extern "C" SEXP GetNCL(SEXP params, SEXP paramsVecR) {
Rcpp::List list(params);
Rcpp::LogicalVector paramsVec(paramsVecR);
bool charall = paramsVec[0];
bool polyconvert = paramsVec[1];
bool levelsUnif = paramsVec[2];
bool returnTrees = paramsVec[3];
bool returnData = paramsVec[4];
int nCharToReturn = 0;
std::vector<std::string> dataTypes; //vector of datatypes for each character block
std::vector<int> nbCharacters; //number of characters for each character block
std::vector<std::string> dataChr; //characters
std::vector<std::string> charLabels; //labels for the characters
std::vector<std::string> stateLabels; //labels for the states
std::vector<int> nbStates; //number of states for each character (for Standard datatype)
std::vector<std::string> trees; //vector of Newick strings holding the names
std::vector<std::string> treeNames; //vector of tree names
std::vector<std::string> taxaNames; //vector of taxa names
std::vector<bool> test(3);
test[0] = charall;
test[1] = polyconvert;
test[2] = levelsUnif;
# if defined(FILENAME_AS_NEXUS)
std::string filename = "'" + list["fileName"] + "'";
# else
std::string filename = list["fileName"];
# endif
MultiFormatReader nexusReader(-1, NxsReader::WARNINGS_TO_STDERR);
/* make NCL less strict */
NxsTreesBlock * treesB = nexusReader.GetTreesBlockTemplate();
treesB->SetAllowImplicitNames(true);
nexusReader.cullIdenticalTaxaBlocks(true);
/* End of making NCL less strict */
nexusReader.ReadFilepath(const_cast < char* > (filename.c_str()), MultiFormatReader::NEXUS_FORMAT);
const unsigned nTaxaBlocks = nexusReader.GetNumTaxaBlocks();
for (unsigned t = 0; t < nTaxaBlocks; ++t) {
/* Get blocks */
const NxsTaxaBlock * taxaBlock = nexusReader.GetTaxaBlock(t);
const unsigned nTreesBlocks = nexusReader.GetNumTreesBlocks(taxaBlock);
const unsigned nCharBlocks = nexusReader.GetNumCharactersBlocks(taxaBlock);
int nTax = taxaBlock->GetNumTaxonLabels();
/* Get taxa names */
for (int j=0; j < nTax; ++j) {
taxaNames.push_back (taxaBlock->GetTaxonLabel(j));
}
/* Get trees */
if (returnTrees) {
if (nTreesBlocks == 0) {
continue;
}
for (unsigned i = 0; i < nTreesBlocks; ++i) {
NxsTreesBlock* treeBlock = nexusReader.GetTreesBlock(taxaBlock, i);
const unsigned nTrees = treeBlock->GetNumTrees();
if (nTrees > 0) {
for (unsigned k = 0; k < nTrees; k++) {
NxsString ts = treeBlock->GetTreeDescription(k);
NxsString trNm = treeBlock->GetTreeName(k);
treeNames.push_back(trNm);
trees.push_back (ts);
}
}
else {
continue;
}
}
}
/* Get data */
if (returnData) {
for (unsigned k = 0; k < nCharBlocks; ++k) {
NxsCharactersBlock * charBlock = nexusReader.GetCharactersBlock(taxaBlock, k);
if (nCharBlocks == 0) {
continue;
}
else {
NxsString dtType = charBlock->GetNameOfDatatype(charBlock->GetDataType());
dataTypes.push_back(dtType);
if (charall) {
nCharToReturn=charBlock->GetNCharTotal();
}
else {
nCharToReturn=charBlock->GetNumIncludedChars();
}
nbCharacters.push_back (nCharToReturn);
for (int eachChar=0; eachChar < nCharToReturn; ++eachChar) { //We only pass the non-eliminated chars
NxsString charLabel=charBlock->GetCharLabel(eachChar);
if (charLabel.length()>1) {
charLabels.push_back (charLabel);
}
else {
charLabels.push_back ("standard_char"); //FIXME: needs to fixed for sequence data
}
NxsString tmpCharString;
if (std::string("Continuous") == dtType) {
tmpCharString = contData(*charBlock, tmpCharString, eachChar, nTax);
nbStates.push_back (0);
}
else {
if (std::string("Standard") == dtType) {
tmpCharString = stdData(*charBlock, tmpCharString, eachChar, nTax,
polyconvert);
unsigned int nCharStates = charBlock->GetNumObsStates(eachChar, false);
nbStates.push_back (nCharStates);
for (unsigned int l=0; l < nCharStates; ++l) {
NxsString label = charBlock->GetStateLabel(eachChar, l);
stateLabels.push_back (label);
}
}
else {
if (std::string("DNA") == dtType) {
for (int taxon=0; taxon < nTax; ++taxon) {
for (int eachChar=0; eachChar < nCharToReturn; ++eachChar) {
unsigned int nCharStates = charBlock->GetNumStates(taxon, eachChar);
if (charBlock->IsGapState(taxon, eachChar)) {
tmpCharString += "-";
}
else {
if (charBlock->IsMissingState(taxon, eachChar)) {
tmpCharString += "?";
}
else {
if (nCharStates == 1) {
tmpCharString += charBlock->GetState(taxon, eachChar, 0);
}
else {
tmpCharString += "?"; //FIXME
}
}
}
}
}
}
else { // other type of data not yet supported
tmpCharString = "";
nbStates.push_back (0);
stateLabels.push_back (std::string(""));
}
}
}
std::string charString = "c(" + tmpCharString + ");";
dataChr.push_back (charString);
}
}
}
}
}
/* Prepare list to return */
Rcpp::List res = Rcpp::List::create(Rcpp::Named("taxaNames") = taxaNames,
Rcpp::Named("treeNames") = treeNames,
Rcpp::Named("trees") = trees,
Rcpp::Named("dataTypes") = dataTypes,
Rcpp::Named("nbCharacters") = nbCharacters,
Rcpp::Named("charLabels") = charLabels,
Rcpp::Named("nbStates") = nbStates,
Rcpp::Named("stateLabels") = stateLabels,
Rcpp::Named("dataChr") = dataChr,
Rcpp::Named("Test") = test);
return res;
}
void AminoacidData::CreateMatrixFromNCL(GarliReader &reader){
NxsCharactersBlock *charblock;
int num=0, numNuc = -1;
do{
charblock = reader.GetCharactersBlock(num);
if(charblock->GetDataType() == NxsCharactersBlock::protein){
if(numNuc < 0) numNuc = num;
else{
throw ErrorException("Multiple characters/data blocks containing protein data found in Nexus datafile!\n\tEither combine the blocks or comment one out.");
}
}
else outman.UserMessage("Ignoring non-protein characters block from Nexus datafile");
num++;
}while(num < reader.NumCharBlocks());
if(numNuc < 0) throw ErrorException("No characters/data blocks containing protein data found in Nexus datafile!");
charblock = reader.GetCharactersBlock(numNuc);
if(charblock->GetNumActiveChar() < charblock->GetNChar()){
outman.UserMessageNoCR("Excluded characters:\n\t");
for(int c=0;c<charblock->GetNCharTotal();c++)
if(charblock->IsExcluded(c)) outman.UserMessageNoCR("%d ", c+1);
outman.UserMessage("");
}
// vector<unsigned> reducedToOrigCharMap = charblock->GetOrigIndexVector();
NxsTaxaBlock *taxablock = reader.GetTaxaBlock();
int numOrigTaxa = charblock->GetNTax();
int numActiveTaxa = charblock->GetNumActiveTaxa();
int numOrigChar = charblock->GetNChar();
int numActiveChar = charblock->GetNumActiveChar();
//int num_chars = reducedToOrigCharMap.size();
//cout << num_chars << endl;
NewMatrix( numActiveTaxa, numActiveChar );
// read in the data, including taxon names
int i=0;
for( int origTaxIndex = 0; origTaxIndex < numOrigTaxa; origTaxIndex++ ) {
if(charblock->IsActiveTaxon(origTaxIndex)){
//internally, blanks in taxon names will be stored as underscores
NxsString tlabel = taxablock->GetTaxonLabel(origTaxIndex);
tlabel.BlanksToUnderscores();
SetTaxonLabel( i, tlabel.c_str());
int j = 0;
bool firstAmbig = true;
for( int origIndex = 0; origIndex < numOrigChar; origIndex++ ) {
if(charblock->IsActiveChar(origIndex)){
unsigned char datum = '\0';
if(charblock->IsGapState(origTaxIndex, origIndex) == true) datum = 20;
else if(charblock->IsMissingState(origTaxIndex, origIndex) == true) datum = 20;
else{
int nstates = charblock->GetNumStates(origTaxIndex, origIndex);
//assert(nstates == 1);
//need to deal with the possibility of multiple states represented in matrix
//just convert to full ambiguity
if(nstates == 1)
datum = CharToDatum(charblock->GetState(origTaxIndex, origIndex, 0));
else{
if(firstAmbig){
outman.UserMessageNoCR("Partially ambiguous characters of taxon %s converted to full ambiguity:\n\t", TaxonLabel(origTaxIndex));
firstAmbig = false;
}
outman.UserMessageNoCR("%d ", origIndex+1);
datum = CharToDatum('?');
}
}
SetMatrix( i, j++, datum );
}
}
if(firstAmbig == false) outman.UserMessage("");
i++;
}
}
}
void DataMatrix::CreateMatrixFromNCL(GarliReader &reader){
NxsCharactersBlock *charblock = reader.GetCharactersBlock();
// vector<unsigned> reducedToOrigCharMap = charblock->GetOrigIndexVector();
NxsTaxaBlock *taxablock = reader.GetTaxaBlock();
int numOrigTaxa = charblock->GetNTax();
int numActiveTaxa = charblock->GetNumActiveTaxa();
int numOrigChar = charblock->GetNChar();
int numActiveChar = charblock->GetNumActiveChar();
//int num_chars = reducedToOrigCharMap.size();
//cout << num_chars << endl;
NewMatrix( numActiveTaxa, numActiveChar );
// read in the data, including taxon names
int i=0;
if(modSpec.IsAminoAcid() == false){
for( int origTaxIndex = 0; origTaxIndex < numOrigTaxa; origTaxIndex++ ) {
if(charblock->IsActiveTaxon(origTaxIndex)){
SetTaxonLabel( i, taxablock->GetTaxonLabel(origTaxIndex).c_str());
int j = 0;
for( int origIndex = 0; origIndex < numOrigChar; origIndex++ ) {
if(charblock->IsActiveChar(origIndex)){
unsigned char datum = '\0';
if(charblock->IsGapState(origTaxIndex, origIndex) == true) datum = 15;
else if(charblock->IsMissingState(origTaxIndex, origIndex) == true) datum = 15;
else{
int nstates = charblock->GetNumStates(origTaxIndex, origIndex);
for(int s=0;s<nstates;s++){
datum += CharToBitwiseRepresentation(charblock->GetState(origTaxIndex, origIndex, s));
}
}
SetMatrix( i, j++, datum );
}
}
i++;
}
}
}
else{
for( int origTaxIndex = 0; origTaxIndex < numOrigTaxa; origTaxIndex++ ) {
if(charblock->IsActiveTaxon(origTaxIndex)){
SetTaxonLabel( i, taxablock->GetTaxonLabel(origTaxIndex).c_str());
int j = 0;
for( int origIndex = 0; origIndex < numOrigChar; origIndex++ ) {
if(charblock->IsActiveChar(origIndex)){
unsigned char datum = '\0';
if(charblock->IsGapState(origTaxIndex, origIndex) == true) datum = 20;
else if(charblock->IsMissingState(origTaxIndex, origIndex) == true) datum = 20;
else{
int nstates = charblock->GetNumStates(origTaxIndex, origIndex);
assert(nstates == 1);
datum = CharToAminoAcidNumber(charblock->GetState(origTaxIndex, origIndex, 0));
/* for(int s=0;s<nstates;s++){
datum += CharToBitwiseRepresentation(charblock->GetState(origTaxIndex, origIndex, s));
}
*/ }
SetMatrix( i, j++, datum );
}
}
i++;
}
}
}
}
void NucleotideData::CreateMatrixFromNCL(GarliReader &reader){
NxsCharactersBlock *charblock;
int num=0, numNuc = -1;
do{
charblock = reader.GetCharactersBlock(num);
if(charblock->GetDataType() == NxsCharactersBlock::nucleotide ||
charblock->GetDataType() == NxsCharactersBlock::dna ||
charblock->GetDataType() == NxsCharactersBlock::rna){
if(numNuc < 0) numNuc = num;
else{
throw ErrorException("Multiple characters/data blocks containing nucleotide data found in Nexus datafile!\n\tEither combine the blocks or comment one out.");
}
}
else outman.UserMessage("Ignoring non-nucleotide characters block from Nexus datafile");
num++;
}while(num < reader.NumCharBlocks());
if(numNuc < 0) throw ErrorException("No characters/data blocks containing nucleotide data found in Nexus datafile!");
charblock = reader.GetCharactersBlock(numNuc);
if(charblock->GetNumActiveChar() < charblock->GetNChar()){
outman.UserMessageNoCR("Excluded characters:\n\t");
for(int c=0;c<charblock->GetNCharTotal();c++)
if(charblock->IsExcluded(c)) outman.UserMessageNoCR("%d ", c+1);
outman.UserMessage("");
}
// vector<unsigned> reducedToOrigCharMap = charblock->GetOrigIndexVector();
NxsTaxaBlock *taxablock = reader.GetTaxaBlock();
int numOrigTaxa = charblock->GetNTax();
int numActiveTaxa = charblock->GetNumActiveTaxa();
int numOrigChar = charblock->GetNChar();
int numActiveChar = charblock->GetNumActiveChar();
//int num_chars = reducedToOrigCharMap.size();
//cout << num_chars << endl;
NewMatrix( numActiveTaxa, numActiveChar );
// read in the data, including taxon names
int i=0;
for( int origTaxIndex = 0; origTaxIndex < numOrigTaxa; origTaxIndex++ ) {
if(charblock->IsActiveTaxon(origTaxIndex)){
//internally, blanks in taxon names will be stored as underscores
NxsString tlabel = taxablock->GetTaxonLabel(origTaxIndex);
tlabel.BlanksToUnderscores();
SetTaxonLabel( i, tlabel.c_str());
int j = 0;
for( int origIndex = 0; origIndex < numOrigChar; origIndex++ ) {
if(charblock->IsActiveChar(origIndex)){
unsigned char datum = '\0';
if(charblock->IsGapState(origTaxIndex, origIndex) == true) datum = 15;
else if(charblock->IsMissingState(origTaxIndex, origIndex) == true) datum = 15;
else{
int nstates = charblock->GetNumStates(origTaxIndex, origIndex);
for(int s=0;s<nstates;s++){
datum += CharToBitwiseRepresentation(charblock->GetState(origTaxIndex, origIndex, s));
}
}
SetMatrix( i, j++, datum );
}
}
i++;
}
}
}
