trec_search.cc

   1 /* trec_search.cc: Example batch search for TREC experiments
   2  *
   3  * ----START-LICENCE----
   4  * Copyright 1999,2000,2001 BrightStation PLC
   5  * Copyright 2003 Olly Betts
   6  * Copyright 2003 Andy MacFarlane, City University
   7  *
   8  * This program is free software; you can redistribute it and/or
   9  * modify it under the terms of the GNU General Public License as
  10  * published by the Free Software Foundation; either version 2 of the
  11  * License, or (at your option) any later version.
  12  *
  13  * This program is distributed in the hope that it will be useful,
  14  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  15  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  16  * GNU General Public License for more details.
  17  *
  18  * You should have received a copy of the GNU General Public License
  19  * along with this program; if not, write to the Free Software
  20  * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307
  21  * USA
  22  * -----END-LICENCE-----
  23  */
  24
  25 #include "config_file.h"
  26 #include <fstream>
  27 #include <xapian.h>
  28 #include <algorithm>
  29 #include <iostream>
  30 #include <string>
  31 #include <time.h>
  32 #include "stopword.h"
  33 #include "split.h"
  34 #include "timer.h"
  35
  36 using namespace Xapian;
  37 using namespace std;
  38
  39 int load_query( std::ifstream & queryfile, int & topicno, SW_STORE sw_store, Xapian::Query & query, Xapian::Stem & stemmer ) {
  40 // load a query and record its terms
  41
  42         if( queryfile.eof() ) return 0;
  43
  44         int found_topicno=0;
  45         string line;
  46         getline(queryfile,line);
  47         line[line.size()-1] ='\0';
  48         vector<string> data;
  49   split(line, ' ', data );
  50         vector <string> terms;
  51         for( vector<string>::const_iterator start = data.begin(); start != data.end(); start++ ) {
  52           string queryword;
  53                 if( !found_topicno ) {
  54                         topicno = atoi( start->c_str());
  55                         found_topicno=1;
  56                 } else if(!IsStopWord( sw_store, (char *) start->c_str() )) {
  57                                         queryword = stemmer(*start);
  58                                         cout << "Queryword: = " << queryword << endl;
  59                                         terms.push_back(queryword);
  60     } // END if
  61
  62         } // END for
  63
  64         // make the query with the terms
  65         Xapian::Query consquery(Xapian::Query::OP_OR, terms.begin(), terms.end());
  66         query = consquery;
  67
  68         return 1;
  69
  70 } // END load_query
  71
  72 int main(int argc, char **argv)
  73 {
  74   // Simplest possible options parsing: we just require two or more
  75   // parameters.
  76   if(argc < 2) {
  77     cout << "usage: " << argv[0] << " <config file>" << endl;
  78     exit(1);
  79   }
  80
  81   // Catch any Xapian::Error exceptions thrown
  82   try {
  83     // load the TREC experiment configuration file
  84     CONFIG_TREC config;
  85     config.setup_config( string(argv[1]) );
  86     config.check_search_config();
  87     Xapian::Stem stemmer( config.get_language() );
  88     struct timeval start_time, finish_time, timelapse;   /* timing variables */
  89
  90     // Make the database
  91     Xapian::Database db(Xapian::Flint::open(config.get_db().c_str()));
  92
  93     // Start an enquire session
  94     Xapian::Enquire enquire(db);
  95
  96     // open the query file
  97     std::ifstream queryfile( config.get_queryfile().c_str() );
  98
  99     // open the results file
 100     std::ofstream resultsfile( config.get_resultsfile().c_str() );
 101
 102     // open the transaction file
 103     std::ofstream transfile( config.get_transfile().c_str() );
 104
 105     // load the stop word list
 106     SW_STORE sw_store;
 107     Read_SW_File( (char *) config.get_stopsfile().c_str(), &sw_store );
 108
 109     // count of no queries done
 110     int count=0;
 111
 112     // total query time
 113     float total_qp_time = 0.0;
 114
 115     // process the queries
 116     while( !queryfile.eof() ) {
 117
 118       int topicno;  // topic number for the query
 119       int len=0;
 120
 121       // Build the query object
 122       Xapian::Query query;
 123       int gotquery = load_query( queryfile, topicno, sw_store, query, stemmer );
 124
 125       if(gotquery && !queryfile.eof()) {
 126
 127         // start the timer
 128         gettimeofday( &start_time, 0 );
 129
 130         cout << "Running " << topicno << ", query = [" << query.get_description() << "] getting " << config.get_noresults() << " docs" << endl;
 131
 132         // Give the query object to the enquire session
 133         enquire.set_query(query);
 134
 135         // Get the top n results of the query
 136         Xapian::MSet matches = enquire.get_mset( 0, config.get_noresults() );
 137
 138         // record the number of matches made in this query
 139         //int queryweightings = enquire.get_totalweightings();
 140         //cout << "W's) for this query is -> " << queryweightings << endl;
 141
 142         // Display the results cout << matches.size() << " results found" << endl;
 143         count++;
 144         if( (count % 1000) == 0 ) cout << "QUERIES PROCESSED) " << count << endl;
 145
 146         // record the results in a 'trec.log' file
 147         for (Xapian::MSetIterator i = matches.begin(); i != matches.end(); i++) {
 148           Xapian::Document doc = i.get_document();
 149           resultsfile << topicno << " Q0 " << doc.get_data() << " " << i.get_rank() << " " <<
 150             i.get_weight() << " " << config.get_runname() << endl;
 151           len++;
 152         } // END for
 153
 154         // record the finish
 155         gettimeofday( &finish_time, 0 );
 156         diff_time( finish_time, start_time, &timelapse );
 157         float qp_time = time_real( timelapse );
 158         total_qp_time += qp_time;
 159         transfile << topicno << "," << qp_time << "," << len << endl;
 160         cout << topicno << "," << qp_time << "," << len << endl;
 161
 162
 163       } // END if
 164     } // END while
 165
 166     // print the total time, and average time per query
 167
 168     float avg_qp_time = total_qp_time /(float) count;
 169     cout << "Average query time for " << count << " Queries is " <<
 170       avg_qp_time << " secs, took a total of " << total_qp_time << " secs" << endl;
 171
 172   } catch( const Xapian::Error &error) {
 173     cout << "Exception: "  << error.get_msg() << endl;
 174   } // END try/catch block
 175
 176 } // END main