|
86
|
|
|
public function getData(): array |
|
87
|
|
|
{ |
|
88
|
|
|
$ld = $this->parseLdJson($this->html); |
|
89
|
|
|
$meta = $this->parseMetaTags($this->html); |
|
90
|
|
|
|
|
91
|
|
|
$meta['html-lang'] = $this->parseHtmlLang($this->html); // <html lang="en"> |
|
92
|
|
|
$meta['html-title'] = $this->parseHtmlTitle($this->html); |
|
93
|
|
|
$meta['html-h1'] = $this->parseHtmlFirstH1($this->html); |
|
94
|
|
|
$meta['html-url'] = $this->url; |
|
95
|
|
|
$meta['prettyDomainName'] = $this->getPrettyDomainName(); |
|
96
|
|
|
$meta['robots'] = $this->getMetaRobotsContent($this->html); |
|
97
|
|
|
|
|
98
|
|
|
return ['JSON-LD' => $ld, 'meta' => $meta]; |
|
99
|
|
|
} |
|
100
|
|
|
|
|
101
|
|
|
/** |
|
102
|
|
|
* extract LD-JSON metadata from <script type="application/ld+json">. |
|
103
|
|
|
* @throws Exception |
|
104
|
|
|
*/ |
|
105
|
|
|
private function parseLdJson(string $html): array |
|
106
|
|
|
{ |
|
107
|
|
|
if (!$this->tagParser instanceof TagParserInterface) { |
|
108
|
|
|
return []; |
|
109
|
|
|
} |
|
110
|
|
|
|
|
111
|
|
|
$results = $this->tagParser->importHtml($html)->xpathResults( |
|
112
|
|
|
'//script[@type="application/ld+json"]' |
|
113
|
|
|
); |
|
114
|
|
|
|
|
115
|
|
|
foreach ($results as $result) { |
|
116
|
|
|
$json = trim($result); |
|
117
|
|
|
// filtrage empty value (todo?) |
|
118
|
|
|
if ($json === '') { |
|
119
|
|
|
continue; |
|
120
|
|
|
} |
|
121
|
|
|
$data = json_decode($json, true, 512, JSON_THROW_ON_ERROR); |
|
122
|
|
|
if (!is_array($data) |
|
123
|
|
|
|| (isset($data['@type']) && is_string($data['@type']) && preg_match('#Breadcrumb#i', $data['@type'])) |
|
124
|
|
|
) { |
|
125
|
|
|
continue; |
|
126
|
|
|
} |
|
127
|
|
|
|
|
128
|
|
|
return $data; |
|
129
|
|
|
} |
|
130
|
|
|
|
|
131
|
|
|
return []; |
|
132
|
|
|
} |
|
133
|
|
|
|
|
134
|
|
|
/** |
|
135
|
|
|
* todo move? /refac/delete? |
|
136
|
|
|
*/ |
|
137
|
|
|
private function parseMetaTags(string $str): array |
|
138
|
|
|
{ |
|
139
|
|
|
$pattern = ' |
|
140
|
|
|
~<\s*meta\s |
|
141
|
|
|
# using lookahead to capture type to $1 |
|
142
|
|
|
(?=[^>]*? |
|
143
|
|
|
\b(?:name|property|http-equiv)\s*=\s* |
|
144
|
|
|
(?|"\s*([^"]*?)\s*"|\'\s*([^\']*?)\s*\'| |
|
145
|
|
|
([^"\'>]*?)(?=\s*/?\s*>|\s\w+\s*=)) |
|
146
|
|
|
) |
|
147
|
|
|
# capture content to $2 |
|
148
|
|
|
[^>]*?\bcontent\s*=\s* |
|
149
|
|
|
(?|"\s*([^"]*?)\s*"|\'\s*([^\']*?)\s*\'| |
|
150
|
|
|
([^"\'>]*?)(?=\s*/?\s*>|\s\w+\s*=)) |
|
151
|
|
|
[^>]*> |
|
152
|
|
|
~ix'; |
|
153
|
|
|
|
|
154
|
|
|
if (preg_match_all($pattern, $str, $out)) { |
|
155
|
|
|
$combine = array_combine($out[1], $out[2]); |
|
156
|
|
|
|
|
157
|
|
|
return $combine ?: []; |
|
158
|
|
|
} |
|
159
|
|
|
|
|
160
|
|
|
return []; |
|
161
|
|
|
} |
|
162
|
|
|
|
|
163
|
|
|
/** |
|
164
|
|
|
* test.com => test.com |
|
165
|
|
|
* bla.test.com => test.com |
|
166
|
|
|
* test.co.uk => test.co.uk (national commercial subdomain) |
|
167
|
|
|
* site.google.com => site.google.com (blog) |
|
168
|
|
|
* bla.site.google.com => site.google.com (blog) |
|
169
|
|
|
*/ |
|
170
|
|
|
public function getPrettyDomainName(): string |
|
171
|
|
|
{ |
|
172
|
|
|
// Parse custom exceptions (free.fr, gouv.fr, etc) |
|
173
|
|
|
$rawDomain = InternetDomainParser::extractSubdomainString($this->url); //only php parsing |
|
174
|
|
|
foreach (self::PRETTY_DOMAIN_EXCLUSION as $end) { |
|
175
|
|
|
if (TextUtil::str_ends_with($rawDomain, $end)) { |
|
176
|
|
|
return $this->sanitizeSubDomain($rawDomain); |
|
177
|
|
|
} |
|
178
|
|
|
} |
|
179
|
|
|
|
|
180
|
|
|
// Parse using InternetDomainParser library |
|
181
|
|
|
return $this->sanitizeSubDomain($this->getRegistrableSubDomain() ?? $rawDomain); // use lib and cached data |
|
182
|
|
|
} |
|
183
|
|
|
|
|
184
|
|
|
/** |
|
185
|
|
|
* "http://www.bla.co.uk/fubar" => "bla.co.uk" |
|
186
|
|
|
* @throws Exception |
|
187
|
|
|
*/ |
|
188
|
|
|
public function getRegistrableSubDomain(): ?string |
|
189
|
|
|
{ |
|
190
|
|
|
try { |
|
191
|
|
|
if (!ExternHttpClient::isHttpURL($this->url)) { |
|
192
|
|
|
throw new Exception('string is not an URL ' . $this->url); |
|
193
|
|
|
} |
|
194
|
|
|
if (!$this->domainParser instanceof InternetDomainParserInterface) { |
|
195
|
|
|
$this->log->notice('InternetDomainParser is not set'); |
|
196
|
|
|
|
|
197
|
|
|
return null; |
|
198
|
|
|
} |
|
199
|
|
|
|
|
200
|
|
|
return $this->domainParser->getRegistrableDomainFromURL($this->url); |
|
201
|
|
|
} catch (Exception $e) { |
|
202
|
|
|
if ($this->log !== null) { |
|
203
|
|
|
$this->log->warning('InternetDomainParser->getRegistrableDomainFromURL NULL ' . $this->url); |
|
204
|
|
|
} |
|
205
|
|
|
throw new Exception('InternetDomainParser->getRegistrableDomainFromURL NULL', $e->getCode(), $e); |
|
206
|
|
|
} |
|
207
|
|
|
} |
|
208
|
|
|
|
|
209
|
|
|
/** |
|
210
|
|
|
* Extract language from <html lang="en-us"> tag. |
|
211
|
|
|
*/ |
|
212
|
|
|
private function parseHtmlLang(string $html): ?string |
|
213
|
|
|
{ |
|
214
|
|
|
if (preg_match('#<html(?: [^>]+)? lang="([A-Z-]{2,15})"(?: [^>]+)?>#i', $html, $matches)) { |
|
215
|
|
|
return $matches[1]; |
|
216
|
|
|
} |
|
217
|
|
|
|
|
218
|
|
|
return null; |
|
219
|
|
|
} |
|
220
|
|
|
|
|
221
|
|
|
/** |
|
222
|
|
|
* Extract webpage title from HTML <title> |
|
223
|
|
|
* not foolproof : example <!-- <title>bla</title> --> |
|
224
|
|
|
*/ |
|
225
|
|
|
private function parseHtmlTitle(string $html): ?string |
|
226
|
|
|
{ |
|
227
|
|
|
if (preg_match('#<title>([^<]+)</title>#i', $html, $matches)) { |
|
228
|
|
|
return trim(strip_tags($matches[1])); |
|
229
|
|
|
} |
|
230
|
|
|
|
|
231
|
|
|
return null; |
|
232
|
|
|
} |
|
233
|
|
|
|
|
234
|
|
|
/** |
|
235
|
|
|
* Extract first <h1> from HTML. |
|
236
|
|
|
*/ |
|
237
|
|
|
private function parseHtmlFirstH1(string $html): ?string |
|
238
|
|
|
{ |
|
239
|
|
|
if (preg_match('#<h1[^>]*>([^<]+)</h1>#i', $html, $matches)) { |
|
240
|
|
|
return trim(strip_tags($matches[1])); |
|
241
|
|
|
} |
|
242
|
|
|
|
|
243
|
|
|
return null; |
|
244
|
|
|
} |
|
245
|
|
|
|
|
246
|
|
|
/** |
|
247
|
|
|
* TODO strip not unicode characters ? |
|
248
|
|
|
* TODO add initial capital letter ? |
|
249
|
|
|
* This method is used to sanitize subdomain name. |
|
250
|
|
|
* WTF ?!?!?! |
|
251
|
|
|
*/ |
|
252
|
|
|
protected function sanitizeSubDomain(string $subDomain): string |
|
253
|
|
|
{ |
|
254
|
|
|
return str_replace('www.', '', $subDomain); |
|
255
|
|
|
} |
|
256
|
|
|
|
|
257
|
|
|
/** |
|
258
|
|
|
* Extract robots meta tag content. |
|
259
|
|
|
* <meta name="robots" content="noindex,noarchive"> |
|
260
|
|
|
*/ |
|
261
|
|
|
private function getMetaRobotsContent(string $html): string |
|
262
|
|
|
{ |
|
263
|
|
|
if (preg_match('#<meta[^>]+name="robots"[^>]+content="([^"]+)"#i', $html, $matches)) { |
|
264
|
|
|
return $matches[1]; |
|
265
|
|
|
} |
|
266
|
|
|
|
|
267
|
|
|
return ''; |
|
268
|
|
|
} |
|
269
|
|
|
} |
|
270
|
|
|
|
Dispositif /
Wikibot
