LemurIndriIndex.hpp

Go to the documentation of this file.

/*==========================================================================
 * Copyright (c) 2004 University of Massachusetts.  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 calls for field(...) and fieldInfoList(...)
 */



#ifndef LEMURINDRI_INDEX_HPP
#define LEMURINDRI_INDEX_HPP

/*
 * NAME DATE - COMMENTS
 * dmf 09/04 - Lemur Index API wrapper on Indri Repository
 */

#include "Index.hpp"
#include "indri/Repository.hpp"

namespace lemur
{
  namespace index 
  {
    
    class LemurIndriIndex : public lemur::api::Index {
    public:
      LemurIndriIndex();

      virtual ~LemurIndriIndex();



      bool open(const std::string& indexName);

      bool open( const char* indexName ) {
        return open(string(indexName));
      }
      void close();



      lemur::api::TERMID_T term(const lemur::api::TERM_T &word) const;

      const lemur::api::TERM_T term(lemur::api::TERMID_T termID) const;

      lemur::api::DOCID_T document(const lemur::api::EXDOCID_T &docIDStr) const;

      const lemur::api::EXDOCID_T document(lemur::api::DOCID_T docID) const;




      lemur::api::COUNT_T docCount() const;

      lemur::api::COUNT_T termCountUnique() const;

      lemur::api::COUNT_T termCount(lemur::api::TERMID_T termID) const;

      lemur::api::COUNT_T termCount() const;

      float docLengthAvg() const;

      lemur::api::COUNT_T docCount(lemur::api::TERMID_T termID) const;

      lemur::api::COUNT_T docLength( lemur::api::DOCID_T documentID ) const;


      lemur::api::DocInfoList* docInfoList(lemur::api::TERMID_T termID) const;

      lemur::api::TermInfoList* termInfoList(lemur::api::DOCID_T docID) const;

      lemur::api::TermInfoList* termInfoListSeq(lemur::api::DOCID_T docID) const;

      const lemur::api::DocumentManager* docManager(lemur::api::DOCID_T docID) const {
        return _docMgr;
      }
  
      const string &getRepositoryName() const {return _repositoryName;}

      const int field(std::string fieldName) const;

      const int field(const char *fieldName) const;

      const std::string field(int fieldID) const;

      virtual lemur::api::FieldInfoList *fieldInfoList(lemur::api::DOCID_T docID) const;

      virtual lemur::api::FieldInfoList *fieldInfoList(lemur::api::
        DOCID_T docID, int fieldID) const;

      virtual std::vector<std::string> documentMetadata(const std::vector< lemur::api::DOCID_T > &documentIDs, const std::string &attributeName);

      virtual std::vector<std::string> documentMetadata(lemur::api::DOCID_T documentID, const std::string &attributeName);

  
    private:
      indri::index::Index* _indexWithDocument( indri::collection::Repository::index_state& indexes, lemur::api::DOCID_T documentID ) const;
      indri::collection::Repository *_repository;
      string _repositoryName;
      lemur::api::DocumentManager *_docMgr;
    };
  }
}

#endif // LEMURINDRI_INDEX_HPP

---