24
|
1 |
/*
|
|
2 |
* Copyright (c) 2010 Nokia Corporation and/or its subsidiary(-ies).
|
|
3 |
* All rights reserved.
|
|
4 |
* This component and the accompanying materials are made available
|
|
5 |
* under the terms of "Eclipse Public License v1.0"
|
|
6 |
* which accompanies this distribution, and is available
|
|
7 |
* at the URL "http://www.eclipse.org/legal/epl-v10.html".
|
|
8 |
*
|
|
9 |
* Initial Contributors:
|
|
10 |
* Nokia Corporation - initial contribution.
|
|
11 |
*
|
|
12 |
* Contributors:
|
|
13 |
*
|
|
14 |
* Description:
|
|
15 |
*
|
|
16 |
*/
|
|
17 |
#include "thaianalysis.h"
|
|
18 |
|
|
19 |
#include "cpixfstools.h"
|
|
20 |
|
|
21 |
#include "CLucene/analysis/standard/StandardTokenizer.h"
|
|
22 |
|
|
23 |
#include <iostream>
|
|
24 |
#include <fstream>
|
|
25 |
|
|
26 |
#include "tinyunicode.h"
|
|
27 |
|
|
28 |
#include "thaistatemachine.h"
|
|
29 |
|
|
30 |
namespace analysis {
|
|
31 |
|
|
32 |
void InitThaiAnalysis(const char* thaiDataFile) {
|
|
33 |
ThaiAnalysisInfra::init(thaiDataFile);
|
|
34 |
}
|
|
35 |
void ShutdownThaiAnalysis() {
|
|
36 |
ThaiAnalysisInfra::shutdown();
|
|
37 |
}
|
|
38 |
|
|
39 |
ThaiAnalysisInfra* ThaiAnalysisInfra::theInstance_ = NULL;
|
|
40 |
|
|
41 |
const char* ThaiAnalysisInfraNotInitialized::what() const throw() {
|
|
42 |
return "Thai analyzer infra was not initialized.";
|
|
43 |
}
|
|
44 |
|
|
45 |
const char* StateMachineFileNotFound::what() const throw() {
|
|
46 |
return "Thai analyzer infra could not find specified StateMachine file.";
|
|
47 |
}
|
|
48 |
|
|
49 |
const char* StateMachineLoadingFailed::what() const throw() {
|
|
50 |
return "Thai analyzer infra failed reading the specified StateMachine file.";
|
|
51 |
}
|
|
52 |
|
|
53 |
|
|
54 |
void ThaiAnalysisInfra::init(const char* dataFile)
|
|
55 |
{
|
|
56 |
shutdown();
|
|
57 |
theInstance_ = new ThaiAnalysisInfra(dataFile);
|
|
58 |
}
|
|
59 |
|
|
60 |
ThaiAnalysisInfra* ThaiAnalysisInfra::theInstance()
|
|
61 |
{
|
|
62 |
if ( !theInstance_ ) throw ThaiAnalysisInfraNotInitialized();
|
|
63 |
return theInstance_;
|
|
64 |
}
|
|
65 |
|
|
66 |
void ThaiAnalysisInfra::shutdown()
|
|
67 |
{
|
|
68 |
delete theInstance_;
|
|
69 |
theInstance_ = 0;
|
|
70 |
}
|
|
71 |
|
|
72 |
std::auto_ptr<BreakIterator> ThaiAnalysisInfra::createBreakIterator()
|
|
73 |
{
|
|
74 |
if ( !blob_.get() )
|
|
75 |
{ // load lazily
|
|
76 |
off_t size = Cpt::filesize(dataFile_.c_str());
|
|
77 |
|
|
78 |
if ( !size ) throw StateMachineFileNotFound();
|
|
79 |
|
|
80 |
blob_.reset( new byte_t[size] );
|
|
81 |
|
|
82 |
std::ifstream in( dataFile_.c_str(), std::ifstream::in | std::ifstream::binary );
|
|
83 |
|
|
84 |
if ( !in ) throw StateMachineFileNotFound();
|
|
85 |
|
|
86 |
in.read( reinterpret_cast<char*>( blob_.get() ), size );
|
|
87 |
|
|
88 |
if ( in.fail() ) throw StateMachineLoadingFailed();
|
|
89 |
|
|
90 |
in.close();
|
|
91 |
|
|
92 |
stateMachine_.reset(blob_.get());
|
|
93 |
}
|
|
94 |
|
|
95 |
return std::auto_ptr<BreakIterator>( new StateMachineBreakIterator<ThaiSmEncoding>( stateMachine_ ) );
|
|
96 |
}
|
|
97 |
|
|
98 |
ThaiAnalysisInfra::ThaiAnalysisInfra(const char* dataFile)
|
|
99 |
: blob_(0),
|
|
100 |
stateMachine_(),
|
|
101 |
dataFile_(dataFile)
|
|
102 |
{
|
|
103 |
// sanity check
|
|
104 |
if ( !Cpt::filesize(dataFile) ) throw StateMachineFileNotFound();
|
|
105 |
}
|
|
106 |
|
|
107 |
ThaiAnalysisInfra::~ThaiAnalysisInfra()
|
|
108 |
{}
|
|
109 |
|
|
110 |
ThaiWordFilter::ThaiWordFilter( lucene::analysis::TokenStream* input,
|
|
111 |
bool deleteTs )
|
|
112 |
: TokenFilter(input, deleteTs),
|
|
113 |
breaks_(),
|
|
114 |
thaiToken_()
|
|
115 |
{
|
|
116 |
breaks_ = ThaiAnalysisInfra::theInstance()->createBreakIterator();
|
|
117 |
}
|
|
118 |
|
|
119 |
using namespace lucene::analysis;
|
|
120 |
|
|
121 |
ThaiWordFilter::~ThaiWordFilter()
|
|
122 |
{}
|
|
123 |
|
|
124 |
#define MAX_BUFSIZE 256
|
|
125 |
|
|
126 |
bool ThaiWordFilter::next(Token* token)
|
|
127 |
{
|
|
128 |
if ( breaks_->hasNext() )
|
|
129 |
{
|
|
130 |
size_t wordBegin = breaks_->current();
|
|
131 |
size_t wordLength = breaks_->next() - wordBegin;
|
|
132 |
|
|
133 |
wchar_t buf[MAX_BUFSIZE];
|
|
134 |
memcpy( buf,
|
|
135 |
thaiToken_.termText()+wordBegin,
|
|
136 |
wordLength * sizeof(wchar_t) );
|
|
137 |
buf[wordLength] = '\0';
|
|
138 |
|
|
139 |
token->set( buf,
|
|
140 |
thaiToken_.startOffset() + wordBegin,
|
|
141 |
thaiToken_.endOffset() + wordBegin + wordLength);
|
|
142 |
return true;
|
|
143 |
}
|
|
144 |
|
|
145 |
if ( input->next( token ) )
|
|
146 |
{
|
|
147 |
if ( unicode::IsThai( token->termText()[0] ) )
|
|
148 |
{
|
|
149 |
thaiToken_.set( token->termText(), token->startOffset(), token->endOffset() );
|
|
150 |
breaks_->setText( thaiToken_.termText()); // reset
|
|
151 |
return next( token );
|
|
152 |
} else {
|
|
153 |
return true;
|
|
154 |
}
|
|
155 |
}
|
|
156 |
|
|
157 |
return false;
|
|
158 |
}
|
|
159 |
|
|
160 |
|
|
161 |
using namespace lucene::analysis::standard;
|
|
162 |
|
|
163 |
ThaiAnalyzer::ThaiAnalyzer()
|
|
164 |
: stopWords_(false)
|
|
165 |
{
|
|
166 |
StopFilter::fillStopTable( &stopWords_,CL_NS(analysis)::StopAnalyzer::ENGLISH_STOP_WORDS);
|
|
167 |
}
|
|
168 |
|
|
169 |
lucene::analysis::TokenStream* ThaiAnalyzer::tokenStream(const wchar_t* fieldName,
|
|
170 |
CL_NS(util)::Reader* reader)
|
|
171 |
{
|
|
172 |
auto_ptr<TokenStream> ret( new StandardTokenizer(reader) );
|
|
173 |
|
|
174 |
ret.reset( new LowerCaseFilter( ret.release(), true ) );
|
|
175 |
ret.reset( new StandardFilter( ret.release(), true ) );
|
|
176 |
ret.reset( new ThaiWordFilter( ret.release(), true ) );
|
|
177 |
ret.reset( new StopFilter( ret.release(), true, &stopWords_ ) );
|
|
178 |
|
|
179 |
return ret.release();
|
|
180 |
}
|
|
181 |
|
|
182 |
}
|