1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
|
<?php
mb\_internal\_encoding("UTF-8");
mb\_regex\_encoding("UTF-8");
class FullTextSearcher {
var $markwords;
var $scorequery;
public function FullTextSearcher() {
}
public function search_sql($statement) {
return "'". $this->to_ngram($statement, 2) ."' IN BOOLEAN MODE";
}
public function getScore() {
return "'".$this->scorequery."'";
}
public function getMarkupWords() {
return $this->markwords;
}
private function to_ngram($string, $n){
$string = mb\_ereg\_replace("^(\\s| )+","", $string);
$string = mb\_ereg\_replace("(\\s| )+$","", $string);
$str_array = preg_split("/(\\s| )+/", $string);
$result = array();
$markwords = array();
$scores = array();
foreach ($str_array as $str){
if ('-' != substr($str, 0, 1)) {
$markwords\[\] = $str;
}
$r = $this->to\_ngram\_query($str, $n);
$result\[\] = $r\[0\];
$scores\[\] = $r\[1\];
}
$this->markwords = $markwords;
$this->scorequery = join(' ', $scores);
return join(' ', $result);
}
private function to\_ngram\_query($string, $n){
$string = trim($string);
if ($string == ''){
return '';
}
if ('-' != substr($string, 0, 1)) {
$plus = true;
} else {
$plus = false;
$string = substr($string, 1);
}
$length = mb_strlen($string);
if ($length < $n){
if ($plus) {
return array("+".$string."*", $string);
} else {
return array("-".$string."*", "");
}
}
$ngrams = array();
$scores = array();
$wa = new WordsAnalyzer();
$wa->loadStr($string);
$wds = $wa->get_indexes();
foreach ($wds as $ngram){
if ($plus) {
$ngrams\[\] = "+" . $ngram;
$scores\[\] = $ngram;
} else {
$ngrams\[\] = "-" . $ngram;
}
}
return array(join(' ', $ngrams), join(' ', $scores));
}
}
class WordsAnalyzer {
private $words;
public function WordsAnalyzer() {
}
public function get_indexes() {
return array_keys($this->words);
}
public function get_fulltext() {
return join(' ', array_keys($this->words));
}
public function loadStr($str) {
$this->words = array();
ereg_replace(13, "", $str);
ereg_replace(10, " ", $str);
$this->analyze_token($this->split_token($str));
}
private function split_token($str) {
$token = array();
while (1) {
$bytes = mb_ereg("\[一-龠\]+|\[ぁ-ん\]+|\[ァ-ヴー\]+|\[a-zA-Z0-9\]+|\[ａ-ｚＡ-Ｚ０-９\]+", $str, $match);
if ($bytes == FALSE)
break;
$match = $match\[0\];
array_push($token, $match);
$pos = strpos($str, $match);
$str = substr($str, $pos+$bytes);
}
return $token;
}
private function analyze_token($token) {
for ($i = 0, $len = count($token); $i < $len; $i++) {
$word = $token\[$i\];
if (mb_ereg("\[一-龠\]", $word, $match) != FALSE) {
if (mb_strlen($word) > 1) {
$this->japanese($word);
} else {
$this->addword($word);
}
} else if (mb_ereg("\[ぁ-ん\]", $word, $match) != FALSE) {
if (mb_strlen($word) > 1) {
$this->hiragana($word);
}
} else if (mb_ereg("\[ァ-ヴー\]", $word, $match) != FALSE) {
if (mb_strlen($word) > 1) {
$this->japanese($word);
}
} else if (mb_ereg("\[a-zA-Z0-9\]", $word, $match) != FALSE) {
if (strlen($word) > 0) {
// camel-gram
$this->english($word);
}
} else if (mb_ereg("\[ａ-ｚＡ-Ｚ０-９\]", $word, $match) != FALSE) {
if (mb_strlen($word) > 0) {
// camel-gram
$word = mb\_convert\_kana($word, 'rn');
$this->english($word);
}
} else {
$this->addword($word);
}
}
}
private function addword($word) {
if (strlen($word) == 0) return;
$this->words{$word} = true;
}
private function japanese($str) {
// bi-gram
$ret = $this->split_ngram($str, 2);
foreach ($ret as $key => $value) {
if (mb_strlen($value) > 1) {
$this->addword($value);
}
}
}
private function hiragana($str) {
// bi-gram
$ret = $this->split_ngram($str, 2);
foreach ($ret as $key => $value) {
if (mb_strlen($value) > 1) {
if (mb_ereg("っ.|を|です|ます|する|ある|いる|から", $value, $match) != FALSE) {
} else {
$this->addword($value);
}
}
}
}
private function english($str) {
$this->addword($str); // そのままも登録
// キャメル記法を分解して登録
while(1){
$r = ereg("(\[a-z\]|\[A-z\]|\[0-9\])\[a-z\]+", $str, $match);
if ($r == FALSE) {
break;
}
$match = $match\[0\];
$this->addword($match);
$pos = strpos($str, $match);
$str = substr($str, $pos + $r);
}
}
private function split_ngram($string, $n){
$ngrams = array();
$string = trim($string);
if ($string == '')
return '';
for ($i = 0, $len = mb_strlen($string); $i < $len; $i++) {
$ngrams\[\] = mb_substr($string, $i, $n);
}
return $ngrams;
}
}
?>
|