|
1
|
|
|
<?php |
|
2
|
|
|
|
|
3
|
|
|
declare(strict_types=1); |
|
4
|
|
|
|
|
5
|
|
|
/** |
|
6
|
|
|
* This file is part of Scout Extended. |
|
7
|
|
|
* |
|
8
|
|
|
* (c) Algolia Team <[email protected]> |
|
9
|
|
|
* |
|
10
|
|
|
* For the full copyright and license information, please view the LICENSE |
|
11
|
|
|
* file that was distributed with this source code. |
|
12
|
|
|
*/ |
|
13
|
|
|
|
|
14
|
|
|
namespace Algolia\ScoutExtended\Splitters; |
|
15
|
|
|
|
|
16
|
|
|
use DOMXPath; |
|
17
|
|
|
use DOMDocument; |
|
18
|
|
|
use Algolia\ScoutExtended\Contracts\SplitterContract; |
|
19
|
|
|
use Algolia\ScoutExtended\Splitters\HtmlSplitterComponent\Queue; |
|
20
|
|
|
use Algolia\ScoutExtended\Splitters\HtmlSplitterComponent\ObjectQueue; |
|
21
|
|
|
|
|
22
|
|
|
final class HtmlSplitter implements SplitterContract |
|
23
|
|
|
{ |
|
24
|
|
|
/** |
|
25
|
|
|
* The list of html tags. |
|
26
|
|
|
* |
|
27
|
|
|
* @var string[] |
|
28
|
|
|
*/ |
|
29
|
|
|
protected $tags = [ |
|
30
|
|
|
'h1', |
|
31
|
|
|
'h2', |
|
32
|
|
|
'h3', |
|
33
|
|
|
'h4', |
|
34
|
|
|
'h5', |
|
35
|
|
|
'h6', |
|
36
|
|
|
'p', |
|
37
|
|
|
]; |
|
38
|
|
|
|
|
39
|
|
|
/** |
|
40
|
|
|
* Creates a new instance of the class. |
|
41
|
|
|
* |
|
42
|
|
|
* @param array $nodes |
|
43
|
|
|
* |
|
44
|
|
|
* @return void |
|
45
|
|
|
*/ |
|
46
|
5 |
|
public function __construct(array $tags = null) |
|
47
|
|
|
{ |
|
48
|
5 |
|
if ($tags !== null) { |
|
49
|
2 |
|
$this->tags = $tags; |
|
50
|
|
|
} |
|
51
|
5 |
|
} |
|
52
|
|
|
|
|
53
|
|
|
/** |
|
54
|
|
|
* Acts a static factory. |
|
55
|
|
|
* |
|
56
|
|
|
* @param string|array<string> $tags |
|
57
|
|
|
* |
|
58
|
|
|
* @return static |
|
59
|
|
|
*/ |
|
60
|
2 |
|
public static function by($tags) |
|
61
|
|
|
{ |
|
62
|
2 |
|
return new static((array) $tags); |
|
63
|
|
|
} |
|
64
|
|
|
|
|
65
|
|
|
/** |
|
66
|
|
|
* Splits the given value. |
|
67
|
|
|
* |
|
68
|
|
|
* @param object $searchable |
|
69
|
|
|
* @param string $value |
|
70
|
|
|
* |
|
71
|
|
|
* @return array |
|
72
|
|
|
*/ |
|
73
|
5 |
|
public function split($searchable, $value): array |
|
74
|
|
|
{ |
|
75
|
5 |
|
$dom = new DOMDocument(); |
|
76
|
|
|
//DOMDocument is only for HTML4, this exception is too avoid errors from HTML5 |
|
77
|
|
|
try { |
|
78
|
5 |
|
$dom->loadHTML($value); |
|
79
|
2 |
|
} catch (\ErrorException $exception) { |
|
|
|
|
|
|
80
|
|
|
} |
|
81
|
5 |
|
$xpath = new DOMXpath($dom); |
|
82
|
5 |
|
$queue = new Queue(); |
|
83
|
5 |
|
$xpathQuery = '//'.implode(' | //', $this->tags); |
|
84
|
5 |
|
$tags = $xpath->query($xpathQuery); |
|
85
|
|
|
|
|
86
|
5 |
|
foreach ($tags as $node) { |
|
87
|
5 |
|
$objectQueue = new ObjectQueue($node->nodeName, $node->textContent); |
|
88
|
5 |
|
$queue->addObjectQueue($objectQueue); |
|
89
|
|
|
} |
|
90
|
|
|
|
|
91
|
5 |
|
return $queue->sanitizeQueue(); |
|
92
|
|
|
} |
|
93
|
|
|
} |
|
94
|
|
|
|
algolia /
scout-extended
Loading history...