SEXP setScoringRules(string method, string parameters){
vector<string> scoringRules;
//if(method == "tfidf" || method == "Okapi" || method == "BM25" ){
// string rule = method + "," + parameters;
//environment.setBaseline(rule);
//scoringRules.push_back("");
//environment.setScoringRules(scoringRules);
//}else{
string rule = "method:" + method + "," + parameters;
scoringRules.push_back(rule);
environment.setScoringRules(scoringRules);
//}
return R_NilValue;
}
static void open_indexes( indri::api::QueryEnvironment& environment,
indri::api::Parameters& param ) {
if( param.exists( "index" ) ) {
indri::api::Parameters indexes = param["index"];
for( unsigned int i=0; i < indexes.size(); i++ ) {
environment.addIndex( std::string(indexes[i]) );
}
}
if( param.exists( "server" ) ) {
indri::api::Parameters servers = param["server"];
for( unsigned int i=0; i < servers.size(); i++ ) {
environment.addServer( std::string(servers[i]) );
}
}
std::vector<std::string> smoothingRules;
if( copy_parameters_to_string_vector( smoothingRules, param, "rule" ) )
environment.setScoringRules( smoothingRules );
}
UINT64 initialize() {
_environment.setSingleBackgroundModel( _parameters.get("singleBackgroundModel", false) );
std::vector<std::string> stopwords;
if( copy_parameters_to_string_vector( stopwords, _parameters, "stopper.word" ) )
_environment.setStopwords(stopwords);
std::vector<std::string> smoothingRules;
if( copy_parameters_to_string_vector( smoothingRules, _parameters, "rule" ) )
_environment.setScoringRules( smoothingRules );
if( _parameters.exists( "index" ) ) {
indri::api::Parameters indexes = _parameters["index"];
for( size_t i=0; i < indexes.size(); i++ ) {
_environment.addIndex( std::string(indexes[i]) );
}
}
if( _parameters.exists( "server" ) ) {
indri::api::Parameters servers = _parameters["server"];
for( size_t i=0; i < servers.size(); i++ ) {
_environment.addServer( std::string(servers[i]) );
}
}
if( _parameters.exists("maxWildcardTerms") )
_environment.setMaxWildcardTerms(_parameters.get("maxWildcardTerms", 100));
_requested = _parameters.get( "count", 1000 );
_initialRequested = _parameters.get( "fbDocs", _requested );
_runID = _parameters.get( "runID", "indri" );
_trecFormat = _parameters.get( "trecFormat" , false );
_inexFormat = _parameters.exists( "inex" );
_printQuery = _parameters.get( "printQuery", false );
_printDocuments = _parameters.get( "printDocuments", false );
_printPassages = _parameters.get( "printPassages", false );
_printSnippets = _parameters.get( "printSnippets", false );
if (_parameters.exists("baseline")) {
// doing a baseline
std::string baseline = _parameters["baseline"];
_environment.setBaseline(baseline);
// need a factory for this...
if( _parameters.get( "fbDocs", 0 ) != 0 ) {
// have to push the method in...
std::string rule = "method:" + baseline;
_parameters.set("rule", rule);
_expander = new indri::query::TFIDFExpander( &_environment, _parameters );
}
} else {
if( _parameters.get( "fbDocs", 0 ) != 0 ) {
_expander = new indri::query::RMExpander( &_environment, _parameters );
}
}
if (_parameters.exists("maxWildcardTerms")) {
_environment.setMaxWildcardTerms((int)_parameters.get("maxWildcardTerms"));
}
return 0;
}
UINT64 initialize() {
try {
_environment.setSingleBackgroundModel( _parameters.get("singleBackgroundModel", false) );
std::vector<std::string> stopwords;
if( copy_parameters_to_string_vector( stopwords, _parameters, "stopper.word" ) )
_environment.setStopwords(stopwords);
std::vector<std::string> smoothingRules;
if( copy_parameters_to_string_vector( smoothingRules, _parameters, "rule" ) )
_environment.setScoringRules( smoothingRules );
if( _parameters.exists( "index" ) ) {
indri::api::Parameters indexes = _parameters["index"];
for( size_t i=0; i < indexes.size(); i++ ) {
_environment.addIndex( std::string(indexes[i]) );
}
}
if( _parameters.exists( "server" ) ) {
indri::api::Parameters servers = _parameters["server"];
for( size_t i=0; i < servers.size(); i++ ) {
_environment.addServer( std::string(servers[i]) );
}
}
if( _parameters.exists("maxWildcardTerms") )
_environment.setMaxWildcardTerms(_parameters.get("maxWildcardTerms", 100));
_requested = _parameters.get( "count", 1000 );
_initialRequested = _parameters.get( "fbDocs", _requested );
_runID = _parameters.get( "runID", "indri" );
_trecFormat = _parameters.get( "trecFormat" , false );
_inexFormat = _parameters.exists( "inex" );
_printQuery = _parameters.get( "printQuery", false );
_printDocuments = _parameters.get( "printDocuments", false );
_printPassages = _parameters.get( "printPassages", false );
_printSnippets = _parameters.get( "printSnippets", false );
if (_parameters.exists("baseline")) {
// doing a baseline
std::string baseline = _parameters["baseline"];
_environment.setBaseline(baseline);
// need a factory for this...
if( _parameters.get( "fbDocs", 0 ) != 0 ) {
// have to push the method in...
std::string rule = "method:" + baseline;
_parameters.set("rule", rule);
_expander = new indri::query::TFIDFExpander( &_environment, _parameters );
}
} else {
if( _parameters.get( "fbDocs", 0 ) != 0 ) {
_expander = new indri::query::RMExpander( &_environment, _parameters );
}
}
if (_parameters.exists("maxWildcardTerms")) {
_environment.setMaxWildcardTerms((int)_parameters.get("maxWildcardTerms"));
}
} catch ( lemur::api::Exception& e ) {
while( _queries.size() ) {
query_t *query = _queries.front();
_queries.pop();
_output.push( new query_t( query->index, query->number, "query: " + query->number + " QueryThread::_initialize exception\n" ) );
_queueEvent.notifyAll();
LEMUR_RETHROW(e, "QueryThread::_initialize");
}
}
return 0;
}