Rcpp::CharacterVector parse_gb_qualifiers(
const std::vector<std::string>& qualifiers )
{
int begin_, end_;
std::vector<std::string> names, values;
names.reserve( qualifiers.size() );
values.reserve( qualifiers.size() );
std::vector<std::string>::const_iterator s_it = std::begin(qualifiers);
for ( ; s_it != std::end( qualifiers ); s_it++ ) {
begin_ = s_it->find_first_not_of("=");
end_ = s_it->find_first_of( "=", begin_ );
names.push_back( s_it->substr( begin_, end_ ) );
values.push_back( trim( s_it->substr(end_ + 1, s_it->length() - end_), "\"") );
}
Rcpp::CharacterVector Values = Rcpp::CharacterVector( std::begin(values), std::end(values) );
Values.names() = names;
return Values;
}
//[[Rcpp::export]]
Rcpp::List checkTreeCpp(Rcpp::S4 obj, Rcpp::List opts) {
std::string err, wrn;
Rcpp::IntegerMatrix ed = obj.slot("edge");
int nrow = ed.nrow();
Rcpp::IntegerVector ances = getAnces(ed);
//Rcpp::IntegerVector desc = getDesc(ed);
int nroots = nRoots(ances);
bool rooted = nroots > 0;
Rcpp::NumericVector edLength = obj.slot("edge.length");
Rcpp::CharacterVector edLengthNm = edLength.names();
Rcpp::CharacterVector label = obj.slot("label");
Rcpp::CharacterVector labelNm = label.names();
Rcpp::CharacterVector edLabel = obj.slot("edge.label");
Rcpp::CharacterVector edLabelNm = edLabel.names();
Rcpp::IntegerVector allnodesSafe = getAllNodesSafe(ed);
Rcpp::IntegerVector allnodesFast = getAllNodesFast(ed, rooted);
int nEdLength = edLength.size();
int nLabel = label.size();
int nEdLabel = edLabel.size();
int nEdges = nrow;
bool hasEdgeLength = !all_naC(edLength);
// check tips
int ntipsSafe = nTipsSafe(ances);
int ntipsFast = nTipsFastCpp(ances);
bool testnTips = ntipsFast == ntipsSafe;
if (! testnTips) {
err.append("Tips incorrectly labeled. ");
}
//check internal nodes
bool testNodes = Rcpp::all(allnodesSafe == allnodesFast).is_true() && // is both ways comparison needed?
Rcpp::all(allnodesFast == allnodesSafe).is_true();
if (! testNodes) {
err.append("Nodes incorrectly labeled. ");
}
// check edge lengths
if (hasEdgeLength) {
if (nEdLength != nEdges) {
err.append("Number of edge lengths do not match number of edges. ");
}
// if (nb_naC(edLength) > nroots) { // not enough! -- best done in R
// err.append("Only the root should have NA as an edge length. ");
// }
if (getRange(edLength, TRUE)[0] < 0) {
err.append("Edge lengths must be non-negative. ");
}
Rcpp::CharacterVector edgeLblSupp = edgeIdCpp(ed, "all");
Rcpp::CharacterVector edgeLblDiff = Rcpp::setdiff(edLengthNm, edgeLblSupp);
if ( edgeLblDiff.size() != 0 ) {
err.append("Edge lengths incorrectly labeled. ");
}
}
// check label names
Rcpp::CharacterVector chrLabelNm = Rcpp::as<Rcpp::CharacterVector>(allnodesFast);
int j = 0;
while (j < nroots) { //remove root(s)
chrLabelNm.erase(0);
j++;
}
bool testLabelNm = isLabelName(labelNm, chrLabelNm);
if (!testLabelNm) {
err.append("Tip and node labels must be a named vector, the names must match the node IDs. ");
err.append("Use tipLabels<- and/or nodeLabels<- to update them. ");
}
// check that tips have labels
Rcpp::CharacterVector tiplabel(ntipsFast);
std::copy (label.begin(), label.begin()+ntipsFast, tiplabel.begin());
bool emptyTipLabel = is_true(any(Rcpp::is_na(tiplabel)));
if ( emptyTipLabel ) {
err.append("All tips must have a label.");
}
// check edgeLabels
Rcpp::CharacterVector chrEdgeLblNm = edgeIdCpp(ed, "all");
bool testEdgeLblNm = isLabelName(edLabelNm, chrEdgeLblNm);
if (!testEdgeLblNm) {
err.append("Edge labels are not labelled correctly. Use the function edgeLabels<- to update them. ");
}
// make sure that tips and node labels are unique
if (hasDuplicatedLabelsCpp(label)) {
std::string labOpt = opts["allow.duplicated.labels"];
if (labOpt == "fail") {
err.append("Labels are not unique. ");
}
if (labOpt == "warn") {
wrn.append("Labels are not unique. ");
}
}
// check for polytomies
if (hasPolytomy(ances)) {
std::string msgPoly = "Tree includes polytomies. ";
std::string polyOpt = opts["poly"];
if (polyOpt == "fail") {
err.append(msgPoly);
}
if (polyOpt == "warn") {
wrn.append(msgPoly);
}
}
// check number of roots
if (nroots > 1) {
std::string msgRoot = "Tree has more than one root. ";
std::string rootOpt = opts["multiroot"];
if (rootOpt == "fail") {
err.append(msgRoot);
}
if (rootOpt == "warn") {
wrn.append(msgRoot);
}
}
// check for singletons
if (hasSingleton(ances)) {
std::string msgSing = "Tree contains singleton nodes. ";
std::string singOpt = opts["singleton"];
if (singOpt == "fail") {
err.append(msgSing);
}
if (singOpt == "warn") {
wrn.append(msgSing);
}
}
return Rcpp::List::create(err, wrn);
}