X-Git-Url: https://git.donarmstrong.com/?a=blobdiff_plain;f=classifyseqscommand.cpp;h=c76b047cbd064a5b7b967ab388109bb4e46cc0cc;hb=49d2b7459c5027557564b21e9487dadafbbbdc96;hp=794b961dcc29ecdc79bf063d116887b4cf8c30ac;hpb=d6c0a11d1cecfac18b323285e7ffadb7f58e848f;p=mothur.git diff --git a/classifyseqscommand.cpp b/classifyseqscommand.cpp index 794b961..c76b047 100644 --- a/classifyseqscommand.cpp +++ b/classifyseqscommand.cpp @@ -85,6 +85,29 @@ string ClassifySeqsCommand::getHelpString(){ } } //********************************************************************************************************************** +string ClassifySeqsCommand::getOutputFileNameTag(string type, string inputName=""){ + try { + string outputFileName = ""; + 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 == "taxonomy") { outputFileName = "taxonomy"; } + else if (type == "accnos") { outputFileName = "flip.accnos"; } + else if (type == "taxsummary") { outputFileName = "tax.summary"; } + else if (type == "matchdist") { outputFileName = "match.dist"; } + else { m->mothurOut("[ERROR]: No definition for type " + type + " output file tag.\n"); m->control_pressed = true; } + } + return outputFileName; + } + catch(exception& e) { + m->errorOut(e, "ClassifySeqsCommand", "getOutputFileNameTag"); + exit(1); + } +} +//********************************************************************************************************************** ClassifySeqsCommand::ClassifySeqsCommand(){ try { abort = true; calledHelp = true; @@ -457,13 +480,15 @@ ClassifySeqsCommand::ClassifySeqsCommand(string option) { search = "kmer"; } - if (namefileNames.size() == 0){ - vector files; files.push_back(fastaFileNames[fastaFileNames.size()-1]); - parser.getNameFile(files); - } - - } - + if (!abort) { + if (namefileNames.size() == 0){ + if (fastaFileNames.size() != 0) { + vector files; files.push_back(fastaFileNames[fastaFileNames.size()-1]); + parser.getNameFile(files); + } + } + } + } } catch(exception& e) { m->errorOut(e, "ClassifySeqsCommand", "ClassifySeqsCommand"); @@ -482,7 +507,7 @@ ClassifySeqsCommand::~ClassifySeqsCommand(){ int ClassifySeqsCommand::execute(){ try { if (abort == true) { if (calledHelp) { return 0; } return 2; } - + if(method == "bayesian"){ classify = new Bayesian(taxonomyFileName, templateFileName, search, kmerSize, cutoff, iters, rand(), flip); } else if(method == "knn"){ classify = new Knn(taxonomyFileName, templateFileName, search, kmerSize, gapOpen, gapExtend, match, misMatch, numWanted, rand()); } else { @@ -500,19 +525,24 @@ int ClassifySeqsCommand::execute(){ string baseTName = taxonomyFileName; if (taxonomyFileName == "saved") {baseTName = rdb->getSavedTaxonomy(); } - string RippedTaxName = m->getRootName(m->getSimpleName(baseTName)); - RippedTaxName = m->getExtension(RippedTaxName.substr(0, RippedTaxName.length()-1)); - if (RippedTaxName[0] == '.') { RippedTaxName = RippedTaxName.substr(1, RippedTaxName.length()); } - RippedTaxName += "."; - + //set rippedTaxName to + string RippedTaxName = ""; + bool foundDot = false; + for (int i = baseTName.length()-1; i >= 0; i--) { + if (foundDot && (baseTName[i] != '.')) { RippedTaxName = baseTName[i] + RippedTaxName; } + else if (foundDot && (baseTName[i] == '.')) { break; } + else if (!foundDot && (baseTName[i] == '.')) { foundDot = true; } + } + if (RippedTaxName != "") { RippedTaxName += "."; } + if (outputDir == "") { outputDir += m->hasPath(fastaFileNames[s]); } - string newTaxonomyFile = outputDir + m->getRootName(m->getSimpleName(fastaFileNames[s])) + RippedTaxName + "taxonomy"; - string newaccnosFile = outputDir + m->getRootName(m->getSimpleName(fastaFileNames[s])) + RippedTaxName + "flip.accnos"; + string newTaxonomyFile = outputDir + m->getRootName(m->getSimpleName(fastaFileNames[s])) + RippedTaxName + getOutputFileNameTag("taxonomy"); + string newaccnosFile = outputDir + m->getRootName(m->getSimpleName(fastaFileNames[s])) + RippedTaxName + getOutputFileNameTag("accnos"); string tempTaxonomyFile = outputDir + m->getRootName(m->getSimpleName(fastaFileNames[s])) + "taxonomy.temp"; - string taxSummary = outputDir + m->getRootName(m->getSimpleName(fastaFileNames[s])) + RippedTaxName + "tax.summary"; + string taxSummary = outputDir + m->getRootName(m->getSimpleName(fastaFileNames[s])) + RippedTaxName + getOutputFileNameTag("taxsummary"); if ((method == "knn") && (search == "distance")) { - string DistName = outputDir + m->getRootName(m->getSimpleName(fastaFileNames[s])) + "match.dist"; + string DistName = outputDir + m->getRootName(m->getSimpleName(fastaFileNames[s])) + getOutputFileNameTag("matchdist"); classify->setDistName(DistName); outputNames.push_back(DistName); outputTypes["matchdist"].push_back(DistName); } @@ -615,7 +645,7 @@ int ClassifySeqsCommand::execute(){ #else vector positions; -#if defined (__APPLE__) || (__MACH__) || (linux) || (__linux) +#if defined (__APPLE__) || (__MACH__) || (linux) || (__linux) || (__linux__) || (__unix__) || (__unix) 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)])); } #else @@ -623,6 +653,7 @@ int ClassifySeqsCommand::execute(){ lines.push_back(new linePair(0, 1000)); }else { positions = m->setFilePosFasta(fastaFileNames[s], numFastaSeqs); + if (positions.size() < processors) { processors = positions.size(); } //figure out how many sequences you have to process int numSeqsPerProcessor = numFastaSeqs / processors; @@ -657,23 +688,8 @@ int ClassifySeqsCommand::execute(){ if(namefile != "") { m->mothurOut("Reading " + namefileNames[s] + "..."); cout.flush(); - nameMap.clear(); //remove old names - - ifstream inNames; - m->openInputFile(namefileNames[s], inNames); - - string firstCol, secondCol; - while(!inNames.eof()) { - inNames >> firstCol >> secondCol; m->gobble(inNames); - - vector temp; - m->splitAtComma(secondCol, temp); - - nameMap[firstCol] = temp; - } - inNames.close(); - + m->readNames(namefileNames[s], nameMap); m->mothurOut(" Done."); m->mothurOutEndLine(); } #endif @@ -821,7 +837,7 @@ int ClassifySeqsCommand::createProcesses(string taxFileName, string tempTaxFile, int num = 0; processIDS.clear(); -#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 @@ -905,16 +921,35 @@ int ClassifySeqsCommand::createProcesses(string taxFileName, string tempTaxFile, } #endif - + vector nonBlankAccnosFiles; + if (!(m->isBlank(accnos))) { nonBlankAccnosFiles.push_back(accnos); } + else { m->mothurRemove(accnos); } //remove so other files can be renamed to it + for(int i=0;iappendFiles((taxFileName + toString(processIDS[i]) + ".temp"), taxFileName); + m->appendFiles((tempTaxFile + toString(processIDS[i]) + ".temp"), tempTaxFile); + if (!(m->isBlank(accnos + toString(processIDS[i]) + ".temp"))) { + nonBlankAccnosFiles.push_back(accnos + toString(processIDS[i]) + ".temp"); + }else { m->mothurRemove((accnos + toString(processIDS[i]) + ".temp")); } + m->mothurRemove((m->getFullPathName(taxFileName) + toString(processIDS[i]) + ".temp")); m->mothurRemove((m->getFullPathName(tempTaxFile) + toString(processIDS[i]) + ".temp")); - m->mothurRemove((m->getFullPathName(accnos) + toString(processIDS[i]) + ".temp")); } + //append accnos files + if (nonBlankAccnosFiles.size() != 0) { + rename(nonBlankAccnosFiles[0].c_str(), accnos.c_str()); + + for (int h=1; h < nonBlankAccnosFiles.size(); h++) { + m->appendFiles(nonBlankAccnosFiles[h], accnos); + m->mothurRemove(nonBlankAccnosFiles[h]); + } + }else { //recreate the accnosfile if needed + ofstream out; + m->openOutputFile(accnos, out); + out.close(); + } + return num; } @@ -923,30 +958,6 @@ int ClassifySeqsCommand::createProcesses(string taxFileName, string tempTaxFile, exit(1); } } -/**************************************************************************************************/ - -void ClassifySeqsCommand::appendTaxFiles(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, "ClassifySeqsCommand", "appendTaxFiles"); - exit(1); - } -} - //********************************************************************************************************************** int ClassifySeqsCommand::driver(linePair* filePos, string taxFName, string tempTFName, string accnos, string filename){ @@ -1002,7 +1013,7 @@ int ClassifySeqsCommand::driver(linePair* filePos, string taxFName, string tempT } delete candidateSeq; - #if defined (__APPLE__) || (__MACH__) || (linux) || (__linux) + #if defined (__APPLE__) || (__MACH__) || (linux) || (__linux) || (__linux__) || (__unix__) || (__unix) unsigned long long pos = inFASTA.tellg(); if ((pos == -1) || (pos >= filePos->end)) { break; } #else