|
1
|
|
|
<?php |
|
2
|
|
|
use Guzzle\Http\Client; |
|
3
|
|
|
|
|
4
|
|
|
/** |
|
5
|
|
|
* Text extractor that calls an Apache Solr instance |
|
6
|
|
|
* and extracts content via the "ExtractingRequestHandler" endpoint. |
|
7
|
|
|
* Does not alter the Solr index itself, but uses it purely |
|
8
|
|
|
* for its file parsing abilities. |
|
9
|
|
|
* |
|
10
|
|
|
* @author ischommer |
|
11
|
|
|
* @see http://wiki.apache.org/solr/ExtractingRequestHandler |
|
12
|
|
|
*/ |
|
13
|
|
|
class SolrCellTextExtractor extends FileTextExtractor |
|
|
|
|
|
|
14
|
|
|
{ |
|
15
|
|
|
/** |
|
16
|
|
|
* Base URL to use for solr text extraction. |
|
17
|
|
|
* E.g. http://localhost:8983/solr/update/extract |
|
18
|
|
|
* |
|
19
|
|
|
* @config |
|
20
|
|
|
* @var string |
|
21
|
|
|
*/ |
|
22
|
|
|
private static $base_url; |
|
|
|
|
|
|
23
|
|
|
|
|
24
|
|
|
private static $priority = 75; |
|
|
|
|
|
|
25
|
|
|
|
|
26
|
|
|
protected $httpClient; |
|
27
|
|
|
|
|
28
|
|
|
public function getHttpClient() |
|
29
|
|
|
{ |
|
30
|
|
|
if (!$this->config()->get('base_url')) { |
|
31
|
|
|
throw new InvalidArgumentException('SolrCellTextExtractor.base_url not specified'); |
|
32
|
|
|
} |
|
33
|
|
|
if (!$this->httpClient) { |
|
34
|
|
|
$this->httpClient = new Client($this->config()->get('base_url')); |
|
35
|
|
|
} |
|
36
|
|
|
return $this->httpClient; |
|
37
|
|
|
} |
|
38
|
|
|
|
|
39
|
|
|
public function setHttpClient($client) |
|
40
|
|
|
{ |
|
41
|
|
|
$this->httpClient = $client; |
|
42
|
|
|
} |
|
43
|
|
|
|
|
44
|
|
|
public function isAvailable() |
|
45
|
|
|
{ |
|
46
|
|
|
$url = $this->config()->get('base_url'); |
|
47
|
|
|
return (boolean) $url; |
|
48
|
|
|
} |
|
49
|
|
|
|
|
50
|
|
|
public function supportsExtension($extension) |
|
51
|
|
|
{ |
|
52
|
|
|
return in_array( |
|
53
|
|
|
strtolower($extension), |
|
54
|
|
|
array( |
|
55
|
|
|
'pdf', 'doc', 'docx', 'xls', 'xlsx', |
|
56
|
|
|
'epub', 'rtf', 'odt', 'fodt', 'ods', 'fods', |
|
57
|
|
|
'ppt', 'pptx', 'odp', 'fodp', 'csv' |
|
58
|
|
|
) |
|
59
|
|
|
); |
|
60
|
|
|
} |
|
61
|
|
|
|
|
62
|
|
|
public function supportsMime($mime) |
|
63
|
|
|
{ |
|
64
|
|
|
// Rely on supportsExtension |
|
65
|
|
|
return false; |
|
66
|
|
|
} |
|
67
|
|
|
|
|
68
|
|
|
public function getContent($path) |
|
69
|
|
|
{ |
|
70
|
|
|
if (!$path) { |
|
71
|
|
|
return ""; |
|
72
|
|
|
} // no file |
|
73
|
|
|
|
|
74
|
|
|
$fileName = basename($path); |
|
75
|
|
|
$client = $this->getHttpClient(); |
|
76
|
|
|
try { |
|
77
|
|
|
$request = $client |
|
78
|
|
|
->post() |
|
79
|
|
|
->addPostFields(array('extractOnly' => 'true', 'extractFormat' => 'text')) |
|
80
|
|
|
->addPostFiles(array('myfile' => $path)); |
|
81
|
|
|
$response = $request->send(); |
|
82
|
|
|
} catch (InvalidArgumentException $e) { |
|
83
|
|
|
SS_Log::log( |
|
84
|
|
|
sprintf( |
|
85
|
|
|
'Error extracting text from "%s" (message: %s)', |
|
86
|
|
|
$path, |
|
87
|
|
|
$e->getMessage() |
|
88
|
|
|
), |
|
89
|
|
|
SS_Log::NOTICE |
|
90
|
|
|
); |
|
91
|
|
|
return null; |
|
92
|
|
|
} |
|
93
|
|
|
// Use preg match to avoid SimpleXML running out of memory on large text nodes |
|
94
|
|
|
preg_match( |
|
95
|
|
|
sprintf('/\<str name\="%s"\>(.*?)\<\/str\>/s', preg_quote($fileName)), |
|
96
|
|
|
(string)$response->getBody(), |
|
97
|
|
|
$matches |
|
98
|
|
|
); |
|
99
|
|
|
|
|
100
|
|
|
return $matches ? $matches[1] : null; |
|
101
|
|
|
} |
|
102
|
|
|
} |
|
103
|
|
|
|
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.