|
1 <?php |
|
2 /** |
|
3 * Zend Framework |
|
4 * |
|
5 * LICENSE |
|
6 * |
|
7 * This source file is subject to the new BSD license that is bundled |
|
8 * with this package in the file LICENSE.txt. |
|
9 * It is also available through the world-wide-web at this URL: |
|
10 * http://framework.zend.com/license/new-bsd |
|
11 * If you did not receive a copy of the license and are unable to |
|
12 * obtain it through the world-wide-web, please send an email |
|
13 * to license@zend.com so we can send you a copy immediately. |
|
14 * |
|
15 * @category Zend |
|
16 * @package Zend_Search_Lucene |
|
17 * @subpackage Analysis |
|
18 * @copyright Copyright (c) 2005-2010 Zend Technologies USA Inc. (http://www.zend.com) |
|
19 * @license http://framework.zend.com/license/new-bsd New BSD License |
|
20 * @version $Id: TextNum.php 20096 2010-01-06 02:05:09Z bkarwin $ |
|
21 */ |
|
22 |
|
23 |
|
24 /** Zend_Search_Lucene_Analysis_Analyzer_Common */ |
|
25 require_once 'Zend/Search/Lucene/Analysis/Analyzer/Common.php'; |
|
26 |
|
27 |
|
28 /** |
|
29 * @category Zend |
|
30 * @package Zend_Search_Lucene |
|
31 * @subpackage Analysis |
|
32 * @copyright Copyright (c) 2005-2010 Zend Technologies USA Inc. (http://www.zend.com) |
|
33 * @license http://framework.zend.com/license/new-bsd New BSD License |
|
34 */ |
|
35 |
|
36 class Zend_Search_Lucene_Analysis_Analyzer_Common_TextNum extends Zend_Search_Lucene_Analysis_Analyzer_Common |
|
37 { |
|
38 /** |
|
39 * Current position in a stream |
|
40 * |
|
41 * @var integer |
|
42 */ |
|
43 private $_position; |
|
44 |
|
45 /** |
|
46 * Reset token stream |
|
47 */ |
|
48 public function reset() |
|
49 { |
|
50 $this->_position = 0; |
|
51 |
|
52 if ($this->_input === null) { |
|
53 return; |
|
54 } |
|
55 |
|
56 // convert input into ascii |
|
57 if (PHP_OS != 'AIX') { |
|
58 $this->_input = iconv($this->_encoding, 'ASCII//TRANSLIT', $this->_input); |
|
59 } |
|
60 $this->_encoding = 'ASCII'; |
|
61 } |
|
62 |
|
63 /** |
|
64 * Tokenization stream API |
|
65 * Get next token |
|
66 * Returns null at the end of stream |
|
67 * |
|
68 * @return Zend_Search_Lucene_Analysis_Token|null |
|
69 */ |
|
70 public function nextToken() |
|
71 { |
|
72 if ($this->_input === null) { |
|
73 return null; |
|
74 } |
|
75 |
|
76 do { |
|
77 if (! preg_match('/[a-zA-Z0-9]+/', $this->_input, $match, PREG_OFFSET_CAPTURE, $this->_position)) { |
|
78 // It covers both cases a) there are no matches (preg_match(...) === 0) |
|
79 // b) error occured (preg_match(...) === FALSE) |
|
80 return null; |
|
81 } |
|
82 |
|
83 $str = $match[0][0]; |
|
84 $pos = $match[0][1]; |
|
85 $endpos = $pos + strlen($str); |
|
86 |
|
87 $this->_position = $endpos; |
|
88 |
|
89 $token = $this->normalize(new Zend_Search_Lucene_Analysis_Token($str, $pos, $endpos)); |
|
90 } while ($token === null); // try again if token is skipped |
|
91 |
|
92 return $token; |
|
93 } |
|
94 } |
|
95 |