00001 /*========================================================================== 00002 * Copyright (c) 2003 Carnegie Mellon University. All Rights Reserved. 00003 * 00004 * Use of the Lemur Toolkit for Language Modeling and Information Retrieval 00005 * is subject to the terms of the software license set forth in the LICENSE 00006 * file included with this software, and also available at 00007 * http://www.lemurproject.org/license.html 00008 * 00009 *========================================================================== 00010 */ 00011 #include "TextHandler.hpp" 00012 00013 #ifndef _BRILLPOSTOK_HPP 00014 #define _BRILLPOSTOK_HPP 00015 00026 00027 class BrillPOSTokenizer : public TextHandler { 00028 00029 public: 00031 BrillPOSTokenizer(); 00032 00034 BrillPOSTokenizer(char s); 00035 00037 void setDelimiter(char s) {splitter = s;}; 00038 00041 char * handleWord(char* word, const char* original, PropertyList* list); 00042 00043 protected: 00044 char splitter; 00045 // keep setting one property. PropertyList copies. 00046 Property pos; 00047 }; 00048 00049 #endif