/**
* Set up a multi-column temp output table for those executors that require one.
* Called from p_init.
*/
void AbstractExecutor::setTempOutputTable(const ExecutorVector& executorVector,
const string tempTableName) {
TupleSchema* schema = m_abstractNode->generateTupleSchema();
int column_count = schema->columnCount();
std::vector<std::string> column_names(column_count);
assert(column_count >= 1);
const std::vector<SchemaColumn*>& outputSchema = m_abstractNode->getOutputSchema();
for (int ctr = 0; ctr < column_count; ctr++) {
column_names[ctr] = outputSchema[ctr]->getColumnName();
}
if (executorVector.isLargeQuery()) {
m_tmpOutputTable = TableFactory::buildLargeTempTable(tempTableName,
schema,
column_names);
}
else {
m_tmpOutputTable = TableFactory::buildTempTable(tempTableName,
schema,
column_names,
executorVector.limits());
}
m_abstractNode->setOutputTable(m_tmpOutputTable);
}
std::shared_ptr<const Table> ShowColumns::_on_execute() {
TableColumnDefinitions column_definitions;
column_definitions.emplace_back("column_name", DataType::String);
column_definitions.emplace_back("column_type", DataType::String);
column_definitions.emplace_back("is_nullable", DataType::Int);
auto out_table = std::make_shared<Table>(column_definitions, TableType::Data);
const auto table = StorageManager::get().get_table(_table_name);
Segments segments;
const auto& column_names = table->column_names();
const auto vs_names = std::make_shared<ValueSegment<pmr_string>>(
tbb::concurrent_vector<pmr_string>(column_names.begin(), column_names.end()));
segments.push_back(vs_names);
const auto& column_types = table->column_data_types();
auto column_types_as_string = tbb::concurrent_vector<pmr_string>{};
for (const auto column_type : column_types) {
column_types_as_string.push_back(pmr_string{data_type_to_string.left.at(column_type)});
}
const auto vs_types = std::make_shared<ValueSegment<pmr_string>>(std::move(column_types_as_string));
segments.push_back(vs_types);
const auto& column_nullables = table->columns_are_nullable();
const auto vs_nullables = std::make_shared<ValueSegment<int32_t>>(
tbb::concurrent_vector<int32_t>(column_nullables.begin(), column_nullables.end()));
segments.push_back(vs_nullables);
out_table->append_chunk(segments);
return out_table;
}
std::vector<std::string> SpatiaLiteDB::geometryTables() {
std::vector<std::string> all_tables = table_names();
// find all tables containing a geometry column
std::vector<std::string> geometry_tables;
for (std::vector<std::string>::iterator table = all_tables.begin();
table != all_tables.end(); table++) {
std::vector<std::string> columns = column_names(*table);
for (std::vector<std::string>::iterator column = columns.begin();
column != columns.end(); column++) {
std::transform(column->begin(), column->end(), column->begin(),
(int(*)(int))std::tolower);
if (!column->compare("geometry")) {
geometry_tables.push_back(*table);
}
}
}
return geometry_tables;
}
// [[Rcpp::export]]
void write_vcf_body_gz( Rcpp::DataFrame fix, Rcpp::DataFrame gt, std::string filename , int mask=0 ) {
// http://stackoverflow.com/a/5649224
// fix DataFrame
Rcpp::StringVector chrom = fix["CHROM"];
Rcpp::StringVector pos = fix["POS"];
Rcpp::StringVector id = fix["ID"];
Rcpp::StringVector ref = fix["REF"];
Rcpp::StringVector alt = fix["ALT"];
Rcpp::StringVector qual = fix["QUAL"];
Rcpp::StringVector filter = fix["FILTER"];
Rcpp::StringVector info = fix["INFO"];
// gt DataFrame
Rcpp::StringMatrix gt_cm = DataFrame_to_StringMatrix(gt);
Rcpp::StringVector column_names(gt.size());
column_names = gt.attr("names");
int i = 0;
int j = 0;
gzFile *fi = (gzFile *)gzopen(filename.c_str(),"ab");
// gzFile *fi = (gzFile *)gzopen(filename.c_str(),"abw");
for(i=0; i<chrom.size(); i++){
Rcpp::checkUserInterrupt();
if(mask == 1 && filter(i) != "PASS" ){
// Don't print variant.
} else {
std::string tmpstring;
tmpstring = chrom(i);
tmpstring = tmpstring + "\t" + pos(i) + "\t";
if(id(i) == NA_STRING){
tmpstring = tmpstring + ".";
} else {
tmpstring = tmpstring + id(i);
}
tmpstring = tmpstring + "\t" + ref(i) + "\t" + alt(i) + "\t";
if(qual(i) == NA_STRING){
tmpstring = tmpstring + "." + "\t";
} else {
tmpstring = tmpstring + qual(i) + "\t";
}
if(filter(i) == NA_STRING){
tmpstring = tmpstring + "." + "\t";
} else {
tmpstring = tmpstring + filter(i) + "\t";
}
tmpstring = tmpstring + info(i);
// gt portion
for(j=0; j<column_names.size(); j++){
if(gt_cm(i, j) == NA_STRING){
tmpstring = tmpstring + "\t" + "./.";
} else {
tmpstring = tmpstring + "\t" + gt_cm(i, j);
}
}
// gzwrite(fi,"my decompressed data",strlen("my decompressed data"));
// gzwrite(fi,"\n",strlen("\n"));
// std::string tmpstring = "test string\n";
gzwrite(fi, (char *)tmpstring.c_str(), tmpstring.size());
gzwrite(fi,"\n",strlen("\n"));
}
}
gzclose(fi);
return;
}
// [[Rcpp::export]]
void write_vcf_body( Rcpp::DataFrame fix, Rcpp::DataFrame gt, std::string filename , int mask=0 ) {
//int write_vcf_body( Rcpp::DataFrame fix, Rcpp::DataFrame gt, std::string filename , int mask=0 ) {
// fix DataFrame
Rcpp::StringVector chrom = fix["CHROM"];
Rcpp::StringVector pos = fix["POS"];
Rcpp::StringVector id = fix["ID"];
Rcpp::StringVector ref = fix["REF"];
Rcpp::StringVector alt = fix["ALT"];
Rcpp::StringVector qual = fix["QUAL"];
Rcpp::StringVector filter = fix["FILTER"];
Rcpp::StringVector info = fix["INFO"];
// gt DataFrame
Rcpp::StringMatrix gt_cm = DataFrame_to_StringMatrix(gt);
Rcpp::StringVector column_names(gt.size());
column_names = gt.attr("names");
// column_names = gt_cm.attr("col.names");
// delete gt;
int i = 0;
int j = 0;
// Uncompressed.
std::ofstream myfile;
myfile.open (filename.c_str(), std::ios::out | std::ios::app | std::ios::binary);
// gzFile *fi = (gzFile *)gzopen("file.gz","wb");
for(i=0; i<chrom.size(); i++){
Rcpp::checkUserInterrupt();
if(mask == 1 && filter(i) == "PASS" ){
// Don't print variant.
} else {
myfile << chrom(i);
myfile << "\t";
myfile << pos(i);
myfile << "\t";
if(id(i) == NA_STRING){
myfile << ".";
myfile << "\t";
} else {
myfile << id(i);
myfile << "\t";
}
myfile << ref(i);
myfile << "\t";
myfile << alt(i);
myfile << "\t";
if(qual(i) == NA_STRING){
myfile << ".";
myfile << "\t";
} else {
myfile << qual(i);
myfile << "\t";
}
if(filter(i) == NA_STRING){
myfile << ".";
myfile << "\t";
} else {
myfile << filter(i);
myfile << "\t";
}
if(info(i) == NA_STRING){
myfile << ".";
myfile << "\t";
} else {
myfile << info(i);
}
// gt region.
myfile << "\t";
myfile << gt_cm(i, 0);
for(j=1; j<column_names.size(); j++){
myfile << "\t";
myfile << gt_cm(i, j);
}
myfile << "\n";
}
}
myfile.close();
return;
}
size_t gl_gframe::num_columns() const { return column_names().size(); }
