void Mase::writeSequences(ostream& output, const SequenceContainer& sc) const throw (Exception)
{
// Checking the existence of specified file, and possibility to open it in write mode
if (!output) { throw IOException ("Mase::write : failed to open file"); }
Comments comments = sc.getGeneralComments();
// Writing all general comments in file
if (comments.size() == 0) {
output << ";;" << endl;
}
for (size_t i = 0 ; i < comments.size() ; i++) {
output << ";;" << comments[i] << endl;
}
string seq, temp = ""; // Initialization
// Main loop : for all sequences
vector<string> names = sc.getSequencesNames();
for (size_t i = 0 ; i < names.size() ; i ++)
{
comments = sc.getComments(names[i]);
// Writing all sequence comments in file
// If no comments are associated with current sequence, an empy commentary line will be writed
if (comments.size() == 0)
{
output << ";" << endl;
}
else
{
for (size_t j = 0 ; j < comments.size() ; j++)
{
output << ";" << comments[j] << endl;
}
}
// Sequence name writing
output << names[i] << endl;
// Sequence cutting to specified characters number per line
seq = sc.toString(names[i]);
while (seq != "")
{
if (seq.size() > charsByLine_)
{
temp = seq;
temp.erase(temp.begin() + static_cast<ptrdiff_t>(charsByLine_), temp.end());
output << temp << endl;
seq.erase(seq.begin(), seq.begin() + static_cast<ptrdiff_t>(charsByLine_));
}
else
{
output << seq << endl;
seq = "";
}
}
}
}
unsigned int MaseTools::getPhase(const Comments & maseFileHeader, const string &setName) throw (Exception)
{
unsigned int phase = 0;
string::size_type index = 0;
for(unsigned int i = 0; i < maseFileHeader.size(); i++) {
string current = maseFileHeader[i];
index = current.find("# of");
if(index < current.npos) {
StringTokenizer st(string(current.begin() + index + 12 , current.end()), " \t\n\f\r=;");
//unsigned int numberOfSegments = TextTools::toInt(st.nextToken());
//cout << "Number of regions: " << st.nextToken() << endl;
string name;
while(st.hasMoreToken()) {
name = st.nextToken();
//cout << "Name of regions: " << name << endl;
}
if(name == setName) {
return phase;
}
}
index = current.find("/codon_start");
if(index < current.npos) {
StringTokenizer st(string(current.begin() + index + 12, current.end()), " \t\n\f\r=;");
phase = TextTools::toInt(st.nextToken());
}
}
throw Exception("PolymorphismSequenceContainer::getPhase: no /codon_start found, or site selection missing.");
}
string PolymorphismSequenceContainerTools::getIngroupSpeciesName(const PolymorphismSequenceContainer & psc)
{
string key;
unsigned int n;
string speciesName;
Comments maseFileHeader = psc.getGeneralComments();
if(!maseFileHeader.size()) return speciesName;
map<string, unsigned int> groupMap = MaseTools::getAvailableSequenceSelections(maseFileHeader);
for(map<string, unsigned int>::iterator mi = groupMap.begin() ; mi != groupMap.end() ; mi++) {
key = mi->first;
n = mi->second;
if (key.compare(0, 7, "INGROUP") == 0 ) {
StringTokenizer * sptk = new StringTokenizer(key, "_");
speciesName = sptk -> getToken(1) + " " + sptk -> getToken(2);
}
}
return speciesName;
}
SiteSelection MaseTools::getSiteSet(const Comments & maseFileHeader, const string & setName) throw (IOException)
{
SiteSelection selection;
for(unsigned int i = 0; i < maseFileHeader.size(); i++) {
string current = maseFileHeader[i];
string::size_type index = current.find("# of");
if(index < current.npos) {
StringTokenizer st(string(current.begin() + index + 4, current.end()), " \t\n\f\r=;");
st.nextToken(); //skip next word: may be 'regions' or 'segments' or else ;-)
unsigned int numberOfSegments = TextTools::toInt(st.nextToken());
string name;
while(st.hasMoreToken()) {
name += st.nextToken();
}
if(name == setName) {
//cout << numberOfSegments << " segments found." << endl;
//Then look for the set definition:
i++;//next line.
unsigned int counter = 0;
while(i < maseFileHeader.size()) {
current = maseFileHeader[i++];
StringTokenizer st2(current);
//st.nextToken(); //Skip ';;'
while(st2.hasMoreToken()) {
StringTokenizer st3(st2.nextToken(), ",");
unsigned int begin = TextTools::toInt(st3.nextToken());
unsigned int end = TextTools::toInt(st3.nextToken());
//WARNING!!! In the mase+ format, sites are numbered from 1 to nbSites,
//Whereas in SiteContainer the index begins at 0.
for(unsigned int j = begin; j <= end; j++) selection.push_back(j - 1);//bounds included.
counter++;
if(counter == numberOfSegments) return selection;
}
}
}
}
}
if(selection.size() == 0) {
throw IOException("Site set " + setName + " has not been found in the sequence file.");
}
return selection;
}
map<string, unsigned int> MaseTools::getAvailableSiteSelections(const Comments & maseHeader)
{
map<string, unsigned int> selections;
for(unsigned int i = 0; i < maseHeader.size(); i++) {
string current = maseHeader[i];
string::size_type index = current.find("# of");
if(index < current.npos) {
StringTokenizer st(string(current.begin() + index + 4, current.end()), " \t\n\f\r=;");
st.nextToken(); //skip next word: may be 'sequences' or else ;-)
unsigned int numberOfSegments = TextTools::toInt(st.nextToken());
string name = st.nextToken();
while(st.hasMoreToken()) {
name += " " +st.nextToken();
}
unsigned int counter = 0;
unsigned nbSites = 0;
while(i < maseHeader.size()) {
i++;
current = maseHeader[i];
StringTokenizer st2(current);
//st.nextToken(); //Skip ';;'
while(st2.hasMoreToken()) {
StringTokenizer st3(st2.nextToken(), ",");
unsigned int begin = TextTools::toInt(st3.nextToken());
unsigned int end = TextTools::toInt(st3.nextToken());
counter++;
nbSites += end - begin + 1;
}
if(counter == numberOfSegments) {
selections[name] = nbSites;
break;
}
}
}
}
return selections;
}
map<string, unsigned int> MaseTools::getAvailableSequenceSelections(const Comments & maseHeader)
{
map<string, unsigned int> selections;
for(unsigned int i = 0; i < maseHeader.size(); i++) {
string current = maseHeader[i];
string::size_type index = current.find("@ of");
if(index < current.npos) {
StringTokenizer st(string(current.begin() + index + 4, current.end()), " \t\n\f\r=;");
st.nextToken(); //skip next word: may be 'sequences' or else ;-)
unsigned int numberOfSequences = TextTools::fromString<unsigned int>(st.nextToken());
string name = st.nextToken();
while(st.hasMoreToken()) {
name += st.nextToken();
}
selections[name] = numberOfSequences;
}
}
return selections;
}
int main()
{
list<string> commands;
list<string> files;
string tempString;
string input;
//Algorithm to split the input and put files in list "files" and comamnds in list "commands"
getline(cin, input);
for (int i = 0; i < input.length(); i++)
{
do
{
tempString.append(1, input[i]);
i++;
} while (input[i] != ' ' && i < input.length());
if (tempString[0] == '-')
{
commands.push_back(tempString);
}
else
{
files.push_back(tempString);
}
tempString.clear();
}
//Foreaching each file with each command all in the requested sequence. Creating object of each class for the needed
// operation and using their properties and methods. On each itteration the input file is copied to the OLD file
// and the input file is being modified depending on the requested command.
for each (string file in files)
{
int counter = 0;
for each (string command in commands)
{
counter++;
string newLine;
ifstream inputFileRead(file);
string newFile;
newFile.append(file);
newFile.append(".old");
ofstream outputFileRead(newFile);
while (getline(inputFileRead, newLine))
{
outputFileRead << newLine << endl;
}
inputFileRead.close();
outputFileRead.close();
newLine.clear();
if (command == "--comments")
{
Comments comment;
ifstream outputFileWrite(newFile);
ofstream inputFileWrite(file);
while (getline(outputFileWrite, newLine))
{
comment.removeComments(newLine);
if (comment.newLineFixed.empty())
{
continue;
}
else
inputFileWrite << comment.newLineFixed << endl;
}
outputFileWrite.close();
inputFileWrite.close();
}
else if (command == "--newlines=CRLF")
{
ifstream outputFileWrite(newFile);
ofstream inputFileWrite(file);
NewLines newlines;
while (getline(outputFileWrite, newLine))
{
newlines.newLinesCRFL(newLine);
inputFileWrite << newlines.newLineFix << endl;
}
}
else if (command == "--newlines=LF")
{
ifstream outputFileWrite(newFile);
ofstream inputFileWrite(file);
NewLines newlines;
while (getline(outputFileWrite, newLine))
{
newlines.newLinesCRFL(newLine);
inputFileWrite << newlines.newLineFix << endl;
}
}
else if (command == "--identation=spaces")
{
ifstream outputFileWrite(newFile);
ofstream inputFileWrite(file);
Identation identation;
while (getline(outputFileWrite, newLine))
{
identation.identationSpaces(newLine);
inputFileWrite << identation.newLineFix << endl;
}
}
else if (command == "--identation=tabs")
{
ifstream outputFileWrite(newFile);
ofstream inputFileWrite(file);
Identation identation;
while (getline(outputFileWrite, newLine))
{
identation.identationTabs(newLine);
inputFileWrite << identation.newLineFix << endl;
}
}
}
