X-Git-Url: https://git.donarmstrong.com/?a=blobdiff_plain;f=filterseqscommand.cpp;h=a677eef9729ef5f1035db03f0a5daff9f7280e36;hb=836150c0a3666899ad58426388f4999d6cf8829a;hp=f3b543c7636af24e333dcf301d897fa86beb9746;hpb=94ba64671398f3c9e591df90536833123de0fd34;p=mothur.git diff --git a/filterseqscommand.cpp b/filterseqscommand.cpp index f3b543c..a677eef 100644 --- a/filterseqscommand.cpp +++ b/filterseqscommand.cpp @@ -10,20 +10,77 @@ #include "filterseqscommand.h" #include "sequence.hpp" -/**************************************************************************************/ +//********************************************************************************************************************** +vector FilterSeqsCommand::setParameters(){ + try { + CommandParameter pfasta("fasta", "InputTypes", "", "", "none", "none", "none",false,true); parameters.push_back(pfasta); + CommandParameter phard("hard", "InputTypes", "", "", "none", "none", "none",false,false); parameters.push_back(phard); + CommandParameter ptrump("trump", "String", "", "*", "", "", "",false,false); parameters.push_back(ptrump); + CommandParameter psoft("soft", "Number", "", "0", "", "", "",false,false); parameters.push_back(psoft); + CommandParameter pvertical("vertical", "Boolean", "", "T", "", "", "",false,false); parameters.push_back(pvertical); + CommandParameter pprocessors("processors", "Number", "", "1", "", "", "",false,false); parameters.push_back(pprocessors); + 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, "FilterSeqsCommand", "setParameters"); + exit(1); + } +} +//********************************************************************************************************************** +string FilterSeqsCommand::getHelpString(){ + try { + string helpString = ""; + helpString += "The filter.seqs command reads a file containing sequences and creates a .filter and .filter.fasta file.\n"; + helpString += "The filter.seqs command parameters are fasta, trump, soft, hard, processors and vertical. \n"; + helpString += "The fasta parameter is required, unless you have a valid current fasta file. You may enter several fasta files to build the filter from and filter, by separating their names with -'s.\n"; + helpString += "For example: fasta=abrecovery.fasta-amazon.fasta \n"; + helpString += "The trump option will remove a column if the trump character is found at that position in any sequence of the alignment. Default=*, meaning no trump. \n"; + helpString += "A soft mask removes any column where the dominant base (i.e. A, T, G, C, or U) does not occur in at least a designated percentage of sequences. Default=0.\n"; + helpString += "The hard parameter allows you to enter a file containing the filter you want to use.\n"; + helpString += "The vertical parameter removes columns where all sequences contain a gap character. The default is T.\n"; + helpString += "The processors parameter allows you to specify the number of processors to use. The default is 1.\n"; + helpString += "The filter.seqs command should be in the following format: \n"; + helpString += "filter.seqs(fasta=yourFastaFile, trump=yourTrump) \n"; + helpString += "Example filter.seqs(fasta=abrecovery.fasta, trump=.).\n"; + helpString += "Note: No spaces between parameter labels (i.e. fasta), '=' and parameters (i.e.yourFasta).\n"; + return helpString; + } + catch(exception& e) { + m->errorOut(e, "FilterSeqsCommand", "getHelpString"); + exit(1); + } +} +//********************************************************************************************************************** +FilterSeqsCommand::FilterSeqsCommand(){ + try { + abort = true; calledHelp = true; + setParameters(); + vector tempOutNames; + outputTypes["fasta"] = tempOutNames; + outputTypes["filter"] = tempOutNames; + } + catch(exception& e) { + m->errorOut(e, "FilterSeqsCommand", "FilterSeqsCommand"); + exit(1); + } +} +/**************************************************************************************/ FilterSeqsCommand::FilterSeqsCommand(string option) { try { - abort = false; + abort = false; calledHelp = false; filterFileName = ""; //allow user to run help - if(option == "help") { help(); abort = true; } + if(option == "help") { help(); abort = true; calledHelp = true; } else { - //valid paramters for this command - string Array[] = {"fasta", "trump", "soft", "hard", "vertical", "outputdir","inputdir", "processors"}; - vector myArray (Array, Array+(sizeof(Array)/sizeof(string))); + vector myArray = setParameters(); OptionParser parser(option); map parameters = parser.getParameters(); @@ -36,6 +93,11 @@ FilterSeqsCommand::FilterSeqsCommand(string option) { if (validParameter.isValidParameter(it->first, myArray, it->second) != true) { abort = true; } } + //initialize outputTypes + vector tempOutNames; + outputTypes["fasta"] = tempOutNames; + outputTypes["filter"] = 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 = ""; } @@ -44,7 +106,7 @@ FilterSeqsCommand::FilterSeqsCommand(string option) { it = parameters.find("fasta"); //user has given a template file if(it != parameters.end()){ - path = hasPath(it->second); + path = m->hasPath(it->second); //if the user has not given a path then, add inputdir. else leave path alone. if (path == "") { parameters["fasta"] = inputDir + it->second; } } @@ -52,7 +114,7 @@ FilterSeqsCommand::FilterSeqsCommand(string option) { it = parameters.find("hard"); //user has given a template file if(it != parameters.end()){ - path = hasPath(it->second); + path = m->hasPath(it->second); //if the user has not given a path then, add inputdir. else leave path alone. if (path == "") { parameters["hard"] = inputDir + it->second; } } @@ -60,30 +122,49 @@ FilterSeqsCommand::FilterSeqsCommand(string option) { //check for required parameters fasta = validParameter.validFile(parameters, "fasta", false); - if (fasta == "not found") { m->mothurOut("fasta is a required parameter for the filter.seqs command."); m->mothurOutEndLine(); abort = true; } + if (fasta == "not found") { + fasta = m->getFastaFile(); + if (fasta != "") { fastafileNames.push_back(fasta); m->mothurOut("Using " + fasta + " as input file for the fasta parameter."); m->mothurOutEndLine(); } + else { m->mothurOut("You have no current fastafile and the fasta parameter is required."); m->mothurOutEndLine(); abort = true; } + } else { - splitAtDash(fasta, fastafileNames); + m->splitAtDash(fasta, fastafileNames); //go through files and make sure they are good, if not, then disregard them for (int i = 0; i < fastafileNames.size(); i++) { if (inputDir != "") { - string path = hasPath(fastafileNames[i]); + string path = m->hasPath(fastafileNames[i]); //if the user has not given a path then, add inputdir. else leave path alone. if (path == "") { fastafileNames[i] = inputDir + fastafileNames[i]; } } ifstream in; - int ableToOpen = openInputFile(fastafileNames[i], in, "noerror"); + int ableToOpen = m->openInputFile(fastafileNames[i], in, "noerror"); //if you can't open it, try default location if (ableToOpen == 1) { if (m->getDefaultPath() != "") { //default path is set - string tryPath = m->getDefaultPath() + getSimpleName(fastafileNames[i]); + string tryPath = m->getDefaultPath() + m->getSimpleName(fastafileNames[i]); m->mothurOut("Unable to open " + fastafileNames[i] + ". Trying default " + tryPath); m->mothurOutEndLine(); - ableToOpen = openInputFile(tryPath, in, "noerror"); + ifstream in2; + ableToOpen = m->openInputFile(tryPath, in2, "noerror"); + in2.close(); fastafileNames[i] = tryPath; } } + + //if you can't open it, try default location + if (ableToOpen == 1) { + if (m->getOutputDir() != "") { //default path is set + string tryPath = m->getOutputDir() + m->getSimpleName(fastafileNames[i]); + m->mothurOut("Unable to open " + fastafileNames[i] + ". Trying output directory " + tryPath); m->mothurOutEndLine(); + ifstream in2; + ableToOpen = m->openInputFile(tryPath, in2, "noerror"); + in2.close(); + fastafileNames[i] = tryPath; + } + } + in.close(); if (ableToOpen == 1) { @@ -92,7 +173,7 @@ FilterSeqsCommand::FilterSeqsCommand(string option) { fastafileNames.erase(fastafileNames.begin()+i); i--; }else{ - string simpleName = getSimpleName(fastafileNames[i]); + string simpleName = m->getSimpleName(fastafileNames[i]); filterFileName += simpleName.substr(0, simpleName.find_first_of('.')); } in.close(); @@ -106,7 +187,7 @@ FilterSeqsCommand::FilterSeqsCommand(string option) { //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"){ outputDir = ""; - outputDir += hasPath(fastafileNames[0]); //if user entered a file with a path then preserve it + outputDir += m->hasPath(fastafileNames[0]); //if user entered a file with a path then preserve it } } //check for optional parameter and set defaults @@ -122,7 +203,8 @@ FilterSeqsCommand::FilterSeqsCommand(string option) { temp = validParameter.validFile(parameters, "soft", false); if (temp == "not found") { soft = 0; } else { soft = (float)atoi(temp.c_str()) / 100.0; } - temp = validParameter.validFile(parameters, "processors", false); if (temp == "not found"){ temp = "1"; } + temp = validParameter.validFile(parameters, "processors", false); if (temp == "not found"){ temp = m->getProcessors(); } + m->setProcessors(temp); convert(temp, processors); vertical = validParameter.validFile(parameters, "vertical", false); @@ -140,41 +222,15 @@ FilterSeqsCommand::FilterSeqsCommand(string option) { exit(1); } } - -//********************************************************************************************************************** - -void FilterSeqsCommand::help(){ - try { - - m->mothurOut("The filter.seqs command reads a file containing sequences and creates a .filter and .filter.fasta file.\n"); - m->mothurOut("The filter.seqs command parameters are fasta, trump, soft, hard and vertical. \n"); - m->mothurOut("The fasta parameter is required. You may enter several fasta files to build the filter from and filter, by separating their names with -'s.\n"); - m->mothurOut("For example: fasta=abrecovery.fasta-amazon.fasta \n"); - m->mothurOut("The trump parameter .... The default is ...\n"); - m->mothurOut("The soft parameter .... The default is ....\n"); - m->mothurOut("The hard parameter allows you to enter a file containing the filter you want to use.\n"); - m->mothurOut("The vertical parameter removes columns where all sequences contain a gap character. The default is T.\n"); - m->mothurOut("The filter.seqs command should be in the following format: \n"); - m->mothurOut("filter.seqs(fasta=yourFastaFile, trump=yourTrump) \n"); - m->mothurOut("Example filter.seqs(fasta=abrecovery.fasta, trump=.).\n"); - m->mothurOut("Note: No spaces between parameter labels (i.e. fasta), '=' and parameters (i.e.yourFasta).\n\n"); - - } - catch(exception& e) { - m->errorOut(e, "FilterSeqsCommand", "help"); - exit(1); - } -} - /**************************************************************************************/ int FilterSeqsCommand::execute() { try { - if (abort == true) { return 0; } + if (abort == true) { if (calledHelp) { return 0; } return 2; } ifstream inFASTA; - openInputFile(fastafileNames[0], inFASTA); + m->openInputFile(fastafileNames[0], inFASTA); Sequence testSeq(inFASTA); alignmentLength = testSeq.getAlignLength(); @@ -187,7 +243,7 @@ int FilterSeqsCommand::execute() { m->mothurOutEndLine(); m->mothurOutEndLine(); - if (m->control_pressed) { return 0; } + if (m->control_pressed) { outputTypes.clear(); return 0; } #ifdef USE_MPI int pid; @@ -198,11 +254,15 @@ int FilterSeqsCommand::execute() { ofstream outFilter; - string filterFile = outputDir + filterFileName + ".filter"; - openOutputFile(filterFile, outFilter); + //prevent giantic file name + string filterFile; + if (fastafileNames.size() > 3) { filterFile = outputDir + "merge.filter"; } + else { filterFile = outputDir + filterFileName + ".filter"; } + + m->openOutputFile(filterFile, outFilter); outFilter << filter << endl; outFilter.close(); - outputNames.push_back(filterFile); + outputNames.push_back(filterFile); outputTypes["filter"].push_back(filterFile); #ifdef USE_MPI } @@ -221,7 +281,7 @@ int FilterSeqsCommand::execute() { if(filter[i] == '1'){ filteredLength++; } } - if (m->control_pressed) { for(int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); } return 0; } + if (m->control_pressed) { outputTypes.clear(); for(int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); } return 0; } m->mothurOutEndLine(); @@ -230,6 +290,12 @@ int FilterSeqsCommand::execute() { m->mothurOut("Length of the original alignment: " + toString(alignmentLength)); m->mothurOutEndLine(); m->mothurOut("Number of sequences used to construct filter: " + toString(numSeqs)); m->mothurOutEndLine(); + //set fasta file as new current fastafile + string current = ""; + itTypes = outputTypes.find("fasta"); + if (itTypes != outputTypes.end()) { + if ((itTypes->second).size() != 0) { current = (itTypes->second)[0]; m->setFastaFile(current); } + } m->mothurOutEndLine(); m->mothurOut("Output File Names: "); m->mothurOutEndLine(); @@ -254,9 +320,9 @@ int FilterSeqsCommand::filterSequences() { for (int i = 0; i < lines.size(); i++) { delete lines[i]; } lines.clear(); - string filteredFasta = outputDir + getRootName(getSimpleName(fastafileNames[s])) + "filter.fasta"; + string filteredFasta = outputDir + m->getRootName(m->getSimpleName(fastafileNames[s])) + "filter.fasta"; #ifdef USE_MPI - int pid, start, end, numSeqsPerProcessor, num; + int pid, numSeqsPerProcessor, num; int tag = 2001; vectorMPIPos; @@ -265,7 +331,6 @@ int FilterSeqsCommand::filterSequences() { MPI_Comm_rank(MPI_COMM_WORLD, &pid); //find out who we are MPI_File outMPI; - MPI_File tempMPI; MPI_File inMPI; int outMode=MPI_MODE_CREATE|MPI_MODE_WRONLY; int inMode=MPI_MODE_RDONLY; @@ -283,7 +348,7 @@ int FilterSeqsCommand::filterSequences() { if (pid == 0) { //you are the root process - MPIPos = setFilePosFasta(fastafileNames[s], num); //fills MPIPos, returns numSeqs + MPIPos = m->setFilePosFasta(fastafileNames[s], num); //fills MPIPos, returns numSeqs numSeqs += num; //send file positions to all processes @@ -305,8 +370,8 @@ int FilterSeqsCommand::filterSequences() { //wait on chidren for(int i = 1; i < processors; i++) { - 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 @@ -326,11 +391,11 @@ int FilterSeqsCommand::filterSequences() { if (m->control_pressed) { MPI_File_close(&inMPI); MPI_File_close(&outMPI); 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); @@ -338,7 +403,7 @@ int FilterSeqsCommand::filterSequences() { MPI_Barrier(MPI_COMM_WORLD); //make everyone wait - just in case #else - vector positions = divideFile(fastafileNames[s], processors); + vector positions = m->divideFile(fastafileNames[s], processors); for (int i = 0; i < (positions.size()-1); i++) { lines.push_back(new linePair(positions[i], positions[(i+1)])); @@ -355,20 +420,20 @@ int FilterSeqsCommand::filterSequences() { //append fasta files for(int i=1;iappendFiles((fastafileNames[s] + toString(processIDS[i]) + ".temp"), filteredFasta); remove((fastafileNames[s] + toString(processIDS[i]) + ".temp").c_str()); } } if (m->control_pressed) { return 1; } #else - numFastaSeqs = driverRunFilter(filter, filteredFasta, fastafileNames[s], lines[0]); + int numFastaSeqs = driverRunFilter(filter, filteredFasta, fastafileNames[s], lines[0]); numSeqs += numFastaSeqs; if (m->control_pressed) { return 1; } #endif #endif - outputNames.push_back(filteredFasta); + outputNames.push_back(filteredFasta); outputTypes["fasta"].push_back(filteredFasta); } return 0; @@ -400,7 +465,7 @@ int FilterSeqsCommand::driverMPIRun(int start, int num, MPI_File& inMPI, MPI_Fil istringstream iss (tempBuf,istringstream::in); delete buf4; - Sequence seq(iss); gobble(iss); + Sequence seq(iss); m->gobble(iss); if (seq.getName() != "") { string align = seq.getAligned(); @@ -456,10 +521,10 @@ int FilterSeqsCommand::driverMPIRun(int start, int num, MPI_File& inMPI, MPI_Fil int FilterSeqsCommand::driverRunFilter(string F, string outputFilename, string inputFilename, linePair* filePos) { try { ofstream out; - openOutputFile(outputFilename, out); + m->openOutputFile(outputFilename, out); ifstream in; - openInputFile(inputFilename, in); + m->openInputFile(inputFilename, in); in.seekg(filePos->start); @@ -470,7 +535,7 @@ int FilterSeqsCommand::driverRunFilter(string F, string outputFilename, string i if (m->control_pressed) { in.close(); out.close(); return 0; } - Sequence seq(in); gobble(in); + Sequence seq(in); m->gobble(in); if (seq.getName() != "") { string align = seq.getAligned(); string filterSeq = ""; @@ -485,8 +550,12 @@ int FilterSeqsCommand::driverRunFilter(string F, string outputFilename, string i count++; } - unsigned long int pos = in.tellg(); - if ((pos == -1) || (pos >= filePos->end)) { break; } + #if defined (__APPLE__) || (__MACH__) || (linux) || (__linux) + unsigned long int pos = in.tellg(); + if ((pos == -1) || (pos >= filePos->end)) { break; } + #else + if (in.eof()) { break; } + #endif //report progress if((count) % 100 == 0){ m->mothurOut(toString(count)); m->mothurOutEndLine(); } @@ -527,13 +596,17 @@ int FilterSeqsCommand::createProcessesRunFilter(string F, string filename) { //pass numSeqs to parent ofstream out; - string tempFile = toString(getpid()) + ".temp"; - openOutputFile(tempFile, out); + string tempFile = filename + toString(getpid()) + ".num.temp"; + m->openOutputFile(tempFile, out); out << num << endl; out.close(); 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."); m->mothurOutEndLine(); + for (int i = 0; i < processIDS.size(); i++) { kill (processIDS[i], SIGINT); } + exit(0); + } } //force parent to wait until all the processes are done @@ -544,8 +617,8 @@ int FilterSeqsCommand::createProcessesRunFilter(string F, string filename) { for (int i = 0; i < processIDS.size(); i++) { ifstream in; - string tempFile = toString(processIDS[i]) + ".temp"; - openInputFile(tempFile, in); + string tempFile = filename + toString(processIDS[i]) + ".num.temp"; + m->openInputFile(tempFile, in); if (!in.eof()) { int tempNum = 0; in >> tempNum; num += tempNum; } in.close(); remove(tempFile.c_str()); } @@ -570,7 +643,7 @@ string FilterSeqsCommand::createFilter() { F.setLength(alignmentLength); - if(trump != '*' || isTrue(vertical) || soft != 0){ + if(trump != '*' || m->isTrue(vertical) || soft != 0){ F.initialize(); } @@ -578,7 +651,7 @@ string FilterSeqsCommand::createFilter() { else { F.setFilter(string(alignmentLength, '1')); } numSeqs = 0; - if(trump != '*' || isTrue(vertical) || soft != 0){ + if(trump != '*' || m->isTrue(vertical) || soft != 0){ for (int s = 0; s < fastafileNames.size(); s++) { for (int i = 0; i < lines.size(); i++) { delete lines[i]; } lines.clear(); @@ -604,7 +677,7 @@ string FilterSeqsCommand::createFilter() { if (m->control_pressed) { MPI_File_close(&inMPI); return 0; } if (pid == 0) { //you are the root process - MPIPos = setFilePosFasta(fastafileNames[s], num); //fills MPIPos, returns numSeqs + MPIPos = m->setFilePosFasta(fastafileNames[s], num); //fills MPIPos, returns numSeqs numSeqs += num; //send file positions to all processes @@ -646,7 +719,7 @@ string FilterSeqsCommand::createFilter() { MPI_Barrier(MPI_COMM_WORLD); //make everyone wait - just in case #else - vector positions = divideFile(fastafileNames[s], processors); + vector positions = m->divideFile(fastafileNames[s], processors); for (int i = 0; i < (positions.size()-1); i++) { lines.push_back(new linePair(positions[i], positions[(i+1)])); @@ -662,7 +735,7 @@ string FilterSeqsCommand::createFilter() { if (m->control_pressed) { return filterString; } #else - numFastaSeqs = driverCreateFilter(F, fastafileNames[s], lines[0]); + int numFastaSeqs = driverCreateFilter(F, fastafileNames[s], lines[0]); numSeqs += numFastaSeqs; if (m->control_pressed) { return filterString; } #endif @@ -679,7 +752,7 @@ string FilterSeqsCommand::createFilter() { MPI_Comm_rank(MPI_COMM_WORLD, &pid); - if(trump != '*' || isTrue(vertical) || soft != 0){ + if(trump != '*' || m->isTrue(vertical) || soft != 0){ if (pid == 0) { //only one process should output the filter @@ -727,8 +800,9 @@ string FilterSeqsCommand::createFilter() { if (pid == 0) { //only one process should output the filter #endif + F.setNumSeqs(numSeqs); - if(isTrue(vertical) == 1) { F.doVertical(); } + if(m->isTrue(vertical) == 1) { F.doVertical(); } if(soft != 0) { F.doSoft(); } filterString = F.getFilter(); @@ -764,7 +838,7 @@ int FilterSeqsCommand::driverCreateFilter(Filters& F, string filename, linePair* try { ifstream in; - openInputFile(filename, in); + m->openInputFile(filename, in); in.seekg(filePos->start); @@ -775,18 +849,22 @@ int FilterSeqsCommand::driverCreateFilter(Filters& F, string filename, linePair* if (m->control_pressed) { in.close(); return 1; } - Sequence seq(in); gobble(in); + Sequence seq(in); m->gobble(in); if (seq.getName() != "") { if (seq.getAligned().length() != alignmentLength) { m->mothurOut("Sequences are not all the same length, please correct."); m->mothurOutEndLine(); m->control_pressed = true; } if(trump != '*') { F.doTrump(seq); } - if(isTrue(vertical) || soft != 0) { F.getFreqs(seq); } + if(m->isTrue(vertical) || soft != 0) { F.getFreqs(seq); } cout.flush(); count++; } - unsigned long int pos = in.tellg(); - if ((pos == -1) || (pos >= filePos->end)) { break; } + #if defined (__APPLE__) || (__MACH__) || (linux) || (__linux) + unsigned long int pos = in.tellg(); + if ((pos == -1) || (pos >= filePos->end)) { break; } + #else + if (in.eof()) { break; } + #endif //report progress if((count) % 100 == 0){ m->mothurOut(toString(count)); m->mothurOutEndLine(); } @@ -831,7 +909,7 @@ int FilterSeqsCommand::MPICreateFilter(int start, int num, Filters& F, MPI_File& if (seq.getAligned().length() != alignmentLength) { cout << "Alignment length is " << alignmentLength << " and sequence " << seq.getName() << " has length " << seq.getAligned().length() << ", please correct." << endl; exit(1); } if(trump != '*'){ F.doTrump(seq); } - if(isTrue(vertical) || soft != 0){ F.getFreqs(seq); } + if(m->isTrue(vertical) || soft != 0){ F.getFreqs(seq); } cout.flush(); //report progress @@ -854,7 +932,7 @@ int FilterSeqsCommand::MPICreateFilter(int start, int num, Filters& F, MPI_File& int FilterSeqsCommand::createProcessesCreateFilter(Filters& F, string filename) { try { #if defined (__APPLE__) || (__MACH__) || (linux) || (__linux) - int process = 0; + int process = 1; int num = 0; processIDS.clear(); @@ -866,12 +944,19 @@ int FilterSeqsCommand::createProcessesCreateFilter(Filters& F, string filename) 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){ + //reset child's filter counts to 0; + F.a.clear(); F.a.resize(alignmentLength, 0); + F.t.clear(); F.t.resize(alignmentLength, 0); + F.g.clear(); F.g.resize(alignmentLength, 0); + F.c.clear(); F.c.resize(alignmentLength, 0); + F.gap.clear(); F.gap.resize(alignmentLength, 0); + num = driverCreateFilter(F, filename, lines[process]); //write out filter counts to file filename += toString(getpid()) + "filterValues.temp"; ofstream out; - openOutputFile(filename, out); + m->openOutputFile(filename, out); out << num << endl; out << F.getFilter() << endl; @@ -881,38 +966,45 @@ int FilterSeqsCommand::createProcessesCreateFilter(Filters& F, string filename) for (int k = 0; k < alignmentLength; k++) { out << F.c[k] << '\t'; } out << endl; for (int k = 0; k < alignmentLength; k++) { out << F.gap[k] << '\t'; } out << endl; - cout << F.getFilter() << endl; + //cout << F.getFilter() << endl; out.close(); 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."); m->mothurOutEndLine(); + for (int i = 0; i < processIDS.size(); i++) { kill (processIDS[i], SIGINT); } + exit(0); + } } + //parent do your part + num = driverCreateFilter(F, filename, lines[0]); + //force parent to wait until all the processes are done - for (int i=0;iopenInputFile(tempFilename, in); int temp, tempNum; string tempFilterString; - in >> tempNum; gobble(in); num += tempNum; + in >> tempNum; m->gobble(in); num += tempNum; in >> tempFilterString; F.mergeFilter(tempFilterString); - for (int k = 0; k < alignmentLength; k++) { in >> temp; F.a[k] += temp; } gobble(in); - for (int k = 0; k < alignmentLength; k++) { in >> temp; F.t[k] += temp; } gobble(in); - for (int k = 0; k < alignmentLength; k++) { in >> temp; F.g[k] += temp; } gobble(in); - for (int k = 0; k < alignmentLength; k++) { in >> temp; F.c[k] += temp; } gobble(in); - for (int k = 0; k < alignmentLength; k++) { in >> temp; F.gap[k] += temp; } gobble(in); + for (int k = 0; k < alignmentLength; k++) { in >> temp; F.a[k] += temp; } m->gobble(in); + for (int k = 0; k < alignmentLength; k++) { in >> temp; F.t[k] += temp; } m->gobble(in); + for (int k = 0; k < alignmentLength; k++) { in >> temp; F.g[k] += temp; } m->gobble(in); + for (int k = 0; k < alignmentLength; k++) { in >> temp; F.c[k] += temp; } m->gobble(in); + for (int k = 0; k < alignmentLength; k++) { in >> temp; F.gap[k] += temp; } m->gobble(in); in.close(); remove(tempFilename.c_str());