X-Git-Url: https://git.donarmstrong.com/?a=blobdiff_plain;f=aligncommand.cpp;h=efc8ce489e79ac5e1d17084f5e952b39590894ba;hb=6c2b1e530a5c0bb87040e58a3e410097acdfcc3d;hp=05ecd46ac6a3e192d834c6f26cdb5e1aa533cb71;hpb=55386dddad84cc1140d736cabaf4dd0ae16f2e01;p=mothur.git diff --git a/aligncommand.cpp b/aligncommand.cpp index 05ecd46..efc8ce4 100644 --- a/aligncommand.cpp +++ b/aligncommand.cpp @@ -75,6 +75,29 @@ string AlignCommand::getHelpString(){ exit(1); } } +//********************************************************************************************************************** +string AlignCommand::getOutputFileNameTag(string type, string inputName=""){ + try { + string tag = ""; + map >::iterator it; + + //is this a type this command creates + it = outputTypes.find(type); + if (it == outputTypes.end()) { m->mothurOut("[ERROR]: this command doesn't create a " + type + " output file.\n"); } + else { + if (type == "fasta") { tag = "align"; } + else if (type == "alignreport") { tag = "align.report"; } + else if (type == "accnos") { tag = "flip.accnos"; } + else { m->mothurOut("[ERROR]: No definition for type " + type + " output file tag.\n"); m->control_pressed = true; } + } + return tag; + } + catch(exception& e) { + m->errorOut(e, "AlignCommand", "getOutputFileName"); + exit(1); + } +} + //********************************************************************************************************************** AlignCommand::AlignCommand(){ try { @@ -224,34 +247,29 @@ AlignCommand::AlignCommand(string option) { // ...at some point should added some additional type checking... string temp; temp = validParameter.validFile(parameters, "ksize", false); if (temp == "not found"){ temp = "8"; } - convert(temp, kmerSize); + m->mothurConvert(temp, kmerSize); temp = validParameter.validFile(parameters, "match", false); if (temp == "not found"){ temp = "1.0"; } - convert(temp, match); + m->mothurConvert(temp, match); temp = validParameter.validFile(parameters, "mismatch", false); if (temp == "not found"){ temp = "-1.0"; } - convert(temp, misMatch); + m->mothurConvert(temp, misMatch); temp = validParameter.validFile(parameters, "gapopen", false); if (temp == "not found"){ temp = "-2.0"; } - convert(temp, gapOpen); + m->mothurConvert(temp, gapOpen); temp = validParameter.validFile(parameters, "gapextend", false); if (temp == "not found"){ temp = "-1.0"; } - convert(temp, gapExtend); + m->mothurConvert(temp, gapExtend); temp = validParameter.validFile(parameters, "processors", false); if (temp == "not found"){ temp = m->getProcessors(); } m->setProcessors(temp); - convert(temp, processors); + m->mothurConvert(temp, processors); temp = validParameter.validFile(parameters, "flip", false); if (temp == "not found"){ temp = "f"; } flip = m->isTrue(temp); temp = validParameter.validFile(parameters, "save", false); if (temp == "not found"){ temp = "f"; } save = m->isTrue(temp); - //this is so the threads can quickly load the reference data - #if defined (__APPLE__) || (__MACH__) || (linux) || (__linux) - #else - if (processors != 1) { save = true; } - #endif rdb->save = save; if (save) { //clear out old references rdb->clearMemory(); @@ -272,7 +290,7 @@ AlignCommand::AlignCommand(string option) { else { if (save) { rdb->setSavedReference(templateFileName); } } temp = validParameter.validFile(parameters, "threshold", false); if (temp == "not found"){ temp = "0.50"; } - convert(temp, threshold); + m->mothurConvert(temp, threshold); search = validParameter.validFile(parameters, "search", false); if (search == "not found"){ search = "kmer"; } if ((search != "suffix") && (search != "kmer") && (search != "blast")) { m->mothurOut("invalid search option: choices are kmer, suffix or blast."); m->mothurOutEndLine(); abort=true; } @@ -310,9 +328,9 @@ int AlignCommand::execute(){ m->mothurOut("Aligning sequences from " + candidateFileNames[s] + " ..." ); m->mothurOutEndLine(); if (outputDir == "") { outputDir += m->hasPath(candidateFileNames[s]); } - string alignFileName = outputDir + m->getRootName(m->getSimpleName(candidateFileNames[s])) + "align"; - string reportFileName = outputDir + m->getRootName(m->getSimpleName(candidateFileNames[s])) + "align.report"; - string accnosFileName = outputDir + m->getRootName(m->getSimpleName(candidateFileNames[s])) + "flip.accnos"; + string alignFileName = outputDir + m->getRootName(m->getSimpleName(candidateFileNames[s])) + getOutputFileNameTag("fasta"); + string reportFileName = outputDir + m->getRootName(m->getSimpleName(candidateFileNames[s])) + getOutputFileNameTag("alignreport"); + string accnosFileName = outputDir + m->getRootName(m->getSimpleName(candidateFileNames[s])) + getOutputFileNameTag("accnos"); bool hasAccnos = true; int numFastaSeqs = 0; @@ -322,7 +340,7 @@ int AlignCommand::execute(){ #ifdef USE_MPI int pid, numSeqsPerProcessor; int tag = 2001; - vector MPIPos; + vector MPIPos; MPIWroteAccnos = false; MPI_Status status; @@ -426,8 +444,8 @@ int AlignCommand::execute(){ #else - vector positions; - #if defined (__APPLE__) || (__MACH__) || (linux) || (__linux) + vector positions; + #if defined (__APPLE__) || (__MACH__) || (linux) || (__linux) || (__linux__) || (__unix__) || (__unix) positions = m->divideFile(candidateFileNames[s], processors); for (int i = 0; i < (positions.size()-1); i++) { lines.push_back(new linePair(positions[i], positions[(i+1)])); } #else @@ -435,7 +453,8 @@ int AlignCommand::execute(){ lines.push_back(new linePair(0, 1000)); }else { positions = m->setFilePosFasta(candidateFileNames[s], numFastaSeqs); - + if (positions.size() < processors) { processors = positions.size(); } + //figure out how many sequences you have to process int numSeqsPerProcessor = numFastaSeqs / processors; for (int i = 0; i < processors; i++) { @@ -553,7 +572,6 @@ int AlignCommand::driver(linePair* filePos, string alignFName, string reportFNam if (candidateSeq->getUnaligned().length() > alignment->getnRows()) { alignment->resize(candidateSeq->getUnaligned().length()+1); } - Sequence temp = templateDB->findClosestSequence(candidateSeq); Sequence* templateSeq = &temp; @@ -622,8 +640,8 @@ int AlignCommand::driver(linePair* filePos, string alignFName, string reportFNam } delete candidateSeq; - #if defined (__APPLE__) || (__MACH__) || (linux) || (__linux) - unsigned long int pos = inFASTA.tellg(); + #if defined (__APPLE__) || (__MACH__) || (linux) || (__linux) || (__linux__) || (__unix__) || (__unix) + unsigned long long pos = inFASTA.tellg(); if ((pos == -1) || (pos >= filePos->end)) { break; } #else if (inFASTA.eof()) { break; } @@ -650,7 +668,7 @@ int AlignCommand::driver(linePair* filePos, string alignFName, string reportFNam } //********************************************************************************************************************** #ifdef USE_MPI -int AlignCommand::driverMPI(int start, int num, MPI_File& inMPI, MPI_File& alignFile, MPI_File& reportFile, MPI_File& accnosFile, vector& MPIPos){ +int AlignCommand::driverMPI(int start, int num, MPI_File& inMPI, MPI_File& alignFile, MPI_File& reportFile, MPI_File& accnosFile, vector& MPIPos){ try { string outputString = ""; MPI_Status statusReport; @@ -831,7 +849,7 @@ int AlignCommand::createProcesses(string alignFileName, string reportFileName, s try { int num = 0; processIDS.resize(0); -#if defined (__APPLE__) || (__MACH__) || (linux) || (__linux) +#if defined (__APPLE__) || (__MACH__) || (linux) || (__linux) || (__linux__) || (__unix__) || (__unix) int process = 1; //loop through and create all the processes you want @@ -879,7 +897,7 @@ int AlignCommand::createProcesses(string alignFileName, string reportFileName, s if (!in.eof()) { int tempNum = 0; in >> tempNum; num += tempNum; } in.close(); m->mothurRemove(tempFile); - appendAlignFiles((alignFileName + toString(processIDS[i]) + ".temp"), alignFileName); + m->appendFiles((alignFileName + toString(processIDS[i]) + ".temp"), alignFileName); m->mothurRemove((alignFileName + toString(processIDS[i]) + ".temp")); appendReportFiles((reportFileName + toString(processIDS[i]) + ".temp"), reportFileName); @@ -896,7 +914,7 @@ int AlignCommand::createProcesses(string alignFileName, string reportFileName, s rename(nonBlankAccnosFiles[0].c_str(), accnosFName.c_str()); for (int h=1; h < nonBlankAccnosFiles.size(); h++) { - appendAlignFiles(nonBlankAccnosFiles[h], accnosFName); + m->appendFiles(nonBlankAccnosFiles[h], accnosFName); m->mothurRemove(nonBlankAccnosFiles[h]); } }else { //recreate the accnosfile if needed @@ -923,7 +941,7 @@ int AlignCommand::createProcesses(string alignFileName, string reportFileName, s string extension = ""; if (i != 0) { extension = toString(i) + ".temp"; } - alignData* tempalign = new alignData((alignFileName + extension), (reportFileName + extension), (accnosFName + extension), filename, align, search, kmerSize, m, lines[i]->start, lines[i]->end, flip, match, misMatch, gapOpen, gapExtend, threshold, i); + alignData* tempalign = new alignData(templateFileName, (alignFileName + extension), (reportFileName + extension), (accnosFName + extension), filename, align, search, kmerSize, m, lines[i]->start, lines[i]->end, flip, match, misMatch, gapOpen, gapExtend, threshold, i); pDataArray.push_back(tempalign); processIDS.push_back(i); @@ -961,7 +979,7 @@ int AlignCommand::createProcesses(string alignFileName, string reportFileName, s else { m->mothurRemove(accnosFName); } //remove so other files can be renamed to it for (int i = 1; i < processors; i++) { - appendAlignFiles((alignFileName + toString(i) + ".temp"), alignFileName); + m->appendFiles((alignFileName + toString(i) + ".temp"), alignFileName); m->mothurRemove((alignFileName + toString(i) + ".temp")); appendReportFiles((reportFileName + toString(i) + ".temp"), reportFileName); @@ -977,7 +995,7 @@ int AlignCommand::createProcesses(string alignFileName, string reportFileName, s rename(nonBlankAccnosFiles[0].c_str(), accnosFName.c_str()); for (int h=1; h < nonBlankAccnosFiles.size(); h++) { - appendAlignFiles(nonBlankAccnosFiles[h], accnosFName); + m->appendFiles(nonBlankAccnosFiles[h], accnosFName); m->mothurRemove(nonBlankAccnosFiles[h]); } }else { //recreate the accnosfile if needed @@ -994,29 +1012,6 @@ int AlignCommand::createProcesses(string alignFileName, string reportFileName, s exit(1); } } -/**************************************************************************************************/ - -void AlignCommand::appendAlignFiles(string temp, string filename) { - try{ - - ofstream output; - ifstream input; - m->openOutputFileAppend(filename, output); - m->openInputFile(temp, input); - - while(char c = input.get()){ - if(input.eof()) { break; } - else { output << c; } - } - - input.close(); - output.close(); - } - catch(exception& e) { - m->errorOut(e, "AlignCommand", "appendAlignFiles"); - exit(1); - } -} //********************************************************************************************************************** void AlignCommand::appendReportFiles(string temp, string filename) { @@ -1029,10 +1024,11 @@ void AlignCommand::appendReportFiles(string temp, string filename) { while (!input.eof()) { char c = input.get(); if (c == 10 || c == 13){ break; } } // get header line - while(char c = input.get()){ - if(input.eof()) { break; } - else { output << c; } - } + char buffer[4096]; + while (!input.eof()) { + input.read(buffer, 4096); + output.write(buffer, input.gcount()); + } input.close(); output.close();