void matIR::QueryStats::init(const std::string& query, indri::api::QueryEnvironment& environment)
{
// Extract only the terms from the query and add to the vector
indri::api::QueryParserWrapper *parser = indri::api::QueryParserFactory::get(query, "indri");
indri::lang::ScoredExtentNode* rootNode = parser->query();
indri::lang::RawScorerNodeExtractor extractor;
rootNode->walk(extractor);
std::vector<indri::lang::RawScorerNode*>& scorerNodes = extractor.getScorerNodes();
for (int i = 0; i < scorerNodes.size(); i++){
std::string qterm = environment.stemTerm(scorerNodes[i]->queryText());
queryString.push_back(qterm);
if(environment.stemCount(qterm) == 0)
continue;
if( _queryTokens.find(qterm) == _queryTokens.end() )
_queryTokens.insert(make_pair( qterm, 1));
else
_queryTokens[qterm] += 1;
}
// Initialize vectors
_query_collectionFrequency.set_size(_queryTokens.size());
_query_documentFrequency.set_size(_queryTokens.size());
// Now obtain the statistics
int i = 0;
map<std::string, int>::const_iterator iter;
for (iter=_queryTokens.begin(); iter != _queryTokens.end(); ++iter) {
std::string stem = environment.stemTerm(iter->first);
_query_collectionFrequency(i) = (double) environment.stemCount(stem);
_query_documentFrequency(i) = (double) environment.documentStemCount(stem);
++i;
}
}
void updateQueryDetails(indri::api::QueryEnvironment& environment,
Results& resultData,
string query){
indri::api::QueryParserWrapper *parser = indri::api::QueryParserFactory::get(query, "indri");
indri::lang::ScoredExtentNode* rootNode = parser->query();
indri::lang::RawScorerNodeExtractor extractor;
rootNode->walk(extractor);
vector<indri::lang::RawScorerNode*>& scorerNodes = extractor.getScorerNodes();
for (int i = 0; i < scorerNodes.size(); i++){
string qterm = environment.stemTerm(scorerNodes[i]->queryText());
if(environment.stemCount(qterm) == 0)
continue;
if( resultData.queryStems.find(qterm) == resultData.queryStems.end() ){
resultData.queryStems.insert(make_pair( qterm, 1));
resultData.queryStemOrder.push_back(qterm);
}
else
resultData.queryStems[qterm] += 1;
}
}
SEXP stemTerm(string _term) {
return Rcpp::wrap(environment.stemTerm(_term));
}