1
|
|
|
<?php |
2
|
|
|
namespace ApacheSolrForTypo3\Solr; |
3
|
|
|
|
4
|
|
|
/*************************************************************** |
5
|
|
|
* Copyright notice |
6
|
|
|
* |
7
|
|
|
* (c) 2009-2015 Ingo Renner <[email protected]> |
8
|
|
|
* All rights reserved |
9
|
|
|
* |
10
|
|
|
* This script is part of the TYPO3 project. The TYPO3 project is |
11
|
|
|
* free software; you can redistribute it and/or modify |
12
|
|
|
* it under the terms of the GNU General Public License as published by |
13
|
|
|
* the Free Software Foundation; either version 2 of the License, or |
14
|
|
|
* (at your option) any later version. |
15
|
|
|
* |
16
|
|
|
* The GNU General Public License can be found at |
17
|
|
|
* http://www.gnu.org/copyleft/gpl.html. |
18
|
|
|
* |
19
|
|
|
* This script is distributed in the hope that it will be useful, |
20
|
|
|
* but WITHOUT ANY WARRANTY; without even the implied warranty of |
21
|
|
|
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the |
22
|
|
|
* GNU General Public License for more details. |
23
|
|
|
* |
24
|
|
|
* This copyright notice MUST APPEAR in all copies of the script! |
25
|
|
|
***************************************************************/ |
26
|
|
|
|
27
|
|
|
use TYPO3\CMS\Core\Utility\GeneralUtility; |
28
|
|
|
|
29
|
|
|
/** |
30
|
|
|
* Solr Service Access |
31
|
|
|
* |
32
|
|
|
* @author Ingo Renner <[email protected]> |
33
|
|
|
*/ |
34
|
|
|
class SolrService extends \Apache_Solr_Service |
35
|
|
|
{ |
36
|
|
|
const LUKE_SERVLET = 'admin/luke'; |
37
|
|
|
const SYSTEM_SERVLET = 'admin/system'; |
38
|
|
|
const PLUGINS_SERVLET = 'admin/plugins'; |
39
|
|
|
const CORES_SERVLET = 'admin/cores'; |
40
|
|
|
const SCHEMA_SERVLET = 'schema'; |
41
|
|
|
const SYNONYMS_SERVLET = 'schema/analysis/synonyms/'; |
42
|
|
|
const STOPWORDS_SERVLET = 'schema/analysis/stopwords/'; |
43
|
|
|
|
44
|
|
|
const SCHEME_HTTP = 'http'; |
45
|
|
|
const SCHEME_HTTPS = 'https'; |
46
|
|
|
|
47
|
|
|
/** |
48
|
|
|
* Server connection scheme. http or https. |
49
|
|
|
* |
50
|
|
|
* @var string |
51
|
|
|
*/ |
52
|
|
|
protected $_scheme = self::SCHEME_HTTP; |
53
|
|
|
|
54
|
|
|
/** |
55
|
|
|
* Constructed servlet URL for Luke |
56
|
|
|
* |
57
|
|
|
* @var string |
58
|
|
|
*/ |
59
|
|
|
protected $_lukeUrl; |
60
|
|
|
|
61
|
|
|
/** |
62
|
|
|
* Constructed servlet URL for plugin information |
63
|
|
|
* |
64
|
|
|
* @var string |
65
|
|
|
*/ |
66
|
|
|
protected $_pluginsUrl; |
67
|
|
|
|
68
|
|
|
protected $_coresUrl; |
69
|
|
|
|
70
|
|
|
protected $_extractUrl; |
71
|
|
|
|
72
|
|
|
protected $_synonymsUrl; |
73
|
|
|
|
74
|
|
|
protected $_stopWordsUrl; |
75
|
|
|
|
76
|
|
|
protected $_schemaUrl; |
77
|
|
|
|
78
|
|
|
protected $debug; |
79
|
|
|
|
80
|
|
|
/** |
81
|
|
|
* @var \Apache_Solr_Response |
82
|
|
|
*/ |
83
|
|
|
protected $responseCache = null; |
84
|
|
|
protected $hasSearched = false; |
85
|
|
|
|
86
|
|
|
protected $lukeData = array(); |
87
|
|
|
protected $systemData = null; |
88
|
|
|
protected $pluginsData = null; |
89
|
|
|
|
90
|
|
|
protected $schemaName = null; |
91
|
|
|
protected $solrconfigName = null; |
92
|
|
|
|
93
|
|
|
/** |
94
|
|
|
* @var array |
95
|
|
|
*/ |
96
|
|
|
protected $configuration; |
97
|
|
|
|
98
|
|
|
/** |
99
|
|
|
* Constructor |
100
|
|
|
* |
101
|
|
|
* @param string $host Solr host |
102
|
|
|
* @param string $port Solr port |
103
|
|
|
* @param string $path Solr path |
104
|
|
|
* @param string $scheme Scheme, defaults to http, can be https |
105
|
|
|
*/ |
106
|
54 |
|
public function __construct( |
107
|
|
|
$host = '', |
108
|
|
|
$port = '8983', |
109
|
|
|
$path = '/solr/', |
110
|
|
|
$scheme = 'http' |
111
|
|
|
) { |
112
|
54 |
|
$this->setScheme($scheme); |
113
|
53 |
|
$this->configuration = Util::getSolrConfiguration(); |
|
|
|
|
114
|
|
|
|
115
|
53 |
|
parent::__construct($host, $port, $path); |
116
|
53 |
|
} |
117
|
|
|
|
118
|
|
|
/** |
119
|
|
|
* Creates a string representation of the Solr connection. Specifically |
120
|
|
|
* will return the Solr URL. |
121
|
|
|
* |
122
|
|
|
* @return string The Solr URL. |
123
|
|
|
*/ |
124
|
4 |
|
public function __toString() |
125
|
|
|
{ |
126
|
4 |
|
return $this->_scheme . '://' . $this->_host . ':' . $this->_port . $this->_path; |
127
|
|
|
} |
128
|
|
|
|
129
|
|
|
/** |
130
|
|
|
* Returns the current time in milliseconds. |
131
|
|
|
* |
132
|
|
|
* @return int |
133
|
|
|
*/ |
134
|
|
|
protected function getMilliseconds() |
135
|
|
|
{ |
136
|
|
|
return GeneralUtility::milliseconds(); |
137
|
|
|
} |
138
|
|
|
|
139
|
|
|
/** |
140
|
|
|
* Performs a search. |
141
|
|
|
* |
142
|
|
|
* @param string $query query string / search term |
143
|
|
|
* @param int $offset result offset for pagination |
144
|
|
|
* @param int $limit number of results to retrieve |
145
|
|
|
* @param array $params additional HTTP GET parameters |
146
|
|
|
* @param string $method The HTTP method (Apache_Solr_Service::METHOD_GET or Apache_Solr_Service::METHOD::POST) |
147
|
|
|
* @return \Apache_Solr_Response Solr response |
148
|
|
|
* @throws \RuntimeException if Solr returns a HTTP status code other than 200 |
149
|
|
|
*/ |
150
|
25 |
|
public function search($query, $offset = 0, $limit = 10, $params = array(), $method = self::METHOD_GET) |
151
|
|
|
{ |
152
|
25 |
|
$response = parent::search($query, $offset, $limit, $params, $method); |
153
|
25 |
|
$this->hasSearched = true; |
154
|
|
|
|
155
|
25 |
|
$this->responseCache = $response; |
156
|
|
|
|
157
|
25 |
|
if ($response->getHttpStatus() != 200) { |
158
|
|
|
throw new \RuntimeException( |
159
|
|
|
'Invalid query. Solr returned an error: ' |
160
|
|
|
. $response->getHttpStatus() . ' ' |
161
|
|
|
. $response->getHttpStatusMessage(), |
162
|
|
|
1293109870 |
163
|
|
|
); |
164
|
|
|
} |
165
|
|
|
|
166
|
25 |
|
return $response; |
167
|
|
|
} |
168
|
|
|
|
169
|
|
|
/** |
170
|
|
|
* Call the /admin/ping servlet, can be used to quickly tell if a connection to the |
171
|
|
|
* server is available. |
172
|
|
|
* |
173
|
|
|
* Simply overrides the SolrPhpClient implementation, changing ping from a |
174
|
|
|
* HEAD to a GET request, see http://forge.typo3.org/issues/44167 |
175
|
|
|
* |
176
|
|
|
* Also does not report the time, see https://forge.typo3.org/issues/64551 |
177
|
|
|
* |
178
|
|
|
* @param float|int $timeout maximum time to wait for ping in seconds, -1 for unlimited (default is 2) |
179
|
|
|
* @return bool TRUE if Solr can be reached, FALSE if not |
180
|
|
|
*/ |
181
|
35 |
|
public function ping($timeout = 2) |
182
|
|
|
{ |
183
|
35 |
|
$httpResponse = $this->performPingRequest($timeout); |
184
|
35 |
|
return ($httpResponse->getStatusCode() === 200); |
185
|
|
|
} |
186
|
|
|
|
187
|
|
|
/** |
188
|
|
|
* Call the /admin/ping servlet, can be used to get the runtime of a ping request. |
189
|
|
|
* |
190
|
|
|
* @param float|int $timeout maximum time to wait for ping in seconds, -1 for unlimited (default is 2) |
191
|
|
|
* @return int runtime in milliseconds |
192
|
|
|
* @throws \ApacheSolrForTypo3\Solr\PingFailedException |
193
|
|
|
*/ |
194
|
|
|
public function getPingRoundTripRuntime($timeout = 2) |
195
|
|
|
{ |
196
|
|
|
$start = $this->getMilliseconds(); |
197
|
|
|
$httpResponse = $this->performPingRequest($timeout); |
198
|
|
|
$end = $this->getMilliseconds(); |
199
|
|
|
|
200
|
|
|
if ($httpResponse->getStatusCode() !== 200) { |
201
|
|
|
$message = 'Solr ping failed with unexpected response code: ' . $httpResponse->getStatusCode(); |
202
|
|
|
/** @var $exception \ApacheSolrForTypo3\Solr\PingFailedException */ |
203
|
|
|
$exception = GeneralUtility::makeInstance('ApacheSolrForTypo3\Solr\PingFailedException', $message); |
204
|
|
|
$exception->setHttpResponse($httpResponse); |
205
|
|
|
throw $exception; |
206
|
|
|
} |
207
|
|
|
|
208
|
|
|
return $end - $start; |
209
|
|
|
} |
210
|
|
|
|
211
|
|
|
/** |
212
|
|
|
* Performs a ping request and returns the result. |
213
|
|
|
* |
214
|
|
|
* @param int $timeout |
215
|
|
|
* @return \Apache_Solr_HttpTransport_Response |
216
|
|
|
*/ |
217
|
35 |
|
protected function performPingRequest($timeout = 2) |
218
|
|
|
{ |
219
|
35 |
|
return $this->getHttpTransport()->performGetRequest($this->_pingUrl, $timeout); |
220
|
|
|
} |
221
|
|
|
|
222
|
|
|
/** |
223
|
|
|
* Performs a content and meta data extraction request. |
224
|
|
|
* |
225
|
|
|
* @param ExtractingQuery $query An extraction query |
226
|
|
|
* @return array An array containing the extracted content [0] and meta data [1] |
227
|
|
|
*/ |
228
|
1 |
|
public function extractByQuery(ExtractingQuery $query) |
229
|
|
|
{ |
230
|
|
|
$headers = array( |
231
|
1 |
|
'Content-Type' => 'multipart/form-data; boundary=' . $query->getMultiPartPostDataBoundary() |
232
|
1 |
|
); |
233
|
|
|
|
234
|
|
|
try { |
235
|
1 |
|
$response = $this->requestServlet( |
236
|
1 |
|
self::EXTRACT_SERVLET, |
237
|
1 |
|
$query->getQueryParameters(), |
238
|
1 |
|
'POST', |
239
|
1 |
|
$headers, |
240
|
1 |
|
$query->getRawPostFileData() |
241
|
1 |
|
); |
242
|
1 |
|
} catch (\Exception $e) { |
243
|
|
|
GeneralUtility::devLog('Extracting text and meta data through Solr Cell over HTTP POST', |
244
|
|
|
'solr', 3, array( |
245
|
|
|
'query' => (array)$query, |
246
|
|
|
'parameters' => $query->getQueryParameters(), |
247
|
|
|
'file' => $query->getFile(), |
248
|
|
|
'headers' => $headers, |
249
|
|
|
'query url' => self::EXTRACT_SERVLET, |
250
|
|
|
'exception' => $e->getMessage() |
251
|
|
|
)); |
252
|
|
|
} |
253
|
|
|
|
254
|
|
|
return array( |
255
|
1 |
|
$response->extracted, |
256
|
1 |
|
(array)$response->extracted_metadata |
257
|
1 |
|
); |
258
|
|
|
} |
259
|
|
|
|
260
|
|
|
/** |
261
|
|
|
* Make a request to a servlet (a path) that's not a standard path. |
262
|
|
|
* |
263
|
|
|
* @param string $servlet Path to be added to the base Solr path. |
264
|
|
|
* @param array $parameters Optional, additional request parameters when constructing the URL. |
265
|
|
|
* @param string $method HTTP method to use, defaults to GET. |
266
|
|
|
* @param array $requestHeaders Key value pairs of header names and values. Should include 'Content-Type' for POST and PUT. |
267
|
|
|
* @param string $rawPost Must be an empty string unless method is POST or PUT. |
268
|
|
|
* @param float|bool $timeout Read timeout in seconds, defaults to FALSE. |
269
|
|
|
* @return \Apache_Solr_Response Response object |
270
|
|
|
* @throws \Apache_Solr_HttpTransportException if returned HTTP status is other than 200 |
271
|
|
|
*/ |
272
|
1 |
|
public function requestServlet( |
273
|
|
|
$servlet, |
274
|
|
|
$parameters = array(), |
275
|
|
|
$method = 'GET', |
276
|
|
|
$requestHeaders = array(), |
277
|
|
|
$rawPost = '', |
278
|
|
|
$timeout = false |
279
|
|
|
) { |
280
|
1 |
|
$httpTransport = $this->getHttpTransport(); |
281
|
1 |
|
$solrResponse = null; |
|
|
|
|
282
|
|
|
|
283
|
1 |
|
if ($method == 'GET' || $method == 'HEAD') { |
284
|
|
|
// Make sure we are not sending a request body. |
285
|
|
|
$rawPost = ''; |
286
|
|
|
} |
287
|
|
|
|
288
|
|
|
// Add default parameters |
289
|
1 |
|
$parameters['wt'] = self::SOLR_WRITER; |
290
|
1 |
|
$parameters['json.nl'] = $this->_namedListTreatment; |
291
|
1 |
|
$url = $this->_constructUrl($servlet, $parameters); |
292
|
|
|
|
293
|
1 |
|
if ($method == self::METHOD_GET) { |
294
|
|
|
$httpResponse = $httpTransport->performGetRequest($url, $timeout); |
295
|
1 |
|
} elseif ($method == self::METHOD_POST) { |
296
|
|
|
// FIXME should respect all headers, not only Content-Type |
297
|
1 |
|
$httpResponse = $httpTransport->performPostRequest($url, $rawPost, |
298
|
1 |
|
$requestHeaders['Content-Type'], $timeout); |
299
|
1 |
|
} |
300
|
|
|
|
301
|
1 |
|
$solrResponse = new \Apache_Solr_Response($httpResponse, |
|
|
|
|
302
|
1 |
|
$this->_createDocuments, $this->_collapseSingleValueArrays); |
303
|
|
|
|
304
|
1 |
|
if ($solrResponse->getHttpStatus() != 200) { |
305
|
|
|
throw new \Apache_Solr_HttpTransportException($solrResponse); |
306
|
|
|
} |
307
|
|
|
|
308
|
1 |
|
return $solrResponse; |
309
|
|
|
} |
310
|
|
|
|
311
|
|
|
/** |
312
|
|
|
* Return a valid http URL given this server's scheme, host, port, and path |
313
|
|
|
* and a provided servlet name. |
314
|
|
|
* |
315
|
|
|
* @param string $servlet Servlet name |
316
|
|
|
* @param array $params Additional URL parameters to attach to the end of the URL |
317
|
|
|
* @return string Servlet URL |
318
|
|
|
*/ |
319
|
53 |
|
protected function _constructUrl($servlet, $params = array()) |
320
|
|
|
{ |
321
|
53 |
|
$url = parent::_constructUrl($servlet, $params); |
322
|
|
|
|
323
|
53 |
|
if (!GeneralUtility::isFirstPartOfStr($url, $this->_scheme)) { |
324
|
1 |
|
$parsedUrl = parse_url($url); |
325
|
|
|
|
326
|
|
|
// unfortunately can't use str_replace as it replace all |
327
|
|
|
// occurrences of $needle and can't be limited to replace only once |
328
|
1 |
|
$url = $this->_scheme . substr($url, strlen($parsedUrl['scheme'])); |
329
|
1 |
|
} |
330
|
|
|
|
331
|
53 |
|
return $url; |
332
|
|
|
} |
333
|
|
|
|
334
|
|
|
/** |
335
|
|
|
* Returns the set scheme |
336
|
|
|
* |
337
|
|
|
* @return string |
338
|
|
|
*/ |
339
|
|
|
public function getScheme() |
340
|
|
|
{ |
341
|
|
|
return $this->_scheme; |
342
|
|
|
} |
343
|
|
|
|
344
|
|
|
/** |
345
|
|
|
* Set the scheme used. If empty will fallback to constants |
346
|
|
|
* |
347
|
|
|
* @param string $scheme Either http or https |
348
|
|
|
* @throws \UnexpectedValueException |
349
|
|
|
*/ |
350
|
54 |
|
public function setScheme($scheme) |
351
|
|
|
{ |
352
|
|
|
// Use the provided scheme or use the default |
353
|
54 |
|
if (empty($scheme)) { |
354
|
1 |
|
throw new \UnexpectedValueException('Scheme parameter is empty', |
355
|
1 |
|
1380756390); |
356
|
|
|
} else { |
357
|
53 |
|
if (in_array($scheme, |
358
|
53 |
|
array(self::SCHEME_HTTP, self::SCHEME_HTTPS))) { |
359
|
53 |
|
$this->_scheme = $scheme; |
360
|
53 |
|
} else { |
361
|
|
|
throw new \UnexpectedValueException('Unsupported scheme parameter, scheme must be http or https', |
362
|
|
|
1380756442); |
363
|
|
|
} |
364
|
|
|
} |
365
|
|
|
|
366
|
53 |
|
if ($this->_urlsInited) { |
367
|
|
|
$this->_initUrls(); |
368
|
|
|
} |
369
|
53 |
|
} |
370
|
|
|
|
371
|
|
|
/** |
372
|
|
|
* get field meta data for the index |
373
|
|
|
* |
374
|
|
|
* @param int $numberOfTerms Number of top terms to fetch for each field |
375
|
|
|
* @return array |
376
|
|
|
*/ |
377
|
|
|
public function getFieldsMetaData($numberOfTerms = 0) |
378
|
|
|
{ |
379
|
|
|
return $this->getLukeMetaData($numberOfTerms)->fields; |
380
|
|
|
} |
381
|
|
|
|
382
|
|
|
/** |
383
|
|
|
* Retrieves meta data about the index from the luke request handler |
384
|
|
|
* |
385
|
|
|
* @param int $numberOfTerms Number of top terms to fetch for each field |
386
|
|
|
* @return \Apache_Solr_Response Index meta data |
387
|
|
|
*/ |
388
|
|
|
public function getLukeMetaData($numberOfTerms = 0) |
389
|
|
|
{ |
390
|
|
|
if (!isset($this->lukeData[$numberOfTerms])) { |
391
|
|
|
$lukeUrl = $this->_constructUrl( |
392
|
|
|
self::LUKE_SERVLET, |
393
|
|
|
array( |
394
|
|
|
'numTerms' => $numberOfTerms, |
395
|
|
|
'wt' => self::SOLR_WRITER, |
396
|
|
|
'fl' => '*' |
397
|
|
|
) |
398
|
|
|
); |
399
|
|
|
|
400
|
|
|
$this->lukeData[$numberOfTerms] = $this->_sendRawGet($lukeUrl); |
401
|
|
|
} |
402
|
|
|
|
403
|
|
|
return $this->lukeData[$numberOfTerms]; |
404
|
|
|
} |
405
|
|
|
|
406
|
|
|
/** |
407
|
|
|
* Central method for making a get operation against this Solr Server |
408
|
|
|
* |
409
|
|
|
* @param string $url |
410
|
|
|
* @param float|bool $timeout Read timeout in seconds |
411
|
|
|
* @return \Apache_Solr_Response |
412
|
|
|
*/ |
413
|
53 |
|
protected function _sendRawGet($url, $timeout = false) |
414
|
|
|
{ |
415
|
53 |
|
$logSeverity = 0; // info |
416
|
|
|
|
417
|
|
|
try { |
418
|
53 |
|
$response = parent::_sendRawGet($url, $timeout); |
419
|
53 |
|
} catch (\Apache_Solr_HttpTransportException $e) { |
|
|
|
|
420
|
1 |
|
$logSeverity = 3; // fatal error |
421
|
1 |
|
$response = $e->getResponse(); |
422
|
|
|
} |
423
|
|
|
|
424
|
53 |
|
if ($this->configuration->getLoggingQueryRawGet() || $response->getHttpStatus() != 200) { |
|
|
|
|
425
|
|
|
$logData = array( |
426
|
1 |
|
'query url' => $url, |
427
|
|
|
'response' => (array)$response |
428
|
1 |
|
); |
429
|
|
|
|
430
|
1 |
|
if (!empty($e)) { |
431
|
1 |
|
$logData['exception'] = $e->__toString(); |
432
|
1 |
|
} else { |
433
|
|
|
// trigger data parsing |
434
|
|
|
$response->response; |
435
|
|
|
$logData['response data'] = print_r($response, true); |
436
|
|
|
} |
437
|
|
|
|
438
|
1 |
|
GeneralUtility::devLog('Querying Solr using GET', 'solr', |
439
|
1 |
|
$logSeverity, $logData); |
440
|
1 |
|
} |
441
|
|
|
|
442
|
53 |
|
return $response; |
443
|
|
|
} |
444
|
|
|
|
445
|
|
|
/** |
446
|
|
|
* Returns whether a search has been executed or not. |
447
|
|
|
* |
448
|
|
|
* @return bool TRUE if a search has been executed, FALSE otherwise |
449
|
|
|
*/ |
450
|
|
|
public function hasSearched() |
451
|
|
|
{ |
452
|
|
|
return $this->hasSearched; |
453
|
|
|
} |
454
|
|
|
|
455
|
|
|
/** |
456
|
|
|
* Gets the most recent response (if any) |
457
|
|
|
* |
458
|
|
|
* @return \Apache_Solr_Response Most recent response, or NULL if a search has not been executed yet. |
459
|
|
|
*/ |
460
|
|
|
public function getResponse() |
461
|
|
|
{ |
462
|
|
|
return $this->responseCache; |
463
|
|
|
} |
464
|
|
|
|
465
|
|
|
/** |
466
|
|
|
* Enable/Disable debug mode |
467
|
|
|
* |
468
|
|
|
* @param bool $debug TRUE to enable debug mode, FALSE to turn off, off by default |
469
|
|
|
*/ |
470
|
|
|
public function setDebug($debug) |
471
|
|
|
{ |
472
|
|
|
$this->debug = (boolean)$debug; |
473
|
|
|
} |
474
|
|
|
|
475
|
|
|
/** |
476
|
|
|
* Gets information about the plugins installed in Solr |
477
|
|
|
* |
478
|
|
|
* @return array A nested array of plugin data. |
479
|
|
|
*/ |
480
|
|
|
public function getPluginsInformation() |
481
|
|
|
{ |
482
|
|
|
if (empty($this->pluginsData)) { |
483
|
|
|
$pluginsInformation = $this->_sendRawGet($this->_pluginsUrl); |
484
|
|
|
|
485
|
|
|
// access a random property to trigger response parsing |
486
|
|
|
$pluginsInformation->responseHeader; |
487
|
|
|
$this->pluginsData = $pluginsInformation; |
488
|
|
|
} |
489
|
|
|
|
490
|
|
|
return $this->pluginsData; |
491
|
|
|
} |
492
|
|
|
|
493
|
|
|
/** |
494
|
|
|
* Gets the name of the schema.xml file installed and in use on the Solr |
495
|
|
|
* server. |
496
|
|
|
* |
497
|
|
|
* @return string Name of the active schema.xml |
498
|
|
|
*/ |
499
|
1 |
|
public function getSchemaName() |
500
|
|
|
{ |
501
|
1 |
|
if (is_null($this->schemaName)) { |
502
|
1 |
|
$systemInformation = $this->getSystemInformation(); |
503
|
1 |
|
$this->schemaName = $systemInformation->core->schema; |
504
|
1 |
|
} |
505
|
|
|
|
506
|
1 |
|
return $this->schemaName; |
507
|
|
|
} |
508
|
|
|
|
509
|
|
|
/** |
510
|
|
|
* Gets information about the Solr server |
511
|
|
|
* |
512
|
|
|
* @return array A nested array of system data. |
513
|
|
|
*/ |
514
|
2 |
|
public function getSystemInformation() |
515
|
|
|
{ |
516
|
2 |
|
if (empty($this->systemData)) { |
517
|
2 |
|
$systemInformation = $this->system(); |
518
|
|
|
|
519
|
|
|
// access a random property to trigger response parsing |
520
|
2 |
|
$systemInformation->responseHeader; |
521
|
2 |
|
$this->systemData = $systemInformation; |
522
|
2 |
|
} |
523
|
|
|
|
524
|
2 |
|
return $this->systemData; |
525
|
|
|
} |
526
|
|
|
|
527
|
|
|
/** |
528
|
|
|
* Gets the name of the solrconfig.xml file installed and in use on the Solr |
529
|
|
|
* server. |
530
|
|
|
* |
531
|
|
|
* @return string Name of the active solrconfig.xml |
532
|
|
|
*/ |
533
|
1 |
|
public function getSolrconfigName() |
534
|
|
|
{ |
535
|
1 |
|
if (is_null($this->solrconfigName)) { |
536
|
1 |
|
$solrconfigXmlUrl = $this->_scheme . '://' |
537
|
1 |
|
. $this->_host . ':' . $this->_port |
538
|
1 |
|
. $this->_path . 'admin/file/?file=solrconfig.xml'; |
539
|
|
|
|
540
|
1 |
|
$solrconfigXml = simplexml_load_file($solrconfigXmlUrl); |
541
|
1 |
|
if ($solrconfigXml === false) { |
542
|
|
|
throw new \InvalidArgumentException('No valid xml response from schema file: ' . $solrconfigXmlUrl); |
543
|
|
|
} |
544
|
1 |
|
$this->solrconfigName = (string)$solrconfigXml->attributes()->name; |
545
|
1 |
|
} |
546
|
|
|
|
547
|
1 |
|
return $this->solrconfigName; |
548
|
|
|
} |
549
|
|
|
|
550
|
|
|
/** |
551
|
|
|
* Gets the Solr server's version number. |
552
|
|
|
* |
553
|
|
|
* @return string Solr version number |
554
|
|
|
*/ |
555
|
1 |
|
public function getSolrServerVersion() |
556
|
|
|
{ |
557
|
1 |
|
$systemInformation = $this->getSystemInformation(); |
558
|
|
|
|
559
|
|
|
// don't know why $systemInformation->lucene->solr-spec-version won't work |
560
|
1 |
|
$luceneInformation = (array)$systemInformation->lucene; |
561
|
1 |
|
return $luceneInformation['solr-spec-version']; |
562
|
|
|
} |
563
|
|
|
|
564
|
|
|
/** |
565
|
|
|
* Deletes all index documents of a certain type and does a commit |
566
|
|
|
* afterwards. |
567
|
|
|
* |
568
|
|
|
* @param string $type The type of documents to delete, usually a table name. |
569
|
|
|
* @param bool $commit Will commit immediately after deleting the documents if set, defaults to TRUE |
570
|
|
|
*/ |
571
|
|
|
public function deleteByType($type, $commit = true) |
572
|
|
|
{ |
573
|
|
|
$this->deleteByQuery('type:' . trim($type)); |
574
|
|
|
|
575
|
|
|
if ($commit) { |
576
|
|
|
$this->commit(false, false, false); |
577
|
|
|
} |
578
|
|
|
} |
579
|
|
|
|
580
|
|
|
/** |
581
|
|
|
* Raw Delete Method. Takes a raw post body and sends it to the update service. Body should be |
582
|
|
|
* a complete and well formed "delete" xml document |
583
|
|
|
* |
584
|
|
|
* @param string $rawPost Expected to be utf-8 encoded xml document |
585
|
|
|
* @param float|int $timeout Maximum expected duration of the delete operation on the server (otherwise, will throw a communication exception) |
586
|
|
|
* @return \Apache_Solr_Response |
587
|
|
|
*/ |
588
|
5 |
|
public function delete($rawPost, $timeout = 3600) |
589
|
|
|
{ |
590
|
5 |
|
$response = $this->_sendRawPost($this->_updateUrl, $rawPost, $timeout); |
591
|
|
|
|
592
|
5 |
|
GeneralUtility::devLog( |
593
|
5 |
|
'Delete Query sent.', |
594
|
5 |
|
'solr', |
595
|
5 |
|
1, |
596
|
|
|
array( |
597
|
5 |
|
'query' => $rawPost, |
598
|
5 |
|
'query url' => $this->_updateUrl, |
599
|
|
|
'response' => (array)$response |
600
|
5 |
|
) |
601
|
5 |
|
); |
602
|
|
|
|
603
|
5 |
|
return $response; |
604
|
|
|
} |
605
|
|
|
|
606
|
|
|
/** |
607
|
|
|
* Central method for making a post operation against this Solr Server |
608
|
|
|
* |
609
|
|
|
* @param string $url |
610
|
|
|
* @param string $rawPost |
611
|
|
|
* @param float|bool $timeout Read timeout in seconds |
612
|
|
|
* @param string $contentType |
613
|
|
|
* @return \Apache_Solr_Response |
614
|
|
|
*/ |
615
|
46 |
|
protected function _sendRawPost( |
616
|
|
|
$url, |
617
|
|
|
$rawPost, |
618
|
|
|
$timeout = false, |
619
|
|
|
$contentType = 'text/xml; charset=UTF-8' |
620
|
|
|
) { |
621
|
46 |
|
$logSeverity = 0; // info |
622
|
|
|
|
623
|
|
|
try { |
624
|
46 |
|
$response = parent::_sendRawPost($url, $rawPost, $timeout, |
625
|
46 |
|
$contentType); |
626
|
46 |
|
} catch (\Apache_Solr_HttpTransportException $e) { |
|
|
|
|
627
|
|
|
$logSeverity = 3; // fatal error |
628
|
|
|
$response = $e->getResponse(); |
629
|
|
|
} |
630
|
|
|
|
631
|
46 |
|
if ($this->configuration->getLoggingQueryRawPost() || $response->getHttpStatus() != 200) { |
|
|
|
|
632
|
|
|
$logData = array( |
633
|
|
|
'query url' => $url, |
634
|
|
|
'content' => $rawPost, |
635
|
|
|
'response' => (array)$response |
636
|
|
|
); |
637
|
|
|
|
638
|
|
|
if (!empty($e)) { |
639
|
|
|
$logData['exception'] = $e->__toString(); |
640
|
|
|
} |
641
|
|
|
|
642
|
|
|
GeneralUtility::devLog('Querying Solr using POST', 'solr', |
643
|
|
|
$logSeverity, $logData); |
644
|
|
|
} |
645
|
|
|
|
646
|
46 |
|
return $response; |
647
|
|
|
} |
648
|
|
|
|
649
|
|
|
/** |
650
|
|
|
* Get currently configured synonyms |
651
|
|
|
* |
652
|
|
|
* @param string $baseWord If given a base word, retrieves the synonyms for that word only |
653
|
|
|
* @return array |
654
|
|
|
*/ |
655
|
|
|
public function getSynonyms($baseWord = '') |
656
|
|
|
{ |
657
|
|
|
$synonymsUrl = $this->_synonymsUrl; |
658
|
|
|
if (!empty($baseWord)) { |
659
|
|
|
$synonymsUrl .= '/' . $baseWord; |
660
|
|
|
} |
661
|
|
|
|
662
|
|
|
$response = $this->_sendRawGet($synonymsUrl); |
663
|
|
|
$decodedResponse = json_decode($response->getRawResponse()); |
664
|
|
|
|
665
|
|
|
$synonyms = array(); |
666
|
|
|
if (!empty($baseWord)) { |
667
|
|
|
if (is_array($decodedResponse->{$baseWord})) { |
668
|
|
|
$synonyms = $decodedResponse->{$baseWord}; |
669
|
|
|
} |
670
|
|
|
} else { |
671
|
|
|
if (isset($decodedResponse->synonymMappings->managedMap)) { |
672
|
|
|
$synonyms = (array)$decodedResponse->synonymMappings->managedMap; |
673
|
|
|
} |
674
|
|
|
} |
675
|
|
|
|
676
|
|
|
return $synonyms; |
677
|
|
|
} |
678
|
|
|
|
679
|
|
|
/** |
680
|
|
|
* Add list of synonyms for base word to managed synonyms map |
681
|
|
|
* |
682
|
|
|
* @param $baseWord |
683
|
|
|
* @param array $synonyms |
684
|
|
|
* |
685
|
|
|
* @return \Apache_Solr_Response |
686
|
|
|
* |
687
|
|
|
* @throws \Apache_Solr_InvalidArgumentException If $baseWord or $synonyms are empty |
688
|
|
|
*/ |
689
|
|
|
public function addSynonym($baseWord, array $synonyms) |
690
|
|
|
{ |
691
|
|
|
if (empty($baseWord) || empty($synonyms)) { |
692
|
|
|
throw new \Apache_Solr_InvalidArgumentException('Must provide base word and synonyms.'); |
693
|
|
|
} |
694
|
|
|
|
695
|
|
|
$rawPut = json_encode(array($baseWord => $synonyms)); |
696
|
|
|
return $this->_sendRawPost($this->_synonymsUrl, $rawPut, |
697
|
|
|
$this->getHttpTransport()->getDefaultTimeout(), 'application/json'); |
698
|
|
|
} |
699
|
|
|
|
700
|
|
|
/** |
701
|
|
|
* Remove a synonym from the synonyms map |
702
|
|
|
* |
703
|
|
|
* @param $baseWord |
704
|
|
|
* @return \Apache_Solr_Response |
705
|
|
|
* @throws \Apache_Solr_InvalidArgumentException |
706
|
|
|
*/ |
707
|
|
|
public function deleteSynonym($baseWord) |
708
|
|
|
{ |
709
|
|
|
if (empty($baseWord)) { |
710
|
|
|
throw new \Apache_Solr_InvalidArgumentException('Must provide base word.'); |
711
|
|
|
} |
712
|
|
|
|
713
|
|
|
return $this->_sendRawDelete($this->_synonymsUrl . '/' . $baseWord); |
714
|
|
|
} |
715
|
|
|
|
716
|
|
|
/** |
717
|
|
|
* Central method for making a HTTP DELETE operation against the Solr server |
718
|
|
|
* |
719
|
|
|
* @param $url |
720
|
|
|
* @param bool|float $timeout Read timeout in seconds |
721
|
|
|
* @return \Apache_Solr_Response |
722
|
|
|
*/ |
723
|
|
|
protected function _sendRawDelete($url, $timeout = false) |
724
|
|
|
{ |
725
|
|
|
$logSeverity = 0; // info |
726
|
|
|
|
727
|
|
|
try { |
728
|
|
|
$httpTransport = $this->getHttpTransport(); |
729
|
|
|
|
730
|
|
|
$httpResponse = $httpTransport->performDeleteRequest($url, |
731
|
|
|
$timeout); |
732
|
|
|
$solrResponse = new \Apache_Solr_Response($httpResponse, |
733
|
|
|
$this->_createDocuments, $this->_collapseSingleValueArrays); |
734
|
|
|
|
735
|
|
|
if ($solrResponse->getHttpStatus() != 200) { |
736
|
|
|
throw new \Apache_Solr_HttpTransportException($solrResponse); |
737
|
|
|
} |
738
|
|
|
} catch (\Apache_Solr_HttpTransportException $e) { |
|
|
|
|
739
|
|
|
$logSeverity = 3; // fatal error |
740
|
|
|
$solrResponse = $e->getResponse(); |
741
|
|
|
} |
742
|
|
|
|
743
|
|
|
if ($this->configuration->getLoggingQueryRawDelete() || $solrResponse->getHttpStatus() != 200) { |
|
|
|
|
744
|
|
|
$logData = array( |
745
|
|
|
'query url' => $url, |
746
|
|
|
'response' => (array)$solrResponse |
747
|
|
|
); |
748
|
|
|
|
749
|
|
|
if (!empty($e)) { |
750
|
|
|
$logData['exception'] = $e->__toString(); |
751
|
|
|
} else { |
752
|
|
|
// trigger data parsing |
753
|
|
|
$solrResponse->response; |
754
|
|
|
$logData['response data'] = print_r($solrResponse, true); |
755
|
|
|
} |
756
|
|
|
|
757
|
|
|
GeneralUtility::devLog('Querying Solr using DELETE', 'solr', |
758
|
|
|
$logSeverity, $logData); |
759
|
|
|
} |
760
|
|
|
|
761
|
|
|
return $solrResponse; |
762
|
|
|
} |
763
|
|
|
|
764
|
|
|
/** |
765
|
|
|
* Get currently configured stop words |
766
|
|
|
* |
767
|
|
|
* @return array |
768
|
|
|
*/ |
769
|
|
|
public function getStopWords() |
770
|
|
|
{ |
771
|
|
|
$stopWords = array(); |
772
|
|
|
|
773
|
|
|
$response = $this->_sendRawGet($this->_stopWordsUrl); |
774
|
|
|
$decodedResponse = json_decode($response->getRawResponse()); |
775
|
|
|
|
776
|
|
|
if (isset($decodedResponse->wordSet->managedList)) { |
777
|
|
|
$stopWords = (array)$decodedResponse->wordSet->managedList; |
778
|
|
|
} |
779
|
|
|
|
780
|
|
|
return $stopWords; |
781
|
|
|
} |
782
|
|
|
|
783
|
|
|
/** |
784
|
|
|
* Adds stop words to the managed stop word list |
785
|
|
|
* |
786
|
|
|
* @param array|string $stopWords string for a single word, array for multiple words |
787
|
|
|
* @return \Apache_Solr_Response |
788
|
|
|
* @throws \Apache_Solr_InvalidArgumentException If $stopWords is empty |
789
|
|
|
*/ |
790
|
|
|
public function addStopWords($stopWords) |
791
|
|
|
{ |
792
|
|
|
if (empty($stopWords)) { |
793
|
|
|
throw new \Apache_Solr_InvalidArgumentException('Must provide stop word.'); |
794
|
|
|
} |
795
|
|
|
|
796
|
|
|
if (is_string($stopWords)) { |
797
|
|
|
$stopWords = array($stopWords); |
798
|
|
|
} |
799
|
|
|
|
800
|
|
|
$stopWords = array_values($stopWords); |
801
|
|
|
$rawPut = json_encode($stopWords); |
802
|
|
|
return $this->_sendRawPost($this->_stopWordsUrl, $rawPut, |
803
|
|
|
$this->getHttpTransport()->getDefaultTimeout(), 'application/json'); |
804
|
|
|
} |
805
|
|
|
|
806
|
|
|
/** |
807
|
|
|
* Deletes a words from the managed stop word list |
808
|
|
|
* |
809
|
|
|
* @param string $stopWord stop word to delete |
810
|
|
|
* @return \Apache_Solr_Response |
811
|
|
|
* @throws \Apache_Solr_InvalidArgumentException If $stopWords is empty |
812
|
|
|
*/ |
813
|
|
|
public function deleteStopWord($stopWord) |
814
|
|
|
{ |
815
|
|
|
if (empty($stopWord)) { |
816
|
|
|
throw new \Apache_Solr_InvalidArgumentException('Must provide stop word.'); |
817
|
|
|
} |
818
|
|
|
|
819
|
|
|
return $this->_sendRawDelete($this->_stopWordsUrl . '/' . $stopWord); |
820
|
|
|
} |
821
|
|
|
|
822
|
|
|
/** |
823
|
|
|
* Reloads the current core |
824
|
|
|
* |
825
|
|
|
* @return \Apache_Solr_Response |
826
|
|
|
*/ |
827
|
|
|
public function reloadCore() |
828
|
|
|
{ |
829
|
|
|
$coreName = array_pop(explode('/', trim($this->_path, '/'))); |
|
|
|
|
830
|
|
|
$coreAdminReloadUrl = $this->_coresUrl . '?action=reload&core=' . $coreName; |
831
|
|
|
|
832
|
|
|
return $this->_sendRawGet($coreAdminReloadUrl); |
833
|
|
|
} |
834
|
|
|
|
835
|
|
|
/** |
836
|
|
|
* initializes various URLs, including the Luke URL |
837
|
|
|
* |
838
|
|
|
* @return void |
839
|
|
|
*/ |
840
|
53 |
|
protected function _initUrls() |
841
|
|
|
{ |
842
|
53 |
|
parent::_initUrls(); |
843
|
|
|
|
844
|
53 |
|
$this->_lukeUrl = $this->_constructUrl( |
845
|
53 |
|
self::LUKE_SERVLET, |
846
|
|
|
array( |
847
|
53 |
|
'numTerms' => '0', |
848
|
|
|
'wt' => self::SOLR_WRITER |
849
|
53 |
|
) |
850
|
53 |
|
); |
851
|
|
|
|
852
|
53 |
|
$this->_pluginsUrl = $this->_constructUrl( |
853
|
53 |
|
self::PLUGINS_SERVLET, |
854
|
53 |
|
array('wt' => self::SOLR_WRITER) |
855
|
53 |
|
); |
856
|
|
|
|
857
|
53 |
|
$pathElements = explode('/', trim($this->_path, '/')); |
858
|
53 |
|
$this->_coresUrl = |
859
|
53 |
|
$this->_scheme . '://' . |
860
|
53 |
|
$this->_host . ':' . |
861
|
53 |
|
$this->_port . '/' . |
862
|
53 |
|
$pathElements[0] . '/' . |
863
|
53 |
|
self::CORES_SERVLET; |
864
|
|
|
|
865
|
53 |
|
$this->_schemaUrl = $this->_constructUrl(self::SCHEMA_SERVLET); |
866
|
|
|
|
867
|
53 |
|
$managedLanguage = $this->getManagedLanguage(); |
868
|
53 |
|
$this->_synonymsUrl = $this->_constructUrl( |
869
|
|
|
self::SYNONYMS_SERVLET |
870
|
53 |
|
) . $managedLanguage; |
871
|
53 |
|
$this->_stopWordsUrl = $this->_constructUrl( |
872
|
|
|
self::STOPWORDS_SERVLET |
873
|
53 |
|
) . $managedLanguage; |
874
|
53 |
|
} |
875
|
|
|
|
876
|
|
|
/** |
877
|
|
|
* Get the language map name for the text field. |
878
|
|
|
* This is necessary to select the correct synonym map. |
879
|
|
|
* |
880
|
|
|
* @return string |
881
|
|
|
*/ |
882
|
53 |
|
protected function getManagedLanguage() |
883
|
|
|
{ |
884
|
53 |
|
$language = 'english'; |
885
|
|
|
|
886
|
53 |
|
$schema = $this->getSchema(); |
887
|
|
|
|
888
|
53 |
|
if (is_object($schema) && isset($schema->fieldTypes)) { |
889
|
52 |
|
foreach ($schema->fieldTypes as $fieldType) { |
890
|
52 |
|
if ($fieldType->name === 'text') { |
891
|
52 |
|
foreach ($fieldType->queryAnalyzer->filters as $filter) { |
892
|
52 |
|
if ($filter->class === 'solr.ManagedSynonymFilterFactory') { |
893
|
52 |
|
$language = $filter->managed; |
894
|
52 |
|
} |
895
|
52 |
|
} |
896
|
52 |
|
} |
897
|
52 |
|
} |
898
|
52 |
|
} |
899
|
|
|
|
900
|
53 |
|
return $language; |
901
|
|
|
} |
902
|
|
|
|
903
|
|
|
/** |
904
|
|
|
* Get the configured schema for the current core |
905
|
|
|
* |
906
|
|
|
* @return \stdClass |
907
|
|
|
*/ |
908
|
53 |
|
protected function getSchema() |
909
|
|
|
{ |
910
|
53 |
|
$response = $this->_sendRawGet($this->_schemaUrl); |
911
|
53 |
|
return json_decode($response->getRawResponse())->schema; |
912
|
|
|
} |
913
|
|
|
} |
914
|
|
|
|
Our type inference engine has found an assignment to a property that is incompatible with the declared type of that property.
Either this assignment is in error or the assigned type should be added to the documentation/type hint for that property..