X-Git-Url: https://git.donarmstrong.com/?a=blobdiff_plain;f=chimeracheckcommand.cpp;h=c8761b9bd51ea60fed4e9f5bc1636a8b0afb1987;hb=2c5386dc0cb6a0e42e8d7256bd6fb18ab66e4480;hp=25f9c0fe4fa68425b9fcf6f407357fbc79027cfb;hpb=52ea8f3a52c05331b9bd7519ae0b518bda233d05;p=mothur.git diff --git a/chimeracheckcommand.cpp b/chimeracheckcommand.cpp index 25f9c0f..c8761b9 100644 --- a/chimeracheckcommand.cpp +++ b/chimeracheckcommand.cpp @@ -8,7 +8,6 @@ */ #include "chimeracheckcommand.h" -#include "chimeracheckrdp.h" //*************************************************************************************************************** @@ -27,7 +26,7 @@ ChimeraCheckCommand::ChimeraCheckCommand(string option) { OptionParser parser(option); map parameters = parser.getParameters(); - ValidParameters validParameter; + ValidParameters validParameter("chimera.check"); map::iterator it; //check to make sure all parameters are valid for command @@ -42,7 +41,7 @@ ChimeraCheckCommand::ChimeraCheckCommand(string option) { it = parameters.find("template"); //user has given a template file if(it != parameters.end()){ - string path = hasPath(it->second); + string path = m->hasPath(it->second); //if the user has not given a path then, add inputdir. else leave path alone. if (path == "") { parameters["template"] = inputDir + it->second; } } @@ -52,12 +51,12 @@ ChimeraCheckCommand::ChimeraCheckCommand(string option) { fastafile = validParameter.validFile(parameters, "fasta", false); if (fastafile == "not found") { fastafile = ""; m->mothurOut("fasta is a required parameter for the chimera.check command."); m->mothurOutEndLine(); abort = true; } else { - splitAtDash(fastafile, fastaFileNames); + m->splitAtDash(fastafile, 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]; } } @@ -65,30 +64,21 @@ ChimeraCheckCommand::ChimeraCheckCommand(string option) { int ableToOpen; ifstream in; - #ifdef USE_MPI - int pid; - MPI_Comm_size(MPI_COMM_WORLD, &processors); //set processors to the number of mpi processes running - MPI_Comm_rank(MPI_COMM_WORLD, &pid); //find out who we are + ableToOpen = m->openInputFile(fastaFileNames[i], in, "noerror"); - if (pid == 0) { - #endif - - ableToOpen = openInputFile(fastaFileNames[i], in); + //if you can't open it, try default location + if (ableToOpen == 1) { + if (m->getDefaultPath() != "") { //default path is set + string tryPath = m->getDefaultPath() + m->getSimpleName(fastaFileNames[i]); + m->mothurOut("Unable to open " + fastaFileNames[i] + ". Trying default " + tryPath); m->mothurOutEndLine(); + ableToOpen = m->openInputFile(tryPath, in, "noerror"); + fastaFileNames[i] = tryPath; + } + } in.close(); - #ifdef USE_MPI - for (int j = 1; j < processors; j++) { - MPI_Send(&ableToOpen, 1, MPI_INT, j, 2001, MPI_COMM_WORLD); - } - }else{ - MPI_Status status; - MPI_Recv(&ableToOpen, 1, MPI_INT, 0, 2001, MPI_COMM_WORLD, &status); - } - - #endif - if (ableToOpen == 1) { - m->mothurOut(fastaFileNames[i] + " will be disregarded."); m->mothurOutEndLine(); + m->mothurOut("Unable to open " + fastaFileNames[i] +". It will be disregarded."); m->mothurOutEndLine(); //erase from file list fastaFileNames.erase(fastaFileNames.begin()+i); i--; @@ -109,12 +99,12 @@ ChimeraCheckCommand::ChimeraCheckCommand(string option) { namefile = validParameter.validFile(parameters, "name", false); if (namefile == "not found") { namefile = ""; } else { - splitAtDash(namefile, nameFileNames); + m->splitAtDash(namefile, nameFileNames); //go through files and make sure they are good, if not, then disregard them for (int i = 0; i < nameFileNames.size(); i++) { if (inputDir != "") { - string path = hasPath(nameFileNames[i]); + string path = m->hasPath(nameFileNames[i]); //if the user has not given a path then, add inputdir. else leave path alone. if (path == "") { nameFileNames[i] = inputDir + nameFileNames[i]; } } @@ -122,30 +112,21 @@ ChimeraCheckCommand::ChimeraCheckCommand(string option) { int ableToOpen; ifstream in; - #ifdef USE_MPI - int pid; - MPI_Comm_size(MPI_COMM_WORLD, &processors); //set processors to the number of mpi processes running - MPI_Comm_rank(MPI_COMM_WORLD, &pid); //find out who we are + ableToOpen = m->openInputFile(nameFileNames[i], in, "noerror"); - if (pid == 0) { - #endif - - ableToOpen = openInputFile(nameFileNames[i], in); + //if you can't open it, try default location + if (ableToOpen == 1) { + if (m->getDefaultPath() != "") { //default path is set + string tryPath = m->getDefaultPath() + m->getSimpleName(nameFileNames[i]); + m->mothurOut("Unable to open " + nameFileNames[i] + ". Trying default " + tryPath); m->mothurOutEndLine(); + ableToOpen = m->openInputFile(tryPath, in, "noerror"); + nameFileNames[i] = tryPath; + } + } in.close(); - #ifdef USE_MPI - for (int j = 1; j < processors; j++) { - MPI_Send(&ableToOpen, 1, MPI_INT, j, 2001, MPI_COMM_WORLD); - } - }else{ - MPI_Status status; - MPI_Recv(&ableToOpen, 1, MPI_INT, 0, 2001, MPI_COMM_WORLD, &status); - } - - #endif - if (ableToOpen == 1) { - m->mothurOut(nameFileNames[i] + " will be disregarded."); m->mothurOutEndLine(); + m->mothurOut("Unable to open " + nameFileNames[i] + ". It will be disregarded."); m->mothurOutEndLine(); //erase from file list nameFileNames.erase(nameFileNames.begin()+i); i--; @@ -169,7 +150,7 @@ ChimeraCheckCommand::ChimeraCheckCommand(string option) { convert(temp, ksize); temp = validParameter.validFile(parameters, "svg", false); if (temp == "not found") { temp = "F"; } - svg = isTrue(temp); + svg = m->isTrue(temp); if (nameFileNames.size() != 0) { svg = true; } temp = validParameter.validFile(parameters, "increment", false); if (temp == "not found") { temp = "10"; } @@ -236,14 +217,15 @@ int ChimeraCheckCommand::execute(){ if (m->control_pressed) { delete chimera; return 0; } - string outputFileName = outputDir + getRootName(getSimpleName(fastaFileNames[i])) + "chimeracheck.chimeras"; + if (outputDir == "") { outputDir = m->hasPath(fastaFileNames[i]); }//if user entered a file with a path then preserve it + string outputFileName = outputDir + m->getRootName(m->getSimpleName(fastaFileNames[i])) + "chimeracheck.chimeras"; outputNames.push_back(outputFileName); #ifdef USE_MPI int pid, end, numSeqsPerProcessor; int tag = 2001; - vector MPIPos; + vector MPIPos; MPI_Status status; MPI_Comm_rank(MPI_COMM_WORLD, &pid); //find out who we are @@ -267,7 +249,7 @@ int ChimeraCheckCommand::execute(){ if (m->control_pressed) { MPI_File_close(&inMPI); MPI_File_close(&outMPI); for (int j = 0; j < outputNames.size(); j++) { remove(outputNames[j].c_str()); } delete chimera; return 0; } if (pid == 0) { //you are the root process - MPIPos = setFilePosFasta(fastaFileNames[i], numSeqs); //fills MPIPos, returns numSeqs + MPIPos = m->setFilePosFasta(fastaFileNames[i], numSeqs); //fills MPIPos, returns numSeqs //send file positions to all processes for(int j = 1; j < processors; j++) { @@ -318,61 +300,29 @@ int ChimeraCheckCommand::execute(){ MPI_Barrier(MPI_COMM_WORLD); //make everyone wait - just in case #else + vector positions = m->divideFile(fastaFileNames[i], processors); + + for (int s = 0; s < (positions.size()-1); s++) { + lines.push_back(new linePair(positions[s], positions[(s+1)])); + } + //break up file #if defined (__APPLE__) || (__MACH__) || (linux) || (__linux) if(processors == 1){ - ifstream inFASTA; - openInputFile(fastaFileNames[i], inFASTA); - getNumSeqs(inFASTA, numSeqs); - inFASTA.close(); + numSeqs = driver(lines[0], outputFileName, fastaFileNames[i]); - lines.push_back(new linePair(0, numSeqs)); - - driver(lines[0], outputFileName, fastaFileNames[i]); - - if (m->control_pressed) { - for (int j = 0; j < outputNames.size(); j++) { remove(outputNames[j].c_str()); } - for (int j = 0; j < lines.size(); j++) { delete lines[j]; } lines.clear(); - delete chimera; - return 0; - } + if (m->control_pressed) { for (int j = 0; j < outputNames.size(); j++) { remove(outputNames[j].c_str()); } for (int j = 0; j < lines.size(); j++) { delete lines[j]; } lines.clear(); delete chimera; return 0; } }else{ - vector positions; processIDS.resize(0); - ifstream inFASTA; - openInputFile(fastaFileNames[i], inFASTA); - - string input; - while(!inFASTA.eof()){ - input = getline(inFASTA); - if (input.length() != 0) { - if(input[0] == '>'){ long int pos = inFASTA.tellg(); positions.push_back(pos - input.length() - 1); } - } - } - inFASTA.close(); - - numSeqs = positions.size(); - - int numSeqsPerProcessor = numSeqs / processors; - - for (int j = 0; j < processors; j++) { - long int startPos = positions[ j * numSeqsPerProcessor ]; - if(j == processors - 1){ - numSeqsPerProcessor = numSeqs - j * numSeqsPerProcessor; - } - lines.push_back(new linePair(startPos, numSeqsPerProcessor)); - } - - - createProcesses(outputFileName, fastaFileNames[i]); + numSeqs = createProcesses(outputFileName, fastaFileNames[i]); rename((outputFileName + toString(processIDS[0]) + ".temp").c_str(), outputFileName.c_str()); //append output files for(int j=1;jappendFiles((outputFileName + toString(processIDS[j]) + ".temp"), outputFileName); remove((outputFileName + toString(processIDS[j]) + ".temp").c_str()); } @@ -385,20 +335,9 @@ int ChimeraCheckCommand::execute(){ } #else - ifstream inFASTA; - openInputFile(fastaFileNames[i], inFASTA); - getNumSeqs(inFASTA, numSeqs); - inFASTA.close(); - lines.push_back(new linePair(0, numSeqs)); - - driver(lines[0], outputFileName, fastaFileNames[i]); + numSeqs = driver(lines[0], outputFileName, fastaFileNames[i]); - if (m->control_pressed) { - for (int j = 0; j < lines.size(); j++) { delete lines[j]; } lines.clear(); - for (int j = 0; j < outputNames.size(); j++) { remove(outputNames[j].c_str()); } - delete chimera; - return 0; - } + if (m->control_pressed) { for (int j = 0; j < lines.size(); j++) { delete lines[j]; } lines.clear(); for (int j = 0; j < outputNames.size(); j++) { remove(outputNames[j].c_str()); } delete chimera; return 0; } #endif #endif delete chimera; @@ -424,23 +363,26 @@ int ChimeraCheckCommand::execute(){ } //********************************************************************************************************************** -int ChimeraCheckCommand::driver(linePair* line, string outputFName, string filename){ +int ChimeraCheckCommand::driver(linePair* filePos, string outputFName, string filename){ try { ofstream out; - openOutputFile(outputFName, out); + m->openOutputFile(outputFName, out); ofstream out2; ifstream inFASTA; - openInputFile(filename, inFASTA); + m->openInputFile(filename, inFASTA); + + inFASTA.seekg(filePos->start); + + bool done = false; + int count = 0; + + while (!done) { - inFASTA.seekg(line->start); - - for(int i=0;inumSeqs;i++){ - if (m->control_pressed) { return 1; } - Sequence* candidateSeq = new Sequence(inFASTA); gobble(inFASTA); + Sequence* candidateSeq = new Sequence(inFASTA); m->gobble(inFASTA); if (candidateSeq->getName() != "") { //incase there is a commented sequence at the end of a file //find chimeras @@ -453,16 +395,23 @@ int ChimeraCheckCommand::driver(linePair* line, string outputFName, string filen } delete candidateSeq; + #if defined (__APPLE__) || (__MACH__) || (linux) || (__linux) + unsigned long int pos = inFASTA.tellg(); + if ((pos == -1) || (pos >= filePos->end)) { break; } + #else + if (inFASTA.eof()) { break; } + #endif + //report progress - if((i+1) % 100 == 0){ m->mothurOut("Processing sequence: " + toString(i+1)); m->mothurOutEndLine(); } + if((count) % 100 == 0){ m->mothurOut("Processing sequence: " + toString(count)); m->mothurOutEndLine(); } } //report progress - if((line->numSeqs) % 100 != 0){ m->mothurOut("Processing sequence: " + toString(line->numSeqs)); m->mothurOutEndLine(); } + if((count) % 100 != 0){ m->mothurOut("Processing sequence: " + toString(count)); m->mothurOutEndLine(); } out.close(); inFASTA.close(); - return 0; + return count; } catch(exception& e) { m->errorOut(e, "ChimeraCheckCommand", "driver"); @@ -471,7 +420,7 @@ int ChimeraCheckCommand::driver(linePair* line, string outputFName, string filen } //********************************************************************************************************************** #ifdef USE_MPI -int ChimeraCheckCommand::driverMPI(int start, int num, MPI_File& inMPI, MPI_File& outMPI, vector& MPIPos){ +int ChimeraCheckCommand::driverMPI(int start, int num, MPI_File& inMPI, MPI_File& outMPI, vector& MPIPos){ try { MPI_File outAccMPI; MPI_Status status; @@ -493,7 +442,7 @@ int ChimeraCheckCommand::driverMPI(int start, int num, MPI_File& inMPI, MPI_File istringstream iss (tempBuf,istringstream::in); delete buf4; - Sequence* candidateSeq = new Sequence(iss); gobble(iss); + Sequence* candidateSeq = new Sequence(iss); m->gobble(iss); if (candidateSeq->getName() != "") { //incase there is a commented sequence at the end of a file //find chimeras @@ -525,7 +474,7 @@ int ChimeraCheckCommand::createProcesses(string outputFileName, string filename) try { #if defined (__APPLE__) || (__MACH__) || (linux) || (__linux) int process = 0; - // processIDS.resize(0); + int num = 0; //loop through and create all the processes you want while (process != processors) { @@ -535,7 +484,15 @@ int ChimeraCheckCommand::createProcesses(string outputFileName, 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){ - driver(lines[process], outputFileName + toString(getpid()) + ".temp", filename); + num = driver(lines[process], outputFileName + toString(getpid()) + ".temp", filename); + + //pass numSeqs to parent + ofstream out; + string tempFile = outputFileName + 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); } } @@ -546,7 +503,15 @@ int ChimeraCheckCommand::createProcesses(string outputFileName, string filename) wait(&temp); } - return 0; + for (int i = 0; i < processIDS.size(); i++) { + ifstream in; + string tempFile = outputFileName + 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()); + } + + return num; #endif } catch(exception& e) {