1
|
|
|
<?php |
2
|
|
|
|
3
|
|
|
namespace Colligator; |
4
|
|
|
|
5
|
|
|
use GuzzleHttp\Exception\TransferException; |
6
|
|
|
|
7
|
|
|
class DescriptionScraper |
8
|
|
|
{ |
9
|
|
|
public $doc; |
10
|
|
|
public $url; |
11
|
|
|
public $client; |
12
|
|
|
protected $scrapers; |
13
|
|
|
public $sleepTime = 7; |
14
|
|
|
|
15
|
|
|
/** |
16
|
|
|
* Create a new scraper. |
17
|
|
|
* |
18
|
|
|
* @param Client $client |
19
|
|
|
*/ |
20
|
|
|
public function __construct() |
21
|
|
|
{ |
22
|
|
|
$this->register([ |
23
|
|
|
Scrapers\BsScraper::class, |
24
|
|
|
Scrapers\LocScraper::class, |
25
|
|
|
Scrapers\FluxScraper::class, |
26
|
|
|
Scrapers\UnivScraper::class, |
27
|
|
|
]); |
28
|
|
|
} |
29
|
|
|
|
30
|
|
|
public function register($scrapers) |
31
|
|
|
{ |
32
|
|
|
$this->scrapers = []; |
33
|
|
|
foreach ($scrapers as $scraper) { |
34
|
|
|
$this->scrapers[] = new $scraper(); // We could do dependency injection here |
35
|
|
|
} |
36
|
|
|
} |
37
|
|
|
|
38
|
|
|
public function notify($msg, $doc, $args, $level = 'warning') |
39
|
|
|
{ |
40
|
|
|
$msg = vsprintf($msg, $args); |
41
|
|
|
$docLink = sprintf('<http://colligator.biblionaut.net/api/documents/%s|#%s> ', $doc->id, $doc->id); |
42
|
|
|
\Slack::attach([ |
43
|
|
|
'fallback' => '#' . $doc->id . ' ' . $msg, |
44
|
|
|
'text' => $docLink . $msg, |
45
|
|
|
'color' => $level, |
46
|
|
|
])->send(); |
47
|
|
|
} |
48
|
|
|
|
49
|
|
|
public function scrape($url) |
50
|
|
|
{ |
51
|
|
|
foreach ($this->scrapers as $scraper) { |
52
|
|
|
if ($scraper->recognizes($url)) { |
53
|
|
|
return $scraper->scrape($scraper->getCrawler($url)); |
54
|
|
|
} |
55
|
|
|
} |
56
|
|
|
|
57
|
|
|
return; |
58
|
|
|
} |
59
|
|
|
|
60
|
|
|
/** |
61
|
|
|
* Execute the job. |
62
|
|
|
* |
63
|
|
|
* @param Document $doc |
64
|
|
|
* @param string $url |
65
|
|
|
*/ |
66
|
|
|
public function updateDocument(Document $doc, $url) |
67
|
|
|
{ |
68
|
|
|
\Log::debug('[DescriptionScraper] Looking for decription for ' . $doc->id . ' at ' . $url); |
69
|
|
|
|
70
|
|
|
if (preg_match('/(damm.no)/', $url)) { |
71
|
|
|
\Log::debug('[DescriptionScraper] Ignoring URL: ' . $url); |
72
|
|
|
|
73
|
|
|
return; |
74
|
|
|
} |
75
|
|
|
|
76
|
|
|
try { |
77
|
|
|
$result = $this->scrape($url); |
78
|
|
|
} catch (TransferException $e) { |
79
|
|
|
\Log::error('[DescriptionScraper] Transfer failed: ' . $e->getMessage()); |
80
|
|
|
$this->notify('*DescriptionScraper* failed to fetch: %s', $doc, [$url]); |
81
|
|
|
|
82
|
|
|
return; |
83
|
|
|
} catch (Scrapers\ScrapeException $e) { |
84
|
|
|
\Log::error('[DescriptionScraper] Scraping of ' . $url . ' failed: ' . $e->getMessage()); |
85
|
|
|
$this->notify('*DescriptionScraper* / %s failed to find a text at: %s', $doc, [$e->getMessage(), $url]); |
86
|
|
|
|
87
|
|
|
return; |
88
|
|
|
} |
89
|
|
|
if (is_null($result)) { |
90
|
|
|
\Log::error('Encountered URL not recognized by any scraper: ' . $url); |
91
|
|
|
$this->notify('*DescriptionScraper* encountered URL not recognized by any sraper: %s', $doc, [$url]); |
92
|
|
|
|
93
|
|
|
return; |
94
|
|
|
} |
95
|
|
|
|
96
|
|
|
$doc->description = [ |
97
|
|
|
'text' => $result['text'], |
98
|
|
|
'source' => $result['source'], |
99
|
|
|
'source_url' => $url, |
100
|
|
|
]; |
101
|
|
|
|
102
|
|
|
sleep($this->sleepTime); |
103
|
|
|
} |
104
|
|
|
} |
105
|
|
|
|