X-Git-Url: https://git.donarmstrong.com/?a=blobdiff_plain;f=distancecommand.cpp;h=b69b93d85c44c1fa89d08d49c0fb478b2f0cb242;hb=19fcbbdba99658f5eca244803280f9ee7f9f6607;hp=60e90500866673aa8e643e024c3c60b840140dce;hpb=f663afa231c9bc1b5e18e0ea3bdd2b2ee784f5b2;p=mothur.git diff --git a/distancecommand.cpp b/distancecommand.cpp index 60e9050..b69b93d 100644 --- a/distancecommand.cpp +++ b/distancecommand.cpp @@ -15,20 +15,80 @@ #include "onegapignore.h" //********************************************************************************************************************** - +vector 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 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 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","compress"}; - - vector myArray (Array, Array+(sizeof(Array)/sizeof(string))); + vector myArray = setParameters(); OptionParser parser(option); map parameters = parser.getParameters(); @@ -41,6 +101,11 @@ DistanceCommand::DistanceCommand(string option) { if (validParameter.isValidParameter(it2->first, myArray, it2->second) != true) { abort = true; } } + //initialize outputTypes + vector 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 = ""; } @@ -73,13 +138,21 @@ DistanceCommand::DistanceCommand(string option) { //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); @@ -89,6 +162,7 @@ DistanceCommand::DistanceCommand(string option) { 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"){ @@ -112,8 +186,9 @@ DistanceCommand::DistanceCommand(string option) { 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); @@ -154,47 +229,12 @@ DistanceCommand::DistanceCommand(string option) { exit(1); } } - -//********************************************************************************************************************** - -DistanceCommand::~DistanceCommand(){ - - for(int i=0;imothurOut("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, compress, 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 compress parameter allows you to indicate that you want the resulting distance file compressed. The default is false.\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); @@ -209,15 +249,18 @@ int DistanceCommand::execute(){ 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) { @@ -225,10 +268,11 @@ int DistanceCommand::execute(){ 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); } @@ -264,17 +308,17 @@ int DistanceCommand::execute(){ //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[5]; MPI_Recv(buf, 5, MPI_CHAR, i, tag, MPI_COMM_WORLD, &status); @@ -283,7 +327,7 @@ int DistanceCommand::execute(){ //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[5]; strcpy(buf, "done"); @@ -303,7 +347,7 @@ int DistanceCommand::execute(){ 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; @@ -322,7 +366,7 @@ int DistanceCommand::execute(){ 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); @@ -368,28 +412,35 @@ int DistanceCommand::execute(){ 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 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::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; @@ -398,7 +449,7 @@ int DistanceCommand::execute(){ #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); @@ -421,17 +472,17 @@ int DistanceCommand::execute(){ 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()); - remove(outputFile.c_str()); + m->mothurRemove(outputFile); outputFile = newOutputName; } } @@ -441,10 +492,23 @@ int DistanceCommand::execute(){ } #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(); @@ -456,7 +520,8 @@ int DistanceCommand::execute(){ 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; @@ -470,7 +535,7 @@ int DistanceCommand::execute(){ 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 @@ -478,20 +543,36 @@ void DistanceCommand::createProcesses(string filename) { 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("[ERROR]: unable to spawn the necessary processes. Error code: " + toString(pid)); m->mothurOutEndLine(); + perror(" : "); + for (int i=0;imothurOut("unable to spawn the necessary processes."); m->mothurOutEndLine(); 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::iterator it = processIDS.begin(); it != processIDS.end(); it++) { - int temp = it->second; + for (int i=0;iappendFiles((filename + toString(processIDS[i]) + ".temp"), filename); + m->mothurRemove((filename + toString(processIDS[i]) + ".temp")); + } #endif } catch(exception& e) { @@ -854,7 +935,7 @@ int DistanceCommand::convertMatrix(string outputFile) { //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); @@ -889,7 +970,7 @@ int DistanceCommand::convertMatrix(string outputFile) { in.close(); out.close(); - remove(outfile.c_str()); + m->mothurRemove(outfile); return 1; @@ -944,7 +1025,7 @@ int DistanceCommand::convertToLowerTriangle(string outputFile) { //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); @@ -983,7 +1064,7 @@ int DistanceCommand::convertToLowerTriangle(string outputFile) { in.close(); out.close(); - remove(outfile.c_str()); + m->mothurRemove(outfile); return 1; @@ -1054,7 +1135,7 @@ bool DistanceCommand::sanityCheck() { 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); @@ -1071,15 +1152,17 @@ bool DistanceCommand::sanityCheck() { 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); } }