Lucene++ - a full-featured, c++ search engine
API Documentation


WhitespaceTokenizer.h
Go to the documentation of this file.
1 // Copyright (c) 2009-2014 Alan Wright. All rights reserved.
3 // Distributable under the terms of either the Apache License (Version 2.0)
4 // or the GNU Lesser General Public License.
6 
7 #ifndef WHITESPACETOKENIZER_H
8 #define WHITESPACETOKENIZER_H
9 
10 #include "CharTokenizer.h"
11 
12 namespace Lucene {
13 
16 class LPPAPI WhitespaceTokenizer : public CharTokenizer {
17 public:
19  WhitespaceTokenizer(const ReaderPtr& input);
20 
22  WhitespaceTokenizer(const AttributeSourcePtr& source, const ReaderPtr& input);
23 
25  WhitespaceTokenizer(const AttributeFactoryPtr& factory, const ReaderPtr& input);
26 
27  virtual ~WhitespaceTokenizer();
28 
30 
31 public:
33  virtual bool isTokenChar(wchar_t c);
34 };
35 
36 }
37 
38 #endif
LUCENE_CLASS
#define LUCENE_CLASS(Name)
Definition: LuceneObject.h:24
Lucene::ReaderPtr
boost::shared_ptr< Reader > ReaderPtr
Definition: LuceneTypes.h:547
Lucene::CharTokenizer
An abstract base class for simple, character-oriented tokenizers.
Definition: CharTokenizer.h:15
CharTokenizer.h
Lucene
Definition: AbstractAllTermDocs.h:12
Lucene::WhitespaceTokenizer
A WhitespaceTokenizer is a tokenizer that divides text at whitespace. Adjacent sequences of non-White...
Definition: WhitespaceTokenizer.h:16
Lucene::AttributeSourcePtr
boost::shared_ptr< AttributeSource > AttributeSourcePtr
Definition: LuceneTypes.h:520
Lucene::AttributeFactoryPtr
boost::shared_ptr< AttributeFactory > AttributeFactoryPtr
Definition: LuceneTypes.h:519

clucene.sourceforge.net