InvFPTermList.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
 *
 *==========================================================================
 */

/*
  10/18/2002 -- dmf Add binReadC, binWriteC, deltaDecode, and deltEncode
  for compression of TermInfoLists.
*/

#ifndef _INVFPTERMLIST_HPP
#define _INVFPTERMLIST_HPP

#include "common_headers.hpp"
#include "InvFPTerm.hpp"
#include "InvFPTypes.hpp"
#include "RVLCompress.hpp"

namespace lemur 
{
  namespace file 
  {
    class File;
  }
}

namespace lemur 
{
  namespace index 
  {
    class InvFPTermList : public lemur::api::TermInfoList {
    public:
      InvFPTermList();
      InvFPTermList(lemur::api::DOCID_T did, int len, vector<LocatedTerm> &tls);
      ~InvFPTermList();

      void startIteration() const;

      bool hasMore() const;

      lemur::api::TermInfo *nextEntry() const;

      virtual int size();

      virtual lemur::api::TermInfo* operator[](int index) const;

      lemur::api::COUNT_T docLength() const{ return length; }

      lemur::api::COUNT_T termCount()  const{ return listlen; }

      lemur::api::DOCID_T docID()  const{ return uid; }

      bool binRead(ifstream& infile);
      bool binReadC(ifstream& infile);
      void binWriteC(ofstream& ofile);

      bool binReadC( lemur::file::File& infile );
      void binWriteC( lemur::file::File& outfile );

      virtual void deltaDecode();
      virtual void deltaEncode();

      void countTerms();

    protected:
      // Helper functions for iterator, subclasses should override
      virtual lemur::api::TermInfo* newElement() const { return new InvFPTerm(); }
      virtual lemur::api::TermInfo* getElement(lemur::api::TermInfo* elem, lemur::api::POS_T position) const;
      virtual void assignElement(lemur::api::TermInfo* to, lemur::api::TermInfo* from) const {
        *static_cast<InvFPTerm*>(to) = *static_cast<InvFPTerm*>(from);
      }
      virtual lemur::api::POS_T beginPosition() const { return (lemur::api::POS_T) 0; }
      virtual lemur::api::POS_T endPosition() const { return (lemur::api::POS_T) listlen; }
      virtual lemur::api::POS_T nextPosition(lemur::api::POS_T position) const;

      lemur::api::DOCID_T uid; // this doc's id
      lemur::api::COUNT_T length;  // length of this document (terms + stopwords)
      LocatedTerm* list; // list of terms and locations
      LLTerm* listcounted; // list of terms and location lists
      lemur::api::COUNT_T listlen; // number of items we have in list  (same as number of terms)
      mutable int index;   // index for iterator
      lemur::api::LOC_T* counts; // keep track of counts of terms for bag of word
      mutable InvFPTerm entry;
      mutable vector<lemur::api::LOC_T> loclist; //list of locations to return

    };
  }
}
#endif

---