void sequence::resize(size_type new_capacity)
// this function copies "this" into a temp object of the desired capacity,
// deletes stale data memory (via the overloaded assignment operator),
// and returns the re-sized object reference to the caller
{
if( new_capacity < used ) // ensure new_capacity is big
new_capacity = used; // enough for current contents
if( new_capacity <= 0 ) // ensure pre-condition is met
new_capacity = 1; // s.t. new_capacity > 0
// create temp object w/new_capacity to hold "this" contents
sequence tempSeq(new_capacity);
tempSeq.used = used;
tempSeq.current_index = current_index;
// copy data from "this" into tempSeq
size_type i;
for( i = 0 ; i < used; i++ )
{
tempSeq.data[i] = data[i];
}
// send "this" and tempSeq objects to the operator=() function to
// * copy the tempSeq obj back into "this"
// * repoint the "this" data pointer to the new memory store
// * delete stale memory
*this = tempSeq;
}
void Database::LoadObservationSequences(const char* gestureName, const Quantizer& quantizer, vector<TimeSlot>& seqVec)
{
seqVec.reserve(seqVecSpace);
//SQL: "SELECT oLength o FROM ObservationSequenceTable WHERE gestureName==?1 AND quantizerName==?2;";
rc = sqlite3_bind_text(loadObservationSequencesStmt, 1, gestureName, -1, SQLITE_TRANSIENT);
if(rc!=SQLITE_OK) std::cout << "Error when binding gestureName in LoadObservationSequences()" << std::endl;
rc = sqlite3_bind_text(loadObservationSequencesStmt, 2, quantizer.name.c_str(), quantizer.name.size(), SQLITE_TRANSIENT);
if(rc!=SQLITE_OK) std::cout << "Error when binding quantizerName in LoadObservationSequences()" << std::endl;
TimeSlot tempSeq(gestureName,quantizer);
size_t oLength;
const size_t* oPtr = 0;
//evaluation
rc = sqlite3_step(loadObservationSequencesStmt);
while(rc==SQLITE_ROW){
oLength= sqlite3_column_int(loadObservationSequencesStmt, 0);
oPtr = (const size_t*)sqlite3_column_blob(loadObservationSequencesStmt, 1);
seqVec.push_back(tempSeq);
seqVec.back().o.assign(oPtr,oPtr+oLength);
rc = sqlite3_step(loadObservationSequencesStmt);
}
if(rc!=SQLITE_DONE) std::cout << "Error when evaluation in LoadObservationSequences()" << std::endl;
//reset
sqlite3_reset(loadObservationSequencesStmt);
sqlite3_clear_bindings(loadObservationSequencesStmt);
}
SequenceParser::SequenceParser(string groupFile, string fastaFile, string nameFile) {
try {
m = MothurOut::getInstance();
int error;
//read group file
groupMap = new GroupMap(groupFile);
error = groupMap->readMap();
if (error == 1) { m->control_pressed = true; }
//initialize maps
vector<string> namesOfGroups = groupMap->getNamesOfGroups();
for (int i = 0; i < namesOfGroups.size(); i++) {
vector<Sequence> temp;
map<string, string> tempMap;
seqs[namesOfGroups[i]] = temp;
nameMapPerGroup[namesOfGroups[i]] = tempMap;
}
//read fasta file making sure each sequence is in the group file
ifstream in;
m->openInputFile(fastaFile, in);
map<string, string> seqName; //stores name -> sequence string so we can make new "unique" sequences when we parse the name file
while (!in.eof()) {
if (m->control_pressed) { break; }
Sequence seq(in); m->gobble(in);
if (seq.getName() != "") {
string group = groupMap->getGroup(seq.getName());
if (group == "not found") { error = 1; m->mothurOut("[ERROR]: " + seq.getName() + " is in your fasta file and not in your groupfile, please correct."); m->mothurOutEndLine(); }
else {
seqs[group].push_back(seq);
seqName[seq.getName()] = seq.getAligned();
}
}
}
in.close();
if (error == 1) { m->control_pressed = true; }
//read name file
ifstream inName;
m->openInputFile(nameFile, inName);
string first, second;
int countName = 0;
set<string> thisnames1;
while(!inName.eof()) {
if (m->control_pressed) { break; }
inName >> first; m->gobble(inName);
inName >> second; m->gobble(inName);
vector<string> names;
m->splitAtChar(second, names, ',');
//get aligned string for these seqs from the fasta file
string alignedString = "";
map<string, string>::iterator itAligned = seqName.find(names[0]);
if (itAligned == seqName.end()) {
error = 1; m->mothurOut("[ERROR]: " + names[0] + " is in your name file and not in your fasta file, please correct."); m->mothurOutEndLine();
}else {
alignedString = itAligned->second;
}
//separate by group - parse one line in name file
map<string, string> splitMap; //group -> name1,name2,...
map<string, string>::iterator it;
for (int i = 0; i < names.size(); i++) {
string group = groupMap->getGroup(names[i]);
if (group == "not found") { error = 1; m->mothurOut("[ERROR]: " + names[i] + " is in your name file and not in your groupfile, please correct."); m->mothurOutEndLine(); }
else {
it = splitMap.find(group);
if (it != splitMap.end()) { //adding seqs to this group
(it->second) += "," + names[i];
thisnames1.insert(names[i]);
countName++;
}else { //first sighting of this group
splitMap[group] = names[i];
countName++;
thisnames1.insert(names[i]);
//is this seq in the fasta file?
if (i != 0) { //if not then we need to add a duplicate sequence to the seqs for this group so the new "fasta" and "name" files will match
Sequence tempSeq(names[i], alignedString); //get the first guys sequence string since he's in the fasta file.
seqs[group].push_back(tempSeq);
}
}
}
allSeqsMap[names[i]] = names[0];
}
//fill nameMapPerGroup - holds all lines in namefile separated by group
for (it = splitMap.begin(); it != splitMap.end(); it++) {
//grab first name
string firstName = "";
for(int i = 0; i < (it->second).length(); i++) {
if (((it->second)[i]) != ',') {
firstName += ((it->second)[i]);
}else { break; }
}
//group1 -> seq1 -> seq1,seq2,seq3
nameMapPerGroup[it->first][firstName] = it->second;
}
}
inName.close();
if (error == 1) { m->control_pressed = true; }
if (countName != (groupMap->getNumSeqs())) {
vector<string> groupseqsnames = groupMap->getNamesSeqs();
for (int i = 0; i < groupseqsnames.size(); i++) {
set<string>::iterator itnamesfile = thisnames1.find(groupseqsnames[i]);
if (itnamesfile == thisnames1.end()){
cout << "missing name " + groupseqsnames[i] << '\t' << allSeqsMap[groupseqsnames[i]] << endl;
}
}
m->mothurOutEndLine();
m->mothurOut("[ERROR]: Your name file contains " + toString(countName) + " valid sequences, and your groupfile contains " + toString(groupMap->getNumSeqs()) + ", please correct.");
m->mothurOutEndLine();
m->control_pressed = true;
}
}
catch(exception& e) {
m->errorOut(e, "SequenceParser", "SequenceParser");
exit(1);
}
}
