#include "onegapignore.h"
//**********************************************************************************************************************
-
+vector<string> DistanceCommand::setParameters(){
+ try {
+ CommandParameter pcolumn("column", "InputTypes", "", "", "none", "none", "OldFastaColumn",false,false); parameters.push_back(pcolumn);
+ CommandParameter poldfasta("oldfasta", "InputTypes", "", "", "none", "none", "OldFastaColumn",false,false); parameters.push_back(poldfasta);
+ CommandParameter pfasta("fasta", "InputTypes", "", "", "none", "none", "none",false,true); parameters.push_back(pfasta);
+ CommandParameter poutput("output", "Multiple", "column-lt-square", "column", "", "", "",false,false); parameters.push_back(poutput);
+ CommandParameter pcalc("calc", "Multiple", "nogaps-eachgap-onegap", "onegap", "", "", "",false,false); parameters.push_back(pcalc);
+ CommandParameter pcountends("countends", "Boolean", "", "T", "", "", "",false,false); parameters.push_back(pcountends);
+ CommandParameter pcompress("compress", "Boolean", "", "F", "", "", "",false,false); parameters.push_back(pcompress);
+ CommandParameter pprocessors("processors", "Number", "", "1", "", "", "",false,false); parameters.push_back(pprocessors);
+ CommandParameter pcutoff("cutoff", "Number", "", "1.0", "", "", "",false,false); parameters.push_back(pcutoff);
+ CommandParameter pinputdir("inputdir", "String", "", "", "", "", "",false,false); parameters.push_back(pinputdir);
+ CommandParameter poutputdir("outputdir", "String", "", "", "", "", "",false,false); parameters.push_back(poutputdir);
+
+ vector<string> myArray;
+ for (int i = 0; i < parameters.size(); i++) { myArray.push_back(parameters[i].name); }
+ return myArray;
+ }
+ catch(exception& e) {
+ m->errorOut(e, "DistanceCommand", "setParameters");
+ exit(1);
+ }
+}
+//**********************************************************************************************************************
+string DistanceCommand::getHelpString(){
+ try {
+ string helpString = "";
+ helpString += "The dist.seqs command reads a file containing sequences and creates a distance file.\n";
+ helpString += "The dist.seqs command parameters are fasta, oldfasta, column, calc, countends, output, compress, cutoff and processors. \n";
+ helpString += "The fasta parameter is required, unless you have a valid current fasta file.\n";
+ helpString += "The oldfasta and column parameters allow you to append the distances calculated to the column file.\n";
+ helpString += "The calc parameter allows you to specify the method of calculating the distances. Your options are: nogaps, onegap or eachgap. The default is onegap.\n";
+ helpString += "The countends parameter allows you to specify whether to include terminal gaps in distance. Your options are: T or F. The default is T.\n";
+ helpString += "The cutoff parameter allows you to specify maximum distance to keep. The default is 1.0.\n";
+ helpString += "The output parameter allows you to specify format of your distance matrix. Options are column, lt, and square. The default is column.\n";
+ helpString += "The processors parameter allows you to specify number of processors to use. The default is 1.\n";
+ helpString += "The compress parameter allows you to indicate that you want the resulting distance file compressed. The default is false.\n";
+ helpString += "The dist.seqs command should be in the following format: \n";
+ helpString += "dist.seqs(fasta=yourFastaFile, calc=yourCalc, countends=yourEnds, cutoff= yourCutOff, processors=yourProcessors) \n";
+ helpString += "Example dist.seqs(fasta=amazon.fasta, calc=eachgap, countends=F, cutoff= 2.0, processors=3).\n";
+ helpString += "Note: No spaces between parameter labels (i.e. calc), '=' and parameters (i.e.yourCalc).\n";
+ return helpString;
+ }
+ catch(exception& e) {
+ m->errorOut(e, "DistanceCommand", "getHelpString");
+ exit(1);
+ }
+}
+//**********************************************************************************************************************
+DistanceCommand::DistanceCommand(){
+ try {
+ abort = true; calledHelp = true;
+ setParameters();
+ vector<string> tempOutNames;
+ outputTypes["phylip"] = tempOutNames;
+ outputTypes["column"] = tempOutNames;
+ }
+ catch(exception& e) {
+ m->errorOut(e, "DistanceCommand", "DistanceCommand");
+ exit(1);
+ }
+}
+//**********************************************************************************************************************
DistanceCommand::DistanceCommand(string option) {
try {
- abort = false;
+ abort = false; calledHelp = false;
Estimators.clear();
//allow user to run help
- if(option == "help") { help(); abort = true; }
+ if(option == "help") { help(); abort = true; calledHelp = true; }
+ else if(option == "citation") { citation(); abort = true; calledHelp = true;}
else {
- //valid paramters for this command
- string Array[] = {"fasta","oldfasta","column", "output", "calc", "countends", "cutoff", "processors", "outputdir","inputdir"};
- vector<string> myArray (Array, Array+(sizeof(Array)/sizeof(string)));
+ vector<string> myArray = setParameters();
OptionParser parser(option);
map<string, string> parameters = parser.getParameters();
if (validParameter.isValidParameter(it2->first, myArray, it2->second) != true) { abort = true; }
}
+ //initialize outputTypes
+ vector<string> tempOutNames;
+ outputTypes["phylip"] = tempOutNames;
+ outputTypes["column"] = tempOutNames;
+
//if the user changes the input directory command factory will send this info to us in the output parameter
string inputDir = validParameter.validFile(parameters, "inputdir", false);
if (inputDir == "not found"){ inputDir = ""; }
//check for required parameters
fastafile = validParameter.validFile(parameters, "fasta", true);
- if (fastafile == "not found") { m->mothurOut("fasta is a required parameter for the dist.seqs command."); m->mothurOutEndLine(); abort = true; }
- else if (fastafile == "not open") { abort = true; }
+ if (fastafile == "not found") {
+ fastafile = m->getFastaFile();
+ if (fastafile != "") { m->mothurOut("Using " + fastafile + " as input file for the fasta parameter."); m->mothurOutEndLine();
+ ifstream inFASTA;
+ m->openInputFile(fastafile, inFASTA);
+ alignDB = SequenceDB(inFASTA);
+ inFASTA.close();
+ }else { m->mothurOut("You have no current fastafile and the fasta parameter is required."); m->mothurOutEndLine(); abort = true; }
+ }else if (fastafile == "not open") { abort = true; }
else{
ifstream inFASTA;
m->openInputFile(fastafile, inFASTA);
alignDB = SequenceDB(inFASTA);
inFASTA.close();
+ m->setFastaFile(fastafile);
}
oldfastafile = validParameter.validFile(parameters, "oldfasta", true);
column = validParameter.validFile(parameters, "column", true);
if (column == "not found") { column = ""; }
else if (column == "not open") { abort = true; }
+ else { m->setColumnFile(column); }
//if the user changes the output directory command factory will send this info to us in the output parameter
outputDir = validParameter.validFile(parameters, "outputdir", false); if (outputDir == "not found"){
temp = validParameter.validFile(parameters, "cutoff", false); if(temp == "not found"){ temp = "1.0"; }
convert(temp, cutoff);
- temp = validParameter.validFile(parameters, "processors", false); if(temp == "not found"){ temp = "1"; }
- convert(temp, processors);
+ temp = validParameter.validFile(parameters, "processors", false); if (temp == "not found"){ temp = m->getProcessors(); }
+ m->setProcessors(temp);
+ convert(temp, processors);
+ temp = validParameter.validFile(parameters, "compress", false); if(temp == "not found"){ temp = "F"; }
+ convert(temp, compress);
+
output = validParameter.validFile(parameters, "output", false); if(output == "not found"){ output = "column"; }
if (((column != "") && (oldfastafile == "")) || ((column == "") && (oldfastafile != ""))) { m->mothurOut("If you provide column or oldfasta, you must provide both."); m->mothurOutEndLine(); abort=true; }
exit(1);
}
}
-
-//**********************************************************************************************************************
-
-DistanceCommand::~DistanceCommand(){
-
- for(int i=0;i<lines.size();i++){
- delete lines[i];
- }
-
-}
-
-//**********************************************************************************************************************
-
-void DistanceCommand::help(){
- try {
- m->mothurOut("The dist.seqs command reads a file containing sequences and creates a distance file.\n");
- m->mothurOut("The dist.seqs command parameters are fasta, oldfasta, column, calc, countends, output, cutoff and processors. \n");
- m->mothurOut("The fasta parameter is required.\n");
- m->mothurOut("The oldfasta and column parameters allow you to append the distances calculated to the column file.\n");
- m->mothurOut("The calc parameter allows you to specify the method of calculating the distances. Your options are: nogaps, onegap or eachgap. The default is onegap.\n");
- m->mothurOut("The countends parameter allows you to specify whether to include terminal gaps in distance. Your options are: T or F. The default is T.\n");
- m->mothurOut("The cutoff parameter allows you to specify maximum distance to keep. The default is 1.0.\n");
- m->mothurOut("The output parameter allows you to specify format of your distance matrix. Options are column, lt, and square. The default is column.\n");
- m->mothurOut("The processors parameter allows you to specify number of processors to use. The default is 1.\n");
- m->mothurOut("The dist.seqs command should be in the following format: \n");
- m->mothurOut("dist.seqs(fasta=yourFastaFile, calc=yourCalc, countends=yourEnds, cutoff= yourCutOff, processors=yourProcessors) \n");
- m->mothurOut("Example dist.seqs(fasta=amazon.fasta, calc=eachgap, countends=F, cutoff= 2.0, processors=3).\n");
- m->mothurOut("Note: No spaces between parameter labels (i.e. calc), '=' and parameters (i.e.yourCalc).\n\n");
- }
- catch(exception& e) {
- m->errorOut(e, "DistanceCommand", "help");
- exit(1);
- }
-}
//**********************************************************************************************************************
int DistanceCommand::execute(){
try {
- if (abort == true) { return 0; }
+ if (abort == true) { if (calledHelp) { return 0; } return 2; }
int startTime = time(NULL);
int numSeqs = alignDB.getNumSeqs();
cutoff += 0.005;
+ if (!alignDB.sameLength()) { m->mothurOut("[ERROR]: your sequences are not the same length, aborting."); m->mothurOutEndLine(); return 0; }
+
string outputFile;
if (output == "lt") { //does the user want lower triangle phylip formatted file
outputFile = outputDir + m->getRootName(m->getSimpleName(fastafile)) + "phylip.dist";
- remove(outputFile.c_str());
+ m->mothurRemove(outputFile); outputTypes["phylip"].push_back(outputFile);
//output numSeqs to phylip formatted dist file
}else if (output == "column") { //user wants column format
outputFile = outputDir + m->getRootName(m->getSimpleName(fastafile)) + "dist";
+ outputTypes["column"].push_back(outputFile);
//so we don't accidentally overwrite
if (outputFile == column) {
rename(column.c_str(), tempcolumn.c_str());
}
- remove(outputFile.c_str());
+ m->mothurRemove(outputFile);
}else { //assume square
outputFile = outputDir + m->getRootName(m->getSimpleName(fastafile)) + "square.dist";
- remove(outputFile.c_str());
+ m->mothurRemove(outputFile);
+ outputTypes["phylip"].push_back(outputFile);
}
//delete filename;
if (pid == 0) { //you are the root process
-
+
//do your part
string outputMyPart;
driverMPI(start, end, outMPI, cutoff);
- if (m->control_pressed) { MPI_File_close(&outMPI); delete distCalculator; return 0; }
+ if (m->control_pressed) { outputTypes.clear(); MPI_File_close(&outMPI); delete distCalculator; return 0; }
//wait on chidren
for(int i = 1; i < processors; i++) {
- if (m->control_pressed) { MPI_File_close(&outMPI); delete distCalculator; return 0; }
+ if (m->control_pressed) { outputTypes.clear(); MPI_File_close(&outMPI); delete distCalculator; return 0; }
- char buf[4];
- MPI_Recv(buf, 4, MPI_CHAR, i, tag, MPI_COMM_WORLD, &status);
+ char buf[5];
+ MPI_Recv(buf, 5, MPI_CHAR, i, tag, MPI_COMM_WORLD, &status);
}
}else { //you are a child process
//do your part
driverMPI(start, end, outMPI, cutoff);
- if (m->control_pressed) { MPI_File_close(&outMPI); delete distCalculator; return 0; }
+ if (m->control_pressed) { outputTypes.clear(); MPI_File_close(&outMPI); delete distCalculator; return 0; }
- char buf[4];
+ char buf[5];
strcpy(buf, "done");
//tell parent you are done.
- MPI_Send(buf, 4, MPI_CHAR, 0, tag, MPI_COMM_WORLD);
+ MPI_Send(buf, 5, MPI_CHAR, 0, tag, MPI_COMM_WORLD);
}
MPI_File_close(&outMPI);
if (output != "square"){ driverMPI(start, end, outputFile, mySize); }
else { driverMPI(start, end, outputFile, mySize, output); }
- if (m->control_pressed) { delete distCalculator; return 0; }
+ if (m->control_pressed) { outputTypes.clear(); delete distCalculator; return 0; }
int amode=MPI_MODE_APPEND|MPI_MODE_WRONLY|MPI_MODE_CREATE; //
MPI_File outMPI;
for(int b = 1; b < processors; b++) {
unsigned long int fileSize;
- if (m->control_pressed) { MPI_File_close(&outMPI); delete distCalculator; return 0; }
+ if (m->control_pressed) { outputTypes.clear(); MPI_File_close(&outMPI); delete distCalculator; return 0; }
MPI_Recv(&fileSize, 1, MPI_LONG, b, tag, MPI_COMM_WORLD, &status);
else { driver(0, numSeqs, outputFile, "square"); }
}else{ //you have multiple processors
+ unsigned long int numDists = 0;
+
+ if (output == "square") {
+ numDists = numSeqs * numSeqs;
+ }else {
+ for(int i=0;i<numSeqs;i++){
+ for(int j=0;j<i;j++){
+ numDists++;
+ if (numDists > processors) { break; }
+ }
+ }
+ }
+
+ if (numDists < processors) { processors = numDists; }
+
for (int i = 0; i < processors; i++) {
- lines.push_back(new linePair());
+ distlinePair tempLine;
+ lines.push_back(tempLine);
if (output != "square") {
- lines[i]->start = int (sqrt(float(i)/float(processors)) * numSeqs);
- lines[i]->end = int (sqrt(float(i+1)/float(processors)) * numSeqs);
+ lines[i].start = int (sqrt(float(i)/float(processors)) * numSeqs);
+ lines[i].end = int (sqrt(float(i+1)/float(processors)) * numSeqs);
}else{
- lines[i]->start = int ((float(i)/float(processors)) * numSeqs);
- lines[i]->end = int ((float(i+1)/float(processors)) * numSeqs);
+ lines[i].start = int ((float(i)/float(processors)) * numSeqs);
+ lines[i].end = int ((float(i+1)/float(processors)) * numSeqs);
}
+
}
-
- createProcesses(outputFile);
-
- map<int, int>::iterator it = processIDS.begin();
- rename((outputFile + toString(it->second) + ".temp").c_str(), outputFile.c_str());
- it++;
- //append and remove temp files
- for (; it != processIDS.end(); it++) {
- m->appendFiles((outputFile + toString(it->second) + ".temp"), outputFile);
- remove((outputFile + toString(it->second) + ".temp").c_str());
- }
+ createProcesses(outputFile);
}
#else
//ifstream inFASTA;
#endif
#endif
- if (m->control_pressed) { delete distCalculator; remove(outputFile.c_str()); return 0; }
+ if (m->control_pressed) { outputTypes.clear(); delete distCalculator; m->mothurRemove(outputFile); return 0; }
#ifdef USE_MPI
MPI_Comm_rank(MPI_COMM_WORLD, &pid);
if (outputFile == column) {
string tempcolumn = column + ".old";
m->appendFiles(tempcolumn, outputFile);
- remove(tempcolumn.c_str());
+ m->mothurRemove(tempcolumn);
}else{
m->appendFiles(outputFile, column);
- remove(outputFile.c_str());
+ m->mothurRemove(outputFile);
outputFile = column;
}
+
+ if (outputDir != "") {
+ string newOutputName = outputDir + m->getSimpleName(outputFile);
+ rename(outputFile.c_str(), newOutputName.c_str());
+ m->mothurRemove(outputFile);
+ outputFile = newOutputName;
+ }
}
}
#endif
- if (m->control_pressed) { delete distCalculator; remove(outputFile.c_str()); return 0; }
+ if (m->control_pressed) { outputTypes.clear(); delete distCalculator; m->mothurRemove(outputFile); return 0; }
delete distCalculator;
+ //set phylip file as new current phylipfile
+ string current = "";
+ itTypes = outputTypes.find("phylip");
+ if (itTypes != outputTypes.end()) {
+ if ((itTypes->second).size() != 0) { current = (itTypes->second)[0]; m->setPhylipFile(current); }
+ }
+
+ //set column file as new current columnfile
+ itTypes = outputTypes.find("column");
+ if (itTypes != outputTypes.end()) {
+ if ((itTypes->second).size() != 0) { current = (itTypes->second)[0]; m->setColumnFile(current); }
+ }
+
m->mothurOutEndLine();
m->mothurOut("Output File Name: "); m->mothurOutEndLine();
m->mothurOut(outputFile); m->mothurOutEndLine();
m->mothurOutEndLine();
m->mothurOut("It took " + toString(time(NULL) - startTime) + " to calculate the distances for " + toString(numSeqs) + " sequences."); m->mothurOutEndLine();
+
+
+ if (m->isTrue(compress)) {
+ m->mothurOut("Compressing..."); m->mothurOutEndLine();
+ m->mothurOut("(Replacing " + outputFile + " with " + outputFile + ".gz)"); m->mothurOutEndLine();
+ system(("gzip -v " + outputFile).c_str());
+ outputNames.push_back(outputFile + ".gz");
+ }else { outputNames.push_back(outputFile); }
+
return 0;
}
void DistanceCommand::createProcesses(string filename) {
try {
#if defined (__APPLE__) || (__MACH__) || (linux) || (__linux)
- int process = 0;
+ int process = 1;
processIDS.clear();
//loop through and create all the processes you want
int pid = fork();
if (pid > 0) {
- processIDS[lines[process]->end] = pid; //create map from line number to pid so you can append files in correct order later
+ processIDS.push_back(pid); //create map from line number to pid so you can append files in correct order later
process++;
}else if (pid == 0){
- if (output != "square") { driver(lines[process]->start, lines[process]->end, filename + toString(getpid()) + ".temp", cutoff); }
- else { driver(lines[process]->start, lines[process]->end, filename + toString(getpid()) + ".temp", "square"); }
+ if (output != "square") { driver(lines[process].start, lines[process].end, filename + toString(getpid()) + ".temp", cutoff); }
+ else { driver(lines[process].start, lines[process].end, filename + toString(getpid()) + ".temp", "square"); }
exit(0);
- }else { m->mothurOut("unable to spawn the necessary processes."); m->mothurOutEndLine(); exit(0); }
+ }else {
+ m->mothurOut("[ERROR]: unable to spawn the necessary processes. Error code: " + toString(pid)); m->mothurOutEndLine();
+ perror(" : ");
+ for (int i=0;i<processIDS.size();i++) { int temp = processIDS[i]; kill (temp, SIGINT); }
+ exit(0);
+ }
}
-
+
+ //parent does its part
+ if (output != "square") { driver(lines[0].start, lines[0].end, filename, cutoff); }
+ else { driver(lines[0].start, lines[0].end, filename, "square"); }
+
+
//force parent to wait until all the processes are done
- for (map<int, int>::iterator it = processIDS.begin(); it != processIDS.end(); it++) {
- int temp = it->second;
+ for (int i=0;i<processIDS.size();i++) {
+ int temp = processIDS[i];
wait(&temp);
}
+
+ //append and remove temp files
+ for (int i=0;i<processIDS.size();i++) {
+ m->appendFiles((filename + toString(processIDS[i]) + ".temp"), filename);
+ m->mothurRemove((filename + toString(processIDS[i]) + ".temp"));
+ }
#endif
}
catch(exception& e) {
//m->openInputFile(outfile, in);
while(!in.eof()) {
- if (m->control_pressed) { in.close(); remove(outfile.c_str()); out.close(); return 0; }
+ if (m->control_pressed) { in.close(); m->mothurRemove(outfile); out.close(); return 0; }
in >> first >> second >> dist; m->gobble(in);
in.close();
out.close();
- remove(outfile.c_str());
+ m->mothurRemove(outfile);
return 1;
//m->openInputFile(outfile, in);
while(!in.eof()) {
- if (m->control_pressed) { in.close(); remove(outfile.c_str()); out.close(); return 0; }
+ if (m->control_pressed) { in.close(); m->mothurRemove(outfile); out.close(); return 0; }
in >> first >> second >> dist; m->gobble(in);
in.close();
out.close();
- remove(outfile.c_str());
+ m->mothurRemove(outfile);
return 1;
string name1, name2;
float dist;
while (!inDist.eof()) {
- if (m->control_pressed) { inDist.close(); outDist.close(); remove(outputFile.c_str()); return good; }
+ if (m->control_pressed) { inDist.close(); outDist.close(); m->mothurRemove(outputFile); return good; }
inDist >> name1 >> name2 >> dist; m->gobble(inDist);
outDist.close();
if (good) {
- remove(column.c_str());
+ m->mothurRemove(column);
rename(outputFile.c_str(), column.c_str());
}else{
- remove(outputFile.c_str()); //temp file is bad because file mismatch above
+ m->mothurRemove(outputFile); //temp file is bad because file mismatch above
}
+ return good;
+
}
catch(exception& e) {
- m->errorOut(e, "DistanceCommand", "m->appendFiles");
+ m->errorOut(e, "DistanceCommand", "sanityCheck");
exit(1);
}
}