1
|
|
|
<?php |
2
|
|
|
/** |
3
|
|
|
* Class URLExtractor |
4
|
|
|
* |
5
|
|
|
* @created 15.08.2019 |
6
|
|
|
* @author smiley <[email protected]> |
7
|
|
|
* @copyright 2019 smiley |
8
|
|
|
* @license MIT |
9
|
|
|
*/ |
10
|
|
|
|
11
|
|
|
namespace chillerlan\HTTP\Psr18; |
12
|
|
|
|
13
|
|
|
use chillerlan\HTTP\Psr17\RequestFactory; |
14
|
|
|
use Psr\Http\Client\ClientInterface; |
15
|
|
|
use Psr\Http\Message\{RequestFactoryInterface, RequestInterface, ResponseInterface, UriInterface}; |
16
|
|
|
|
17
|
|
|
use function array_key_last; |
18
|
|
|
use function array_pop; |
19
|
|
|
use function count; |
20
|
|
|
use function in_array; |
21
|
|
|
|
22
|
|
|
/** |
23
|
|
|
* A client that follows redirects until it reaches a non-30x response, e.g. to extract shortened URLs |
24
|
|
|
* |
25
|
|
|
* The given HTTP client needs to be set up accordingly: |
26
|
|
|
* |
27
|
|
|
* - CURLOPT_FOLLOWLOCATION must be set to false so that we can intercept the 30x responses |
28
|
|
|
* - CURLOPT_MAXREDIRS should be set to a value > 1 |
29
|
|
|
*/ |
30
|
|
|
class URLExtractor implements ClientInterface{ |
31
|
|
|
|
32
|
|
|
/** @var \Psr\Http\Message\ResponseInterface[] */ |
33
|
|
|
protected array $responses = []; |
34
|
|
|
|
35
|
|
|
protected ClientInterface $http; |
36
|
|
|
|
37
|
|
|
protected RequestFactoryInterface $requestFactory; |
38
|
|
|
|
39
|
|
|
/** |
40
|
|
|
* URLExtractor constructor. |
41
|
|
|
*/ |
42
|
|
|
public function __construct(ClientInterface $http, RequestFactoryInterface $requestFactory = null){ |
43
|
|
|
$this->http = $http; |
44
|
|
|
$this->requestFactory = $requestFactory ?? new RequestFactory; |
45
|
|
|
} |
46
|
|
|
|
47
|
|
|
/** |
48
|
|
|
* @inheritDoc |
49
|
|
|
*/ |
50
|
|
|
public function sendRequest(RequestInterface $request):ResponseInterface{ |
51
|
|
|
|
52
|
|
|
do{ |
53
|
|
|
// fetch the response for the current request |
54
|
|
|
$response = $this->http->sendRequest($request); |
55
|
|
|
$this->responses[] = $response; |
56
|
|
|
// set up a new request to the location header of the last response |
57
|
|
|
$request = $this->requestFactory->createRequest($request->getMethod(), $response->getHeaderLine('location')); |
58
|
|
|
} |
59
|
|
|
while(in_array($response->getStatusCode(), [301, 302, 303, 307, 308], true)); |
60
|
|
|
|
61
|
|
|
return $response; |
62
|
|
|
} |
63
|
|
|
|
64
|
|
|
/** |
65
|
|
|
* extract the given URL and return the last valid location header |
66
|
|
|
*/ |
67
|
|
|
public function extract(UriInterface|string $shortURL):?string{ |
68
|
|
|
$request = $this->requestFactory->createRequest('GET', $shortURL); |
69
|
|
|
$response = $this->sendRequest($request); |
70
|
|
|
|
71
|
|
|
if($response->getStatusCode() !== 200 || empty($this->responses)){ |
72
|
|
|
return null; |
73
|
|
|
} |
74
|
|
|
|
75
|
|
|
$count = count($this->responses) - 2; |
76
|
|
|
|
77
|
|
|
while($count >= 0){ |
78
|
|
|
|
79
|
|
|
$url = $this->responses[$count]?->getHeaderLine('location'); |
80
|
|
|
|
81
|
|
|
if(!empty($url)){ |
82
|
|
|
return $url; |
83
|
|
|
} |
84
|
|
|
|
85
|
|
|
$count--; |
86
|
|
|
} |
87
|
|
|
|
88
|
|
|
return null; |
89
|
|
|
} |
90
|
|
|
|
91
|
|
|
/** |
92
|
|
|
* @return \Psr\Http\Message\ResponseInterface[] |
93
|
|
|
*/ |
94
|
|
|
public function getResponses():array{ |
95
|
|
|
return $this->responses; |
96
|
|
|
} |
97
|
|
|
|
98
|
|
|
} |
99
|
|
|
|