---

QryBasedSampler.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.cs.cmu.edu/~lemur/license.html
 *
 *==========================================================================
*/

#ifndef _QRYBASEDSAMPLER_HPP
#define _QRYBASEDSAMPLER_HPP



#include "FreqCounter.hpp"
#include "DBManager.hpp"

typedef stringset docidset;

#define T_NDOCS 1

#define T_NWORDS 2

#define T_NQRYS 4

class QryBasedSampler {
public:
  QryBasedSampler();
  ~QryBasedSampler();

  bool probe(char * initQuery);
  
  void setDBManager(DBManager * database);

  DBManager * getDBManager();


  void setFreqCounter(FreqCounter * counter);

  FreqCounter * getFreqCounter();


  void setOutputPrefix(char * prefix);
  
  char * getOutputPrefix();

  void setNumDocs(int n);

  int getNumDocs();


  void setNumWords(int n);

  int getNumWords();


  void setNumQueries(int n);

  int getNumQueries();


  void setTermMode(int m);

  int getTermMode();
  

  void setDocsPerQuery(int n);
  
  int getDocsPerQuery();


private:

  /* for querying a db */
  DBManager * db;


  /* for building a description of a db */
  FreqCounter * freqCounter;


  /* output prefix for filenames */
  char * outputPrefix;


  /* termination mode of the probe -
   * either T_NDOCS or T_NWORDS */
  int termMode;
  
  /* number unique docs to retrieve - only used if
   * termMode == T_NDOCS */
  int numDocs;

  /* number unique words to retrieve - only used if
   * termMode == T_NWORDS */
  int numWords;

  /* number of queries to run - only used if
   * termMode == T_NQRYS */
  int numQueries;

  /* documents per query to use */
  int docsPerQuery;

  /* stores the ids of the document already retrieved
   * from the system.  used to prevent parsing
   * a document multiple times */
  docidset seenDocs;
};

#endif

---