void Tree::reorderSeqs (vguard<FastSeq>& seq) const {
Assert (seq.size() == nodes(), "Number of sequences doesn't match number of nodes in tree");
map<string,size_t> name2seq;
for (size_t n = 0; n < seq.size(); ++n) {
Assert (name2seq.find (seq[n].name) == name2seq.end(), "Duplicate sequence name: %s", seq[n].name.c_str());
name2seq[seq[n].name] = n;
}
vguard<size_t> new2old (seq.size()), old2new (seq.size());
for (TreeNodeIndex n = 0; n < nodes(); ++n) {
Assert (name2seq.find (seqName(n)) != name2seq.end(), "Tree node %s is absent from sequence dataset", seqName(n).c_str());
const size_t old_n = name2seq[seqName(n)];
new2old[n] = old_n;
old2new[old_n] = n;
}
for (size_t n = 0; n < new2old.size(); ++n) {
const size_t o = new2old[n], m = old2new[n];
swap (seq[n], seq[o]);
swap (old2new[n], old2new[o]);
swap (new2old[n], new2old[m]);
}
}
string Tree::seqName (TreeNodeIndex n) const {
string s = nodeName(n);
if (s.size() == 0) {
vguard<string> cs;
for (auto c : node[n].child) {
ostringstream o;
o << seqName(c) << ':' << defaultfloat << branchLength(c);
cs.push_back (o.str());
}
s = "(" + join(cs,",") + ")";
}
return s;
}
bool Tree::seqNamesBijective (const vguard<FastSeq>& seq) const {
if (!allNodesNamed())
return false;
if (seq.size() != nodes())
return false;
map<string,size_t> name2seq;
for (size_t n = 0; n < seq.size(); ++n) {
if (name2seq.find (seq[n].name) != name2seq.end())
return false;
name2seq[seq[n].name] = n;
}
for (TreeNodeIndex n = 0; n < nodes(); ++n)
if (name2seq.find (seqName(n)) == name2seq.end())
return false;
return true;
}
void Tree::assignInternalNodeNames (vguard<FastSeq>& seq, const char* prefix) {
reorderSeqs (seq); // make sure that nodes match rows
assignInternalNodeNames (prefix);
for (size_t n = 0; n < nodes(); ++n)
seq[n].name = seqName(n);
}
void Tree::validateBranchLengths() const {
for (size_t n = 0; n + 1 < node.size(); ++n) {
Require (branchLength(n) >= 0, "Node in tree is missing branch length: %s", seqName(n).c_str());
Require (branchLength(n) >= minBranchLength, "Node in tree has a lower-than-minimal branch length: %s", seqName(n).c_str());
}
}
static void load(IOAdapter* io, const U2DbiRef& dbiRef, QList<GObject*>& objects, const QVariantMap& fs, U2OpStatus& os) {
DbiOperationsBlock opBlock(dbiRef, os);
CHECK_OP(os, );
Q_UNUSED(opBlock);
U2SequenceImporter seqImporter(fs, true);
const QString folder = fs.value(DocumentFormat::DBI_FOLDER_HINT, U2ObjectDbi::ROOT_FOLDER).toString();
QByteArray readBuffer(DocumentFormat::READ_BUFF_SIZE, '\0');
char* buff = readBuffer.data();
QBitArray ALPHAS = TextUtils::ALPHA_NUMS;
ALPHAS['-'] = true;
QByteArray seq;
QString seqName(io->getURL().baseFileName());
//reading sequence
QBuffer writer(&seq);
writer.open(QIODevice::WriteOnly);
TmpDbiObjects dbiObjects(dbiRef, os);
bool ok = true;
int len = 0;
bool isStarted = false;
int sequenceCounter = 0;
bool terminatorFound = false;
bool isSplit = fs.value((DocumentReadingMode_SequenceAsSeparateHint), false).toBool();
while (ok && !io->isEof()) {
len = io->readLine(buff, DocumentFormat::READ_BUFF_SIZE, &terminatorFound);
if (len <= 0){
continue;
}
seq.clear();
bool isSeek = writer.seek(0);
assert(isSeek); Q_UNUSED(isSeek);
if (os.isCoR()) {
break;
}
for (int i=0; i<len && ok; i++) {
char c = buff[i];
if (ALPHAS[(uchar)c]) {
ok = writer.putChar(c);
}
}
if(seq.size()>0 && isStarted == false ){
QString name = sequenceCounter == 0 ? seqName : seqName + QString("_%1").arg(sequenceCounter);
isStarted = true;
seqImporter.startSequence(dbiRef, folder, name, false, os);
}
if(isStarted){
seqImporter.addBlock(seq.data(),seq.size(),os);
}
if (seq.size()>0 && isStarted && terminatorFound && isSplit){
finishSequence(objects, io, os, dbiRef, fs, dbiObjects, seqImporter);
sequenceCounter++;
isStarted = false;
}
if (os.isCoR()) {
break;
}
os.setProgress(io->getProgress());
}
writer.close();
CHECK_OP(os, );
if (sequenceCounter == 0){
CHECK_EXT(isStarted == true, os.setError(RawDNASequenceFormat::tr("Sequence is empty")), );
}
