X-Git-Url: https://git.donarmstrong.com/?a=blobdiff_plain;f=chimeraccodecommand.cpp;h=ab10eb8f56bb04dffceb9f4a309ff59157c708d2;hb=2c5386dc0cb6a0e42e8d7256bd6fb18ab66e4480;hp=962c15b3134a836b1e2d25fcbeb983a914bb92fe;hpb=b4f80c1d2be78a8743a408a2b6d462b07f9f71ff;p=mothur.git diff --git a/chimeraccodecommand.cpp b/chimeraccodecommand.cpp index 962c15b..ab10eb8 100644 --- a/chimeraccodecommand.cpp +++ b/chimeraccodecommand.cpp @@ -43,7 +43,7 @@ ChimeraCcodeCommand::ChimeraCcodeCommand(string option) { it = parameters.find("template"); //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["template"] = inputDir + it->second; } } @@ -53,12 +53,12 @@ ChimeraCcodeCommand::ChimeraCcodeCommand(string option) { fastafile = validParameter.validFile(parameters, "fasta", false); if (fastafile == "not found") { fastafile = ""; m->mothurOut("fasta is a required parameter for the chimera.ccode 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]; } } @@ -66,38 +66,19 @@ ChimeraCcodeCommand::ChimeraCcodeCommand(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 - - if (pid == 0) { - #endif - - ableToOpen = openInputFile(fastaFileNames[i], in, "noerror"); + 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"); + 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("Unable to open " + fastaFileNames[i] + ". It will be disregarded."); m->mothurOutEndLine(); //erase from file list @@ -111,10 +92,7 @@ ChimeraCcodeCommand::ChimeraCcodeCommand(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(fastafile); //if user entered a file with a path then preserve it - } + outputDir = validParameter.validFile(parameters, "outputdir", false); if (outputDir == "not found"){ outputDir = ""; } templatefile = validParameter.validFile(parameters, "template", true); if (templatefile == "not open") { abort = true; } @@ -124,20 +102,20 @@ ChimeraCcodeCommand::ChimeraCcodeCommand(string option) { if (maskfile == "not found") { maskfile = ""; } else if (maskfile != "default") { if (inputDir != "") { - string path = hasPath(maskfile); + string path = m->hasPath(maskfile); //if the user has not given a path then, add inputdir. else leave path alone. if (path == "") { maskfile = inputDir + maskfile; } } ifstream in; - int ableToOpen = openInputFile(maskfile, in); + int ableToOpen = m->openInputFile(maskfile, in); if (ableToOpen == 1) { abort = true; } in.close(); } string temp; temp = validParameter.validFile(parameters, "filter", false); if (temp == "not found") { temp = "F"; } - filter = isTrue(temp); + filter = m->isTrue(temp); temp = validParameter.validFile(parameters, "processors", false); if (temp == "not found") { temp = "1"; } convert(temp, processors); @@ -211,16 +189,17 @@ int ChimeraCcodeCommand::execute(){ if (chimera->getUnaligned()) { m->mothurOut("Your template sequences are different lengths, please correct."); m->mothurOutEndLine(); delete chimera; return 0; } templateSeqsLength = chimera->getLength(); + if (outputDir == "") { outputDir = m->hasPath(fastaFileNames[s]); }//if user entered a file with a path then preserve it string outputFileName, accnosFileName; if (maskfile != "") { - outputFileName = outputDir + getRootName(getSimpleName(fastaFileNames[s])) + maskfile + ".ccode.chimeras"; - accnosFileName = outputDir + getRootName(getSimpleName(fastaFileNames[s])) + maskfile + ".ccode.accnos"; + outputFileName = outputDir + m->getRootName(m->getSimpleName(fastaFileNames[s])) + maskfile + ".ccode.chimeras"; + accnosFileName = outputDir + m->getRootName(m->getSimpleName(fastaFileNames[s])) + maskfile + ".ccode.accnos"; }else { - outputFileName = outputDir + getRootName(getSimpleName(fastaFileNames[s])) + "ccode.chimeras"; - accnosFileName = outputDir + getRootName(getSimpleName(fastaFileNames[s])) + "ccode.accnos"; + outputFileName = outputDir + m->getRootName(m->getSimpleName(fastaFileNames[s])) + "ccode.chimeras"; + accnosFileName = outputDir + m->getRootName(m->getSimpleName(fastaFileNames[s])) + "ccode.accnos"; } - string mapInfo = outputDir + getRootName(getSimpleName(fastaFileNames[s])) + "mapinfo"; + string mapInfo = outputDir + m->getRootName(m->getSimpleName(fastaFileNames[s])) + "mapinfo"; if (m->control_pressed) { delete chimera; for (int j = 0; j < outputNames.size(); j++) { remove(outputNames[j].c_str()); } return 0; } @@ -267,7 +246,7 @@ int ChimeraCcodeCommand::execute(){ MPI_File_write_shared(outMPI, buf2, length, MPI_CHAR, &status); delete buf2; - MPIPos = setFilePosFasta(fastaFileNames[s], numSeqs); //fills MPIPos, returns numSeqs + MPIPos = m->setFilePosFasta(fastaFileNames[s], numSeqs); //fills MPIPos, returns numSeqs //send file positions to all processes for(int i = 1; i < processors; i++) { @@ -312,78 +291,44 @@ int ChimeraCcodeCommand::execute(){ #else ofstream outHeader; - string tempHeader = outputDir + getRootName(getSimpleName(fastaFileNames[s])) + maskfile + "ccode.chimeras.tempHeader"; - openOutputFile(tempHeader, outHeader); + string tempHeader = outputDir + m->getRootName(m->getSimpleName(fastaFileNames[s])) + maskfile + "ccode.chimeras.tempHeader"; + m->openOutputFile(tempHeader, outHeader); outHeader << "For full window mapping info refer to " << mapInfo << endl << endl; outHeader.close(); + 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)])); + } + //break up file #if defined (__APPLE__) || (__MACH__) || (linux) || (__linux) if(processors == 1){ - ifstream inFASTA; - openInputFile(fastaFileNames[s], inFASTA); - getNumSeqs(inFASTA, numSeqs); - inFASTA.close(); - - lines.push_back(new linePair(0, numSeqs)); + + numSeqs = driver(lines[0], outputFileName, fastaFileNames[s], accnosFileName); - driver(lines[0], outputFileName, fastaFileNames[s], accnosFileName); - - if (m->control_pressed) { - remove(outputFileName.c_str()); - remove(tempHeader.c_str()); - remove(accnosFileName.c_str()); - for (int j = 0; j < outputNames.size(); j++) { remove(outputNames[j].c_str()); } - for (int i = 0; i < lines.size(); i++) { delete lines[i]; } lines.clear(); - delete chimera; - return 0; - } + if (m->control_pressed) { remove(outputFileName.c_str()); remove(tempHeader.c_str()); remove(accnosFileName.c_str()); for (int j = 0; j < outputNames.size(); j++) { remove(outputNames[j].c_str()); } for (int i = 0; i < lines.size(); i++) { delete lines[i]; } lines.clear(); delete chimera; return 0; } }else{ - vector positions; processIDS.resize(0); - ifstream inFASTA; - openInputFile(fastaFileNames[s], inFASTA); - - string input; - while(!inFASTA.eof()){ - input = getline(inFASTA); - if (input.length() != 0) { - if(input[0] == '>'){ unsigned long int pos = inFASTA.tellg(); positions.push_back(pos - input.length() - 1); } - } - } - inFASTA.close(); - - numSeqs = positions.size(); - - int numSeqsPerProcessor = numSeqs / processors; - - for (int i = 0; i < processors; i++) { - unsigned long int startPos = positions[ i * numSeqsPerProcessor ]; - if(i == processors - 1){ - numSeqsPerProcessor = numSeqs - i * numSeqsPerProcessor; - } - lines.push_back(new linePair(startPos, numSeqsPerProcessor)); - } - - - createProcesses(outputFileName, fastaFileNames[s], accnosFileName); + numSeqs = createProcesses(outputFileName, fastaFileNames[s], accnosFileName); rename((outputFileName + toString(processIDS[0]) + ".temp").c_str(), outputFileName.c_str()); rename((accnosFileName + toString(processIDS[0]) + ".temp").c_str(), accnosFileName.c_str()); //append output files for(int i=1;iappendFiles((outputFileName + toString(processIDS[i]) + ".temp"), outputFileName); remove((outputFileName + toString(processIDS[i]) + ".temp").c_str()); } //append output files for(int i=1;iappendFiles((accnosFileName + toString(processIDS[i]) + ".temp"), accnosFileName); remove((accnosFileName + toString(processIDS[i]) + ".temp").c_str()); } @@ -399,27 +344,13 @@ int ChimeraCcodeCommand::execute(){ } #else - ifstream inFASTA; - openInputFile(fastaFileNames[s], inFASTA); - getNumSeqs(inFASTA, numSeqs); - inFASTA.close(); - lines.push_back(new linePair(0, numSeqs)); + numSeqs = driver(lines[0], outputFileName, fastaFileNames[s], accnosFileName); - driver(lines[0], outputFileName, fastaFileNames[s], accnosFileName); - - if (m->control_pressed) { - remove(outputFileName.c_str()); - remove(tempHeader.c_str()); - remove(accnosFileName.c_str()); - for (int j = 0; j < outputNames.size(); j++) { remove(outputNames[j].c_str()); } - for (int i = 0; i < lines.size(); i++) { delete lines[i]; } lines.clear(); - delete chimera; - return 0; - } + if (m->control_pressed) { remove(outputFileName.c_str()); remove(tempHeader.c_str()); remove(accnosFileName.c_str()); for (int j = 0; j < outputNames.size(); j++) { remove(outputNames[j].c_str()); } for (int i = 0; i < lines.size(); i++) { delete lines[i]; } lines.clear(); delete chimera; return 0; } #endif - appendFiles(outputFileName, tempHeader); + m->appendFiles(outputFileName, tempHeader); remove(outputFileName.c_str()); rename(tempHeader.c_str(), outputFileName.c_str()); @@ -451,24 +382,27 @@ int ChimeraCcodeCommand::execute(){ } //********************************************************************************************************************** -int ChimeraCcodeCommand::driver(linePair* line, string outputFName, string filename, string accnos){ +int ChimeraCcodeCommand::driver(linePair* filePos, string outputFName, string filename, string accnos){ try { ofstream out; - openOutputFile(outputFName, out); + m->openOutputFile(outputFName, out); ofstream out2; - openOutputFile(accnos, out2); + m->openOutputFile(accnos, out2); ifstream inFASTA; - openInputFile(filename, inFASTA); + m->openInputFile(filename, inFASTA); - inFASTA.seekg(line->start); - - for(int i=0;inumSeqs;i++){ + inFASTA.seekg(filePos->start); + + bool done = false; + int count = 0; + + while (!done) { 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 @@ -483,20 +417,28 @@ int ChimeraCcodeCommand::driver(linePair* line, string outputFName, string filen //print results chimera->print(out, out2); } + count++; } 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(); out2.close(); inFASTA.close(); - return 0; + return count; } catch(exception& e) { m->errorOut(e, "ChimeraCcodeCommand", "driver"); @@ -528,7 +470,7 @@ int ChimeraCcodeCommand::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 @@ -568,7 +510,7 @@ int ChimeraCcodeCommand::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) { @@ -578,7 +520,15 @@ int ChimeraCcodeCommand::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, accnos + toString(getpid()) + ".temp"); + num = driver(lines[process], outputFileName + toString(getpid()) + ".temp", filename, accnos + toString(getpid()) + ".temp"); + + //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); } } @@ -589,7 +539,15 @@ int ChimeraCcodeCommand::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) {