|
1
|
|
|
<?php |
|
2
|
|
|
|
|
3
|
|
|
namespace Pilipinews\Website\Gma; |
|
4
|
|
|
|
|
5
|
|
|
use Pilipinews\Common\Article; |
|
6
|
|
|
use Pilipinews\Common\Client; |
|
7
|
|
|
use Pilipinews\Common\Converter; |
|
8
|
|
|
use Pilipinews\Common\Crawler as DomCrawler; |
|
9
|
|
|
use Pilipinews\Common\Interfaces\ScraperInterface; |
|
10
|
|
|
use Pilipinews\Common\Scraper as AbstractScraper; |
|
11
|
|
|
|
|
12
|
|
|
/** |
|
13
|
|
|
* GMA News Scraper |
|
14
|
|
|
* |
|
15
|
|
|
* @package Pilipinews |
|
16
|
|
|
* @author Rougin Gutib <[email protected]> |
|
17
|
|
|
*/ |
|
18
|
|
|
class Scraper extends AbstractScraper implements ScraperInterface |
|
19
|
|
|
{ |
|
20
|
|
|
/** |
|
21
|
|
|
* Returns the contents of an article. |
|
22
|
|
|
* |
|
23
|
|
|
* @param string $link |
|
24
|
|
|
* @return \Pilipinews\Common\Article |
|
25
|
|
|
*/ |
|
26
|
9 |
|
public function scrape($link) |
|
27
|
|
|
{ |
|
28
|
9 |
|
$this->prepare(mb_strtolower($link)); |
|
29
|
|
|
|
|
30
|
9 |
|
$title = $this->json['headline']; |
|
31
|
|
|
|
|
32
|
9 |
|
$title = str_replace(' | News |', '', $title); |
|
33
|
|
|
|
|
34
|
9 |
|
$converter = new Converter; |
|
35
|
|
|
|
|
36
|
9 |
|
$title = $converter->convert($title); |
|
37
|
|
|
|
|
38
|
9 |
|
$body = $this->tweet($this->crawler); |
|
39
|
|
|
|
|
40
|
9 |
|
$html = (string) $this->html($body); |
|
41
|
|
|
|
|
42
|
9 |
|
return new Article($title, $html, $link); |
|
43
|
1 |
|
} |
|
44
|
|
|
|
|
45
|
|
|
/** |
|
46
|
|
|
* Initializes the crawler instance. |
|
47
|
|
|
* |
|
48
|
|
|
* @param string $link |
|
49
|
|
|
* @return void |
|
50
|
|
|
*/ |
|
51
|
9 |
|
protected function prepare($link) |
|
52
|
1 |
|
{ |
|
53
|
9 |
|
$response = (string) Client::request((string) $link); |
|
54
|
|
|
|
|
55
|
9 |
|
$html = trim(preg_replace('/\s+/', ' ', $response)); |
|
56
|
|
|
|
|
57
|
9 |
|
$html = str_replace('<p> <strong>', '<p><strong>', $html); |
|
58
|
|
|
|
|
59
|
9 |
|
$html = str_replace('<br /> ', '<br />', $html); |
|
60
|
|
|
|
|
61
|
9 |
|
preg_match('/<script type="application\/ld\+json"\>(.*?)<\/script\>/i', $html, $match); |
|
62
|
|
|
|
|
63
|
9 |
|
$this->json = json_decode($match[1], true); |
|
|
|
|
|
|
64
|
|
|
|
|
65
|
9 |
|
$content = (string) $this->json['articleBody']; |
|
66
|
|
|
|
|
67
|
9 |
|
$this->crawler = new DomCrawler((string) $content); |
|
68
|
9 |
|
} |
|
69
|
|
|
} |
|
70
|
|
|
|
pilipinews /
gma
Loading history...