Index.hpp

Go to the documentation of this file.

/*==========================================================================
 * Copyright (c) 2001 Carnegie Mellon University.  All Rights Reserved.
 *
 * Use of the Lemur Toolkit for Language Modeling and Information Retrieval
 * is subject to the terms of the software license set forth in the LICENSE
 * file included with this software, and also available at
 * http://www.lemurproject.org/license.html
 *
 *==========================================================================
 *
 * 26 Sep 2007 - mjh - added call for fieldInfoList
 */


#ifndef _INDEX_HPP
#define _INDEX_HPP
// C. Zhai 02/08/2001

#include "TermInfoList.hpp"
#include "FieldInfoList.hpp"
#include "DocInfoList.hpp"
#include "DocumentManager.hpp"
#include "CollectionProps.hpp"
#include "lemur-platform.h"
#include "lemur-compat.hpp"

namespace lemur 
{
  namespace api 
  {
    

    class Index {
    public:

      virtual ~Index() {}; 



      virtual bool open(const string &indexName)=0;



      virtual TERMID_T term (const TERM_T &word) const=0;

      virtual const TERM_T term (TERMID_T termID) const=0;

      virtual DOCID_T document (const EXDOCID_T &docIDStr) const=0;

      virtual const EXDOCID_T document (DOCID_T docID) const=0;

      //  virtual const char* docManager(int docID) { return NULL;}
      virtual const DocumentManager* docManager(DOCID_T docID) const {return NULL;}

      virtual const int field(std::string fieldName) const { return 0; }
      virtual const int field(const char *fieldName) const { return 0; }

      virtual const std::string field(int fieldID) const { return ""; }


      virtual const string termLexiconID() const { return "";} 




      virtual COUNT_T docCount () const=0;

      virtual COUNT_T termCountUnique () const=0;

      virtual COUNT_T termCount (TERMID_T termID) const=0;

      virtual COUNT_T termCount () const=0;

      virtual float docLengthAvg() const=0;

      virtual COUNT_T docCount(TERMID_T termID) const=0;

      virtual COUNT_T docLength (DOCID_T docID) const=0;




      virtual DocInfoList *docInfoList(TERMID_T termID) const=0;

      virtual TermInfoList *termInfoList(DOCID_T docID) const=0;

      virtual FieldInfoList *fieldInfoList(DOCID_T docID) const { return NULL; }

      virtual FieldInfoList *fieldInfoList(DOCID_T docID, int fieldID) const { return NULL; }

      


      virtual std::vector<std::string> documentMetadata(const std::vector< lemur::api::DOCID_T > &documentIDs, const std::string &attributeName) { std::vector<std::string> _blankVector; _blankVector.clear(); return _blankVector; }

      virtual std::vector<std::string> documentMetadata(lemur::api::DOCID_T documentID, const std::string &attributeName) { std::vector<std::string> _blankVector; _blankVector.clear(); return _blankVector; }


      // returns TermInfoList is sequential representation (not bag of words)
      // return NULL list when sequence is not available.
      virtual TermInfoList *termInfoListSeq(DOCID_T docID) const { return NULL; }

      virtual const lemur::parse::CollectionProps* collectionProps() const { return NULL; }
    };
  }
}


#endif

---