|
1
|
|
|
<?php |
|
2
|
|
|
|
|
3
|
|
|
/** |
|
4
|
|
|
* Text extractor that uses php function strip_tags to get just the text. OK for indexing, not the best for readable text. |
|
5
|
|
|
* @author mstephens |
|
6
|
|
|
* |
|
7
|
|
|
*/ |
|
8
|
|
|
class HTMLTextExtractor extends FileTextExtractor |
|
|
|
|
|
|
9
|
|
|
{ |
|
10
|
|
|
public function isAvailable() |
|
11
|
|
|
{ |
|
12
|
|
|
return true; |
|
13
|
|
|
} |
|
14
|
|
|
|
|
15
|
|
|
public function supportsExtension($extension) |
|
16
|
|
|
{ |
|
17
|
|
|
return in_array( |
|
18
|
|
|
strtolower($extension), |
|
19
|
|
|
array("html", "htm", "xhtml") |
|
20
|
|
|
); |
|
21
|
|
|
} |
|
22
|
|
|
|
|
23
|
|
|
public function supportsMime($mime) |
|
24
|
|
|
{ |
|
25
|
|
|
return strtolower($mime) === 'text/html'; |
|
26
|
|
|
} |
|
27
|
|
|
|
|
28
|
|
|
/** |
|
29
|
|
|
* Lower priority because its not the most clever HTML extraction. If there is something better, use it |
|
30
|
|
|
* |
|
31
|
|
|
* @config |
|
32
|
|
|
* @var integer |
|
33
|
|
|
*/ |
|
34
|
|
|
private static $priority = 10; |
|
|
|
|
|
|
35
|
|
|
|
|
36
|
|
|
/** |
|
37
|
|
|
* Extracts content from regex, by using strip_tags() |
|
38
|
|
|
* combined with regular expressions to remove non-content tags like <style> or <script>, |
|
39
|
|
|
* as well as adding line breaks after block tags. |
|
40
|
|
|
* |
|
41
|
|
|
* @param string $path |
|
42
|
|
|
* @return string |
|
43
|
|
|
*/ |
|
44
|
|
|
public function getContent($path) |
|
45
|
|
|
{ |
|
46
|
|
|
$content = file_get_contents($path); |
|
47
|
|
|
// Yes, yes, regex'ing HTML is evil. |
|
48
|
|
|
// Since we don't care about well-formedness or markup here, it does the job. |
|
49
|
|
|
$content = preg_replace( |
|
50
|
|
|
array( |
|
51
|
|
|
// Remove invisible content |
|
52
|
|
|
'@<head[^>]*?>.*?</head>@siu', |
|
53
|
|
|
'@<style[^>]*?>.*?</style>@siu', |
|
54
|
|
|
'@<script[^>]*?.*?</script>@siu', |
|
55
|
|
|
'@<object[^>]*?.*?</object>@siu', |
|
56
|
|
|
'@<embed[^>]*?.*?</embed>@siu', |
|
57
|
|
|
'@<applet[^>]*?.*?</applet>@siu', |
|
58
|
|
|
'@<noframes[^>]*?.*?</noframes>@siu', |
|
59
|
|
|
'@<noscript[^>]*?.*?</noscript>@siu', |
|
60
|
|
|
'@<noembed[^>]*?.*?</noembed>@siu', |
|
61
|
|
|
// Add line breaks before and after blocks |
|
62
|
|
|
'@</?((address)|(blockquote)|(center)|(del))@iu', |
|
63
|
|
|
'@</?((div)|(h[1-9])|(ins)|(isindex)|(p)|(pre))@iu', |
|
64
|
|
|
'@</?((dir)|(dl)|(dt)|(dd)|(li)|(menu)|(ol)|(ul))@iu', |
|
65
|
|
|
'@</?((table)|(th)|(td)|(caption))@iu', |
|
66
|
|
|
'@</?((form)|(button)|(fieldset)|(legend)|(input))@iu', |
|
67
|
|
|
'@</?((label)|(select)|(optgroup)|(option)|(textarea))@iu', |
|
68
|
|
|
'@</?((frameset)|(frame)|(iframe))@iu', |
|
69
|
|
|
), |
|
70
|
|
|
array( |
|
71
|
|
|
' ', ' ', ' ', ' ', ' ', ' ', ' ', ' ', ' ', "$0", "$0", "$0", "$0", "$0", "$0", "$0", "$0", |
|
72
|
|
|
), |
|
73
|
|
|
$content |
|
74
|
|
|
); |
|
75
|
|
|
return strip_tags($content); |
|
76
|
|
|
} |
|
77
|
|
|
} |
|
78
|
|
|
|
silverstripe /
silverstripe-textextraction
Loading history...
You can fix this by adding a namespace to your class:
When choosing a vendor namespace, try to pick something that is not too generic to avoid conflicts with other libraries.