|
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
|
|
|
namespace App\Domain\ExternLink; |
|
11
|
|
|
|
|
12
|
|
|
use App\Application\InfrastructurePorts\HttpClientInterface; |
|
13
|
|
|
use App\Application\Utils\HttpUtil; |
|
14
|
|
|
use App\Domain\InfrastructurePorts\InternetDomainParserInterface; |
|
15
|
|
|
use App\Infrastructure\TagParser; |
|
16
|
|
|
use DomainException; |
|
17
|
|
|
use Exception; |
|
18
|
|
|
use Psr\Log\LoggerInterface; |
|
19
|
|
|
use Psr\Log\NullLogger; |
|
20
|
|
|
|
|
21
|
|
|
class ExternPageFactory |
|
22
|
|
|
{ |
|
23
|
|
|
public function __construct(protected HttpClientInterface $client, protected LoggerInterface $log = new NullLogger()) |
|
24
|
|
|
{ |
|
25
|
|
|
} |
|
26
|
|
|
|
|
27
|
|
|
/** |
|
28
|
|
|
* @throws Exception |
|
29
|
|
|
*/ |
|
30
|
|
|
public function fromURL(string $url, InternetDomainParserInterface $domainParser): ExternPage |
|
31
|
|
|
{ |
|
32
|
|
|
if (!HttpUtil::isHttpURL($url)) { |
|
33
|
|
|
throw new Exception('string is not an URL ' . $url); |
|
34
|
|
|
} |
|
35
|
|
|
|
|
36
|
|
|
$html = $this->getHTML($url, true); |
|
37
|
|
|
if (empty($html)) { |
|
38
|
|
|
throw new DomainException('No HTML from requested URL ' . $url); |
|
39
|
|
|
} |
|
40
|
|
|
|
|
41
|
|
|
return new ExternPage($url, $html, new TagParser(), $domainParser, $this->log); |
|
42
|
|
|
} |
|
43
|
|
|
|
|
44
|
|
|
/** |
|
45
|
|
|
* import source from URL with Guzzle. |
|
46
|
|
|
* todo abstract + refac async request |
|
47
|
|
|
*/ |
|
48
|
|
|
public function getHTML(string $url, ?bool $normalized = false): ?string |
|
49
|
|
|
{ |
|
50
|
|
|
// todo : check banned domains ? |
|
51
|
|
|
// todo : check DNS record => ban ? |
|
52
|
|
|
// todo : accept non-ascii URL ? |
|
53
|
|
|
// idn_to_ascii($url); |
|
54
|
|
|
// idn_to_ascii('teßt.com',IDNA_NONTRANSITIONAL_TO_ASCII,INTL_IDNA_VARIANT_UTS46) |
|
55
|
|
|
// checkdnsrr($string, "A") // check DNS record |
|
56
|
|
|
if (!HttpUtil::isHttpURL($url)) { |
|
57
|
|
|
throw new DomainException('URL not compatible : ' . $url); |
|
58
|
|
|
} |
|
59
|
|
|
$response = $this->client->get($url, [ |
|
60
|
|
|
'timeout' => 20, |
|
61
|
|
|
'allow_redirects' => true, /* note : marche pas en mode proxy Tor */ |
|
62
|
|
|
'headers' => ['User-Agent' => getenv('USER_AGENT')], |
|
63
|
|
|
'verify' => false, |
|
64
|
|
|
// 'http_errors' => true, // TRUE: Exception on 4xx 5xx |
|
65
|
|
|
]); |
|
66
|
|
|
|
|
67
|
|
|
if (200 !== $response->getStatusCode()) { |
|
68
|
|
|
$this->log->error('[z49] HTTP error ' . $response->getStatusCode() . ' ' . $response->getReasonPhrase()); |
|
69
|
|
|
|
|
70
|
|
|
return null; |
|
71
|
|
|
} |
|
72
|
|
|
|
|
73
|
|
|
$contentType = $response->getHeader('Content-Type'); |
|
74
|
|
|
if (in_array('application/pdf', explode(';', $contentType[0]))) { |
|
75
|
|
|
$this->log->debug('Incompatible application/pdf content-type'); |
|
76
|
|
|
return null; |
|
77
|
|
|
} |
|
78
|
|
|
|
|
79
|
|
|
$html = $response->getBody()->getContents() ?? ''; |
|
80
|
|
|
|
|
81
|
|
|
return ($normalized) ? HttpUtil::normalizeHtml($html, $url) : $html; |
|
82
|
|
|
} |
|
83
|
|
|
} |
|
84
|
|
|
|