bool dataframesAreEqual(Rcpp::DataFrame &a, Rcpp::DataFrame &b) {
Rcpp::NumericVector a1 = a["a"];
Rcpp::CharacterVector b1 = a["b"];
Rcpp::DoubleVector c1 = a["c"];
Rcpp::NumericVector a2 = b["a"];
Rcpp::CharacterVector b2 = b["b"];
Rcpp::DoubleVector c2 = b["c"];
if(a.size() != b.size()) {
DLOG(INFO) << "sizes different";
return false;
}
for(int i = 0; i < a1.size(); i++) {
if(a1[i] != a2[i]) {
DLOG(INFO) << a1[i] << " vs " << a2[i];
return false;
}
if(b1[i] != b2[i]) {
DLOG(INFO) << b1[i] << " vs " << b2[i];
return false;
}
if(c1[i] != c2[i]) {
DLOG(INFO) << c1[i] << " vs " << c2[i];
return false;
}
}
return true;
}
Interaction::Interaction(vector<double> const& instance,vector<bool> const& factor,
Rcpp::DataFrame const& class_data,double epsilon_cont,double epsilon_cat,int p_depth,int nb_class,double p_radius){
vector<InteractionItem> new_items(class_data.size());
Rcpp::CharacterVector names = class_data.names();
for(int i=0;i<class_data.size();++i){
vector<double> attr_data = Rcpp::as<vector<double> >(class_data[i]);
string name = Rcpp::as<string>(names[i]);
if(factor[i]){
InteractionItem item(name,instance[i],i,epsilon_cat,factor[i],attr_data);
new_items[i] = item;
}
else{
InteractionItem item(name,instance[i],i,epsilon_cont,factor[i],attr_data);
new_items[i] = item;
}
}
items = new_items;
depth = p_depth;
sims_valid = false;
vector<double> new_sims(nb_class);
sims = new_sims;
radius = p_radius;
}
vector<double> random_instance(Rcpp::DataFrame const& x, Generator& prng){
// Select random row number
int row = prng();
vector<double> instance(x.size());
for(int i=0;i<x.size();++i){
vector<double> col = Rcpp::as<vector<double> >(x[i]);
instance[i] = col[row];
}
return instance;
}
Rcpp::StringMatrix DataFrame_to_StringMatrix( Rcpp::DataFrame df ){
Rcpp::StringVector sv = df(0);
Rcpp::StringMatrix sm(sv.size(), df.size());
sm.attr("col.names") = df.attr("col.names");
sm.attr("row.names") = df.attr("row.names");
for(int i=0; i < df.size(); i++){
sv = df(i);
for(int j=0; j < sv.size(); j++){
sm(j, i) = sv(j);
}
}
return sm;
}
void helper(const std::string& url,
const bool dropCaches,
const std::string& objType) {
if(dropCaches) {
LOG(INFO) << "Dropping caches ...";
int res, res2;
if ((res = system ("sync")) == -1) {
throw std::runtime_error("Error running sync");
}
if ((res2 = system ("sudo sh -c 'echo 3 > /proc/sys/vm/drop_caches'")) == -1) {
throw std::runtime_error("Error dropping caches");
}
}
//std::string extension = base::utils::getExtension(url);
std::string protocol = base::utils::getProtocol(url);
std::string filename = base::utils::stripProtocol(url);
base::IFilePtr file = hdfsutils::FileFactory::makeFile(protocol, filename);
if( hdfsutils::isHdfs(protocol)) {
hdfsutils::HdfsFile *p = (hdfsutils::HdfsFile *)file.get();
base::ConfigurationMap hdfsconf;
hdfsconf["hdfsConfigurationFile"] = std::string("../ddc/test/data/server.conf");
p->configure(hdfsconf);
}
base::FileStatus status = file->stat();
base::ConfigurationMap conf;
std::string schema = "000:int64,"
"001:int64,002:int64,003:int64,004:int64,005:int64,006:int64,007:int64,008:int64,"
"009:int64,010:int64,011:int64,012:int64,013:int64,014:int64,015:int64,016:int64,"
"017:int64,018:int64,019:int64,020:int64,021:int64,022:int64,023:int64,024:int64,"
"025:int64,026:int64,027:int64,028:int64,029:int64,030:int64,031:int64,032:int64,"
"033:int64,034:int64,035:int64,036:int64,037:int64,038:int64,039:int64,040:int64,"
"041:int64,042:int64,043:int64,044:int64,045:int64,046:int64,047:int64,048:int64,"
"049:int64,050:int64,051:int64,052:int64,053:int64,054:int64,055:int64,056:int64,"
"057:int64,058:int64,059:int64,060:int64,061:int64,062:int64,063:int64,064:int64,"
"065:int64,066:int64,067:int64,068:int64,069:int64,070:int64,071:int64,072:int64,"
"073:int64,074:int64,075:int64,076:int64,077:int64,078:int64,079:int64,080:int64,"
"081:int64,082:int64,083:int64,084:int64,085:int64,086:int64,087:int64,088:int64,"
"089:int64,090:int64,091:int64,092:int64,093:int64,094:int64,095:int64,096:int64,"
"097:int64,098:int64,099:int64,100:int64,101:int64,102:int64,103:int64,104:int64,"
"105:int64,106:int64,107:int64,108:int64,109:int64,110:int64,111:int64,112:int64,"
"113:int64,114:int64,115:int64,116:int64,117:int64,118:int64,119:int64,120:int64,"
"121:int64,122:int64,123:int64,124:int64,125:int64,126:int64,127:int64";
conf["schemaUrl"] = schema;
conf["chunkStart"] = (uint64_t)0;
conf["chunkEnd"] = (uint64_t)status.length;
conf["hdfsConfigurationFile"] = std::string("../ddc/test/data/server.conf") ;
boost::shared_ptr<Rcpp::DataFrame> dfptr =
boost::any_cast<boost::shared_ptr<Rcpp::DataFrame>>(ddc_read(url,
objType,
conf));
Rcpp::DataFrame df = Rcpp::DataFrame(*(dfptr.get()));
LOG(INFO) << "Got dataframe of size " << df.size();
}
// [[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;
}