|
1
|
|
|
<?php |
|
2
|
|
|
namespace ApacheSolrForTypo3\Solr\Domain\Search\Query\Helper; |
|
3
|
|
|
|
|
4
|
|
|
/*************************************************************** |
|
5
|
|
|
* Copyright notice |
|
6
|
|
|
* |
|
7
|
|
|
* (c) 2017 Timo Hund <[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
|
|
|
/** |
|
28
|
|
|
* The EscpaeService is responsible to escape the querystring as ecpected for Apache Solr. |
|
29
|
|
|
* |
|
30
|
|
|
* @author Timo Hund <[email protected]> |
|
31
|
|
|
*/ |
|
32
|
|
|
class EscapeService { |
|
33
|
|
|
|
|
34
|
|
|
/** |
|
35
|
|
|
* Quote and escape search strings |
|
36
|
|
|
* |
|
37
|
|
|
* @param string|int|double $string String to escape |
|
38
|
|
|
* @return string|int|double The escaped/quoted string |
|
39
|
|
|
*/ |
|
40
|
|
|
public function escape($string) |
|
41
|
|
|
{ |
|
42
|
|
|
// when we have a numeric string only, nothing needs to be done |
|
43
|
|
|
if (is_numeric($string)) { |
|
44
|
|
|
return $string; |
|
45
|
|
|
} |
|
46
|
|
|
|
|
47
|
|
|
// when no whitespaces are in the query we can also just escape the special characters |
|
48
|
|
|
if (preg_match('/\W/', $string) != 1) { |
|
49
|
|
|
return $this->escapeSpecialCharacters($string); |
|
50
|
|
|
} |
|
51
|
|
|
|
|
52
|
|
|
// when there are no quotes inside the query string we can also just escape the whole string |
|
53
|
|
|
$hasQuotes = strrpos($string, '"') !== false; |
|
54
|
|
|
if (!$hasQuotes) { |
|
55
|
|
|
return $this->escapeSpecialCharacters($string); |
|
56
|
|
|
} |
|
57
|
|
|
|
|
58
|
|
|
$result = $this->tokenizeByQuotesAndEscapeDependingOnContext($string); |
|
59
|
|
|
|
|
60
|
|
|
return $result; |
|
61
|
|
|
} |
|
62
|
|
|
|
|
63
|
|
|
/** |
|
64
|
|
|
* This method is used to escape the content in the query string surrounded by quotes |
|
65
|
|
|
* different then when it is not in a quoted context. |
|
66
|
|
|
* |
|
67
|
|
|
* @param string $string |
|
68
|
|
|
* @return string |
|
69
|
|
|
*/ |
|
70
|
|
|
protected function tokenizeByQuotesAndEscapeDependingOnContext($string) |
|
71
|
|
|
{ |
|
72
|
|
|
$result = ''; |
|
73
|
|
|
$quotesCount = substr_count($string, '"'); |
|
74
|
|
|
$isEvenAmountOfQuotes = $quotesCount % 2 === 0; |
|
75
|
|
|
|
|
76
|
|
|
// go over all quote segments and apply escapePhrase inside a quoted |
|
77
|
|
|
// context and escapeSpecialCharacters outside the quoted context. |
|
78
|
|
|
$segments = explode('"', $string); |
|
79
|
|
|
$segmentsIndex = 0; |
|
80
|
|
|
foreach ($segments as $segment) { |
|
81
|
|
|
$isInQuote = $segmentsIndex % 2 !== 0; |
|
82
|
|
|
$isLastQuote = $segmentsIndex === $quotesCount; |
|
83
|
|
|
|
|
84
|
|
|
if ($isLastQuote && !$isEvenAmountOfQuotes) { |
|
85
|
|
|
$result .= '\"'; |
|
86
|
|
|
} |
|
87
|
|
|
|
|
88
|
|
|
if ($isInQuote && !$isLastQuote) { |
|
89
|
|
|
$result .= $this->escapePhrase($segment); |
|
90
|
|
|
} else { |
|
91
|
|
|
$result .= $this->escapeSpecialCharacters($segment); |
|
92
|
|
|
} |
|
93
|
|
|
|
|
94
|
|
|
$segmentsIndex++; |
|
95
|
|
|
} |
|
96
|
|
|
|
|
97
|
|
|
return $result; |
|
98
|
|
|
} |
|
99
|
|
|
|
|
100
|
|
|
/** |
|
101
|
|
|
* Escapes a value meant to be contained in a phrase with characters with |
|
102
|
|
|
* special meanings in Lucene query syntax. |
|
103
|
|
|
* |
|
104
|
|
|
* @param string $value Unescaped - "dirty" - string |
|
105
|
|
|
* @return string Escaped - "clean" - string |
|
106
|
|
|
*/ |
|
107
|
|
|
protected function escapePhrase($value) |
|
108
|
|
|
{ |
|
109
|
|
|
$pattern = '/("|\\\)/'; |
|
110
|
|
|
$replace = '\\\$1'; |
|
111
|
|
|
|
|
112
|
|
|
return '"' . preg_replace($pattern, $replace, $value) . '"'; |
|
113
|
|
|
} |
|
114
|
|
|
|
|
115
|
|
|
/** |
|
116
|
|
|
* Escapes characters with special meanings in Lucene query syntax. |
|
117
|
|
|
* |
|
118
|
|
|
* @param string $value Unescaped - "dirty" - string |
|
119
|
|
|
* @return string Escaped - "clean" - string |
|
120
|
|
|
*/ |
|
121
|
|
|
protected function escapeSpecialCharacters($value) |
|
122
|
|
|
{ |
|
123
|
|
|
// list taken from http://lucene.apache.org/core/4_4_0/queryparser/org/apache/lucene/queryparser/classic/package-summary.html#package_description |
|
124
|
|
|
// which mentions: + - && || ! ( ) { } [ ] ^ " ~ * ? : \ / |
|
125
|
|
|
// of which we escape: ( ) { } [ ] ^ " ~ : \ / |
|
126
|
|
|
// and explicitly don't escape: + - && || ! * ? |
|
127
|
|
|
$pattern = '/(\\(|\\)|\\{|\\}|\\[|\\]|\\^|"|~|\:|\\\\|\\/)/'; |
|
128
|
|
|
$replace = '\\\$1'; |
|
129
|
|
|
|
|
130
|
|
|
return preg_replace($pattern, $replace, $value); |
|
131
|
|
|
} |
|
132
|
|
|
} |
timohund /
ext-solr
