1
|
|
|
<?php |
2
|
|
|
|
3
|
|
|
namespace Spatie\Crawler; |
4
|
|
|
|
5
|
|
|
use Tree\Node\Node; |
6
|
|
|
use GuzzleHttp\Psr7\Uri; |
7
|
|
|
use InvalidArgumentException; |
8
|
|
|
use Psr\Http\Message\UriInterface; |
9
|
|
|
use Symfony\Component\DomCrawler\Link; |
10
|
|
|
use Symfony\Component\DomCrawler\Crawler as DomCrawler; |
11
|
|
|
|
12
|
|
|
class LinkAdder |
13
|
|
|
{ |
14
|
|
|
/** @var \Spatie\Crawler\Crawler */ |
15
|
|
|
protected $crawler; |
16
|
|
|
|
17
|
|
|
public function __construct(Crawler $crawler) |
18
|
|
|
{ |
19
|
|
|
$this->crawler = $crawler; |
20
|
|
|
} |
21
|
|
|
|
22
|
|
|
public function addFromHtml(string $html, UriInterface $foundOnUrl) |
23
|
|
|
{ |
24
|
|
|
$allLinks = $this->extractLinksFromHtml($html, $foundOnUrl); |
25
|
|
|
|
26
|
|
|
collect($allLinks) |
27
|
|
|
->filter(function (UriInterface $url) { |
28
|
|
|
return $this->hasCrawlableScheme($url); |
29
|
|
|
}) |
30
|
|
|
->map(function (UriInterface $url) { |
31
|
|
|
return $this->normalizeUrl($url); |
32
|
|
|
}) |
33
|
|
|
->filter(function (UriInterface $url) use ($foundOnUrl) { |
34
|
|
|
$node = $this->crawler->addToDepthTree($url, $foundOnUrl); |
35
|
|
|
|
36
|
|
|
return $this->shouldCrawl($node); |
37
|
|
|
}) |
38
|
|
|
->filter(function (UriInterface $url) { |
39
|
|
|
return strpos($url->getPath(), '/tel:') === false; |
40
|
|
|
}) |
41
|
|
|
->each(function (UriInterface $url) use ($foundOnUrl) { |
42
|
|
|
if ($this->crawler->maximumCrawlCountReached()) { |
43
|
|
|
return; |
44
|
|
|
} |
45
|
|
|
|
46
|
|
|
$crawlUrl = CrawlUrl::create($url, $foundOnUrl); |
47
|
|
|
|
48
|
|
|
$this->crawler->addToCrawlQueue($crawlUrl); |
49
|
|
|
}); |
50
|
|
|
} |
51
|
|
|
|
52
|
|
|
/** |
53
|
|
|
* @param string $html |
54
|
|
|
* @param \Psr\Http\Message\UriInterface $foundOnUrl |
55
|
|
|
* |
56
|
|
|
* @return \Illuminate\Support\Collection|\Tightenco\Collect\Support\Collection|null |
57
|
|
|
*/ |
58
|
|
|
protected function extractLinksFromHtml(string $html, UriInterface $foundOnUrl) |
59
|
|
|
{ |
60
|
|
|
if ($this->crawler->mayExecuteJavaScript()) { |
61
|
|
|
$html = $this->getBodyAfterExecutingJavaScript($foundOnUrl); |
62
|
|
|
} |
63
|
|
|
|
64
|
|
|
$domCrawler = new DomCrawler($html, $foundOnUrl); |
65
|
|
|
|
66
|
|
|
return collect($domCrawler->filterXpath('//a')->links()) |
67
|
|
|
->reject(function (Link $link) { |
68
|
|
|
return $link->getNode()->getAttribute('rel') === 'nofollow'; |
69
|
|
|
}) |
70
|
|
|
->map(function (Link $link) { |
71
|
|
|
try { |
72
|
|
|
return new Uri($link->getUri()); |
73
|
|
|
} catch (InvalidArgumentException $exception) { |
74
|
|
|
return; |
75
|
|
|
} |
76
|
|
|
}) |
77
|
|
|
->filter(); |
78
|
|
|
} |
79
|
|
|
|
80
|
|
|
protected function hasCrawlableScheme(UriInterface $uri): bool |
81
|
|
|
{ |
82
|
|
|
return in_array($uri->getScheme(), ['http', 'https']); |
83
|
|
|
} |
84
|
|
|
|
85
|
|
|
protected function normalizeUrl(UriInterface $url): UriInterface |
86
|
|
|
{ |
87
|
|
|
return $url->withFragment(''); |
88
|
|
|
} |
89
|
|
|
|
90
|
|
|
protected function shouldCrawl(Node $node): bool |
91
|
|
|
{ |
92
|
|
|
if ($this->crawler->mustRespectRobots()) { |
93
|
|
|
return $this->crawler->getRobotsTxt()->allows($node->getValue()); |
94
|
|
|
} |
95
|
|
|
|
96
|
|
|
$maximumDepth = $this->crawler->getMaximumDepth(); |
97
|
|
|
|
98
|
|
|
if (is_null($maximumDepth)) { |
99
|
|
|
return true; |
100
|
|
|
} |
101
|
|
|
|
102
|
|
|
return $node->getDepth() <= $maximumDepth; |
103
|
|
|
} |
104
|
|
|
|
105
|
|
|
protected function getBodyAfterExecutingJavaScript(UriInterface $foundOnUrl): string |
106
|
|
|
{ |
107
|
|
|
$browsershot = $this->crawler->getBrowsershot(); |
108
|
|
|
|
109
|
|
|
$html = $browsershot->setUrl((string) $foundOnUrl)->bodyHtml(); |
110
|
|
|
|
111
|
|
|
return html_entity_decode($html); |
112
|
|
|
} |
113
|
|
|
} |
114
|
|
|
|