X-Git-Url: https://git.donarmstrong.com/?a=blobdiff_plain;f=chimeraccodecommand.cpp;h=286681a51d78de92590d7677a2e67a3c70fcaec1;hb=15cde0905641f8adddc1cc704f8c064f760e7461;hp=0b96f2172359efc9b7ed5ab5c9dcc15e7223dbbc;hpb=a61980e36d61937acccadc41be260165a1e389f4;p=mothur.git diff --git a/chimeraccodecommand.cpp b/chimeraccodecommand.cpp index 0b96f21..286681a 100644 --- a/chimeraccodecommand.cpp +++ b/chimeraccodecommand.cpp @@ -10,8 +10,55 @@ #include "chimeraccodecommand.h" #include "ccode.h" +//********************************************************************************************************************** +vector ChimeraCcodeCommand::getValidParameters(){ + try { + string AlignArray[] = {"fasta", "filter", "processors", "window", "template", "mask", "numwanted", "outputdir","inputdir" }; + vector myArray (AlignArray, AlignArray+(sizeof(AlignArray)/sizeof(string))); + return myArray; + } + catch(exception& e) { + m->errorOut(e, "ChimeraCcodeCommand", "getValidParameters"); + exit(1); + } +} +//********************************************************************************************************************** +ChimeraCcodeCommand::ChimeraCcodeCommand(){ + try { + vector tempOutNames; + outputTypes["chimera"] = tempOutNames; + outputTypes["mapinfo"] = tempOutNames; + outputTypes["accnos"] = tempOutNames; + } + catch(exception& e) { + m->errorOut(e, "ChimeraCcodeCommand", "ChimeraCcodeCommand"); + exit(1); + } +} +//********************************************************************************************************************** +vector ChimeraCcodeCommand::getRequiredParameters(){ + try { + string AlignArray[] = {"template","fasta"}; + vector myArray (AlignArray, AlignArray+(sizeof(AlignArray)/sizeof(string))); + return myArray; + } + catch(exception& e) { + m->errorOut(e, "ChimeraCcodeCommand", "getRequiredParameters"); + exit(1); + } +} +//********************************************************************************************************************** +vector ChimeraCcodeCommand::getRequiredFiles(){ + try { + vector myArray; + return myArray; + } + catch(exception& e) { + m->errorOut(e, "ChimeraCcodeCommand", "getRequiredFiles"); + exit(1); + } +} //*************************************************************************************************************** - ChimeraCcodeCommand::ChimeraCcodeCommand(string option) { try { abort = false; @@ -21,13 +68,13 @@ ChimeraCcodeCommand::ChimeraCcodeCommand(string option) { else { //valid paramters for this command - string Array[] = {"fasta", "filter", "processors", "window", "template", "mask", "numwanted", "outputdir","inputdir", }; + string Array[] = {"fasta", "filter", "processors", "window", "template", "mask", "numwanted", "outputdir","inputdir" }; vector myArray (Array, Array+(sizeof(Array)/sizeof(string))); OptionParser parser(option); map parameters = parser.getParameters(); - ValidParameters validParameter; + ValidParameters validParameter("chimera.ccode"); map::iterator it; //check to make sure all parameters are valid for command @@ -35,39 +82,84 @@ ChimeraCcodeCommand::ChimeraCcodeCommand(string option) { if (validParameter.isValidParameter(it->first, myArray, it->second) != true) { abort = true; } } + vector tempOutNames; + outputTypes["chimera"] = tempOutNames; + outputTypes["mapinfo"] = tempOutNames; + outputTypes["accnos"] = 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 = ""; } else { string path; - it = parameters.find("fasta"); - //user has given a template file - if(it != parameters.end()){ - path = 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; } - } - 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; } } } - //check for required parameters - fastafile = validParameter.validFile(parameters, "fasta", true); - if (fastafile == "not open") { abort = true; } - else if (fastafile == "not found") { fastafile = ""; m->mothurOut("fasta is a required parameter for the chimera.ccode command."); m->mothurOutEndLine(); abort = true; } + 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 { + 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 = 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]; } + } + + int ableToOpen; + ifstream in; + + 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() + m->getSimpleName(fastaFileNames[i]); + m->mothurOut("Unable to open " + fastaFileNames[i] + ". Trying default " + tryPath); m->mothurOutEndLine(); + 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) { + m->mothurOut("Unable to open " + fastaFileNames[i] + ". It will be disregarded."); m->mothurOutEndLine(); + //erase from file list + fastaFileNames.erase(fastaFileNames.begin()+i); + i--; + } + } + + //make sure there is at least one valid file left + if (fastaFileNames.size() == 0) { m->mothurOut("no valid files."); m->mothurOutEndLine(); abort = true; } + } //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; } @@ -77,20 +169,20 @@ 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); @@ -117,6 +209,7 @@ void ChimeraCcodeCommand::help(){ m->mothurOut("This command was created using the algorythms described in the 'Evaluating putative chimeric sequences from PCR-amplified products' paper by Juan M. Gonzalez, Johannes Zimmerman and Cesareo Saiz-Jimenez.\n"); m->mothurOut("The chimera.ccode command parameters are fasta, template, filter, mask, processors, window and numwanted.\n"); m->mothurOut("The fasta parameter allows you to enter the fasta file containing your potentially chimeric sequences, and is required. \n"); + m->mothurOut("You may enter multiple fasta files by separating their names with dashes. ie. fasta=abrecovery.fasta-amzon.fasta \n"); m->mothurOut("The template parameter allows you to enter a template file containing known non-chimeric sequences, and is required. \n"); m->mothurOut("The filter parameter allows you to specify if you would like to apply a vertical and 50% soft filter. \n"); m->mothurOut("The processors parameter allows you to specify how many processors you would like to use. The default is 1. \n"); @@ -128,7 +221,7 @@ void ChimeraCcodeCommand::help(){ m->mothurOut("The numwanted parameter allows you to specify how many sequences you would each query sequence compared with.\n"); m->mothurOut("The chimera.ccode command should be in the following format: \n"); m->mothurOut("chimera.ccode(fasta=yourFastaFile, template=yourTemplate) \n"); - m->mothurOut("Example: chimera.seqs(fasta=AD.align, template=core_set_aligned.imputed.fasta) \n"); + m->mothurOut("Example: chimera.ccode(fasta=AD.align, template=core_set_aligned.imputed.fasta) \n"); m->mothurOut("Note: No spaces between parameter labels (i.e. fasta), '=' and parameters (i.e.yourFastaFile).\n\n"); } catch(exception& e) { @@ -148,287 +241,203 @@ int ChimeraCcodeCommand::execute(){ if (abort == true) { return 0; } - int start = time(NULL); - - //set user options - if (maskfile == "default") { m->mothurOut("I am using the default 236627 EU009184.1 Shigella dysenteriae str. FBD013."); m->mothurOutEndLine(); } - - chimera = new Ccode(fastafile, templatefile, filter, maskfile, window, numwanted, outputDir); - - //is your template aligned? - if (chimera->getUnaligned()) { m->mothurOut("Your template sequences are different lengths, please correct."); m->mothurOutEndLine(); delete chimera; return 0; } - templateSeqsLength = chimera->getLength(); - - string outputFileName, accnosFileName; - if (maskfile != "") { - outputFileName = outputDir + getRootName(getSimpleName(fastafile)) + maskfile + ".ccode.chimeras"; - accnosFileName = outputDir + getRootName(getSimpleName(fastafile)) + maskfile + ".ccode.accnos"; - }else { - outputFileName = outputDir + getRootName(getSimpleName(fastafile)) + "ccode.chimeras"; - accnosFileName = outputDir + getRootName(getSimpleName(fastafile)) + "ccode.accnos"; - } - - string mapInfo = outputDir + getRootName(getSimpleName(fastafile)) + "mapinfo"; - bool hasAccnos = true; - - if (m->control_pressed) { delete chimera; return 0; } + for (int s = 0; s < fastaFileNames.size(); s++) { + + m->mothurOut("Checking sequences from " + fastaFileNames[s] + " ..." ); m->mothurOutEndLine(); - #ifdef USE_MPI - - int pid, end, numSeqsPerProcessor; - int tag = 2001; - vector MPIPos; - MPIWroteAccnos = false; + int start = time(NULL); - MPI_Status status; - MPI_Comm_rank(MPI_COMM_WORLD, &pid); //find out who we are - MPI_Comm_size(MPI_COMM_WORLD, &processors); + //set user options + if (maskfile == "default") { m->mothurOut("I am using the default 236627 EU009184.1 Shigella dysenteriae str. FBD013."); m->mothurOutEndLine(); } - MPI_File inMPI; - MPI_File outMPI; - MPI_File outMPIAccnos; + chimera = new Ccode(fastaFileNames[s], templatefile, filter, maskfile, window, numwanted, outputDir); - int outMode=MPI_MODE_CREATE|MPI_MODE_WRONLY; - int inMode=MPI_MODE_RDONLY; - - //char* outFilename = new char[outputFileName.length()]; - //memcpy(outFilename, outputFileName.c_str(), outputFileName.length()); + //is your template aligned? + if (chimera->getUnaligned()) { m->mothurOut("Your template sequences are different lengths, please correct."); m->mothurOutEndLine(); delete chimera; return 0; } + templateSeqsLength = chimera->getLength(); - char outFilename[1024]; - strcpy(outFilename, outputFileName.c_str()); - - //char* outAccnosFilename = new char[accnosFileName.length()]; - //memcpy(outAccnosFilename, accnosFileName.c_str(), accnosFileName.length()); - - char outAccnosFilename[1024]; - strcpy(outAccnosFilename, accnosFileName.c_str()); + 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 + m->getRootName(m->getSimpleName(fastaFileNames[s])) + maskfile + ".ccode.chimeras"; + accnosFileName = outputDir + m->getRootName(m->getSimpleName(fastaFileNames[s])) + maskfile + ".ccode.accnos"; + }else { + outputFileName = outputDir + m->getRootName(m->getSimpleName(fastaFileNames[s])) + "ccode.chimeras"; + accnosFileName = outputDir + m->getRootName(m->getSimpleName(fastaFileNames[s])) + "ccode.accnos"; + } - //char* inFileName = new char[fastafile.length()]; - //memcpy(inFileName, fastafile.c_str(), fastafile.length()); + string mapInfo = outputDir + m->getRootName(m->getSimpleName(fastaFileNames[s])) + "mapinfo"; - char inFileName[1024]; - strcpy(inFileName, fastafile.c_str()); - - MPI_File_open(MPI_COMM_WORLD, inFileName, inMode, MPI_INFO_NULL, &inMPI); //comm, filename, mode, info, filepointer - MPI_File_open(MPI_COMM_WORLD, outFilename, outMode, MPI_INFO_NULL, &outMPI); - MPI_File_open(MPI_COMM_WORLD, outAccnosFilename, outMode, MPI_INFO_NULL, &outMPIAccnos); + if (m->control_pressed) { delete chimera; for (int j = 0; j < outputNames.size(); j++) { remove(outputNames[j].c_str()); } outputTypes.clear(); return 0; } - //delete inFileName; - //delete outFilename; - //delete outAccnosFilename; - - if (m->control_pressed) { MPI_File_close(&inMPI); MPI_File_close(&outMPI); MPI_File_close(&outMPIAccnos); delete chimera; return 0; } + #ifdef USE_MPI - if (pid == 0) { //you are the root process - string outTemp = "For full window mapping info refer to " + mapInfo + "\n\n"; - - //print header - int length = outTemp.length(); - char* buf2 = new char[length]; - memcpy(buf2, outTemp.c_str(), length); - - MPI_File_write_shared(outMPI, buf2, length, MPI_CHAR, &status); - delete buf2; - - MPIPos = setFilePosFasta(fastafile, numSeqs); //fills MPIPos, returns numSeqs - - //send file positions to all processes - MPI_Bcast(&numSeqs, 1, MPI_INT, 0, MPI_COMM_WORLD); //send numSeqs - MPI_Bcast(&MPIPos[0], (numSeqs+1), MPI_LONG, 0, MPI_COMM_WORLD); //send file pos - - //figure out how many sequences you have to align - numSeqsPerProcessor = numSeqs / processors; - if(pid == (processors - 1)){ numSeqsPerProcessor = numSeqs - pid * numSeqsPerProcessor; } - int startIndex = pid * numSeqsPerProcessor; - - //align your part - driverMPI(startIndex, numSeqsPerProcessor, inMPI, outMPI, outMPIAccnos, MPIPos); - - if (m->control_pressed) { MPI_File_close(&inMPI); MPI_File_close(&outMPI); MPI_File_close(&outMPIAccnos); remove(outputFileName.c_str()); remove(accnosFileName.c_str()); delete chimera; return 0; } + int pid, end, numSeqsPerProcessor; + int tag = 2001; + vector MPIPos; + + MPI_Status status; + MPI_Comm_rank(MPI_COMM_WORLD, &pid); //find out who we are + MPI_Comm_size(MPI_COMM_WORLD, &processors); - for (int i = 1; i < processors; i++) { - bool tempResult; - MPI_Recv(&tempResult, 1, MPI_INT, i, tag, MPI_COMM_WORLD, &status); - if (tempResult != 0) { MPIWroteAccnos = true; } - } - }else{ //you are a child process - MPI_Bcast(&numSeqs, 1, MPI_INT, 0, MPI_COMM_WORLD); //get numSeqs - MPIPos.resize(numSeqs+1); - MPI_Bcast(&MPIPos[0], (numSeqs+1), MPI_LONG, 0, MPI_COMM_WORLD); //get file positions + MPI_File inMPI; + MPI_File outMPI; + MPI_File outMPIAccnos; - //figure out how many sequences you have to align - numSeqsPerProcessor = numSeqs / processors; - if(pid == (processors - 1)){ numSeqsPerProcessor = numSeqs - pid * numSeqsPerProcessor; } - int startIndex = pid * numSeqsPerProcessor; + int outMode=MPI_MODE_CREATE|MPI_MODE_WRONLY; + int inMode=MPI_MODE_RDONLY; - //align your part - driverMPI(startIndex, numSeqsPerProcessor, inMPI, outMPI, outMPIAccnos, MPIPos); + char outFilename[1024]; + strcpy(outFilename, outputFileName.c_str()); - if (m->control_pressed) { MPI_File_close(&inMPI); MPI_File_close(&outMPI); MPI_File_close(&outMPIAccnos); delete chimera; return 0; } + char outAccnosFilename[1024]; + strcpy(outAccnosFilename, accnosFileName.c_str()); - MPI_Send(&MPIWroteAccnos, 1, MPI_INT, 0, tag, MPI_COMM_WORLD); - } - - //close files - MPI_File_close(&inMPI); - MPI_File_close(&outMPI); - MPI_File_close(&outMPIAccnos); + char inFileName[1024]; + strcpy(inFileName, fastaFileNames[s].c_str()); + + MPI_File_open(MPI_COMM_WORLD, inFileName, inMode, MPI_INFO_NULL, &inMPI); //comm, filename, mode, info, filepointer + MPI_File_open(MPI_COMM_WORLD, outFilename, outMode, MPI_INFO_NULL, &outMPI); + MPI_File_open(MPI_COMM_WORLD, outAccnosFilename, outMode, MPI_INFO_NULL, &outMPIAccnos); + + if (m->control_pressed) { MPI_File_close(&inMPI); MPI_File_close(&outMPI); MPI_File_close(&outMPIAccnos); for (int j = 0; j < outputNames.size(); j++) { remove(outputNames[j].c_str()); } outputTypes.clear(); delete chimera; return 0; } - //delete accnos file if blank - if (pid == 0) { - if (!MPIWroteAccnos) { - //MPI_Info info; - //MPI_File_delete(outAccnosFilename, info); - hasAccnos = false; - remove(accnosFileName.c_str()); - } - } - - #else - ofstream outHeader; - string tempHeader = outputDir + getRootName(getSimpleName(fastafile)) + maskfile + "ccode.chimeras.tempHeader"; - openOutputFile(tempHeader, outHeader); - - outHeader << "For full window mapping info refer to " << mapInfo << endl << endl; + if (pid == 0) { //you are the root process + string outTemp = "For full window mapping info refer to " + mapInfo + "\n\n"; + + //print header + int length = outTemp.length(); + char* buf2 = new char[length]; + memcpy(buf2, outTemp.c_str(), length); - outHeader.close(); - - //break up file - #if defined (__APPLE__) || (__MACH__) || (linux) || (__linux) - if(processors == 1){ - ifstream inFASTA; - openInputFile(fastafile, inFASTA); - numSeqs=count(istreambuf_iterator(inFASTA),istreambuf_iterator(), '>'); - inFASTA.close(); - - lines.push_back(new linePair(0, numSeqs)); - - driver(lines[0], outputFileName, fastafile, accnosFileName); - - if (m->control_pressed) { - remove(outputFileName.c_str()); - remove(tempHeader.c_str()); - remove(accnosFileName.c_str()); - for (int i = 0; i < lines.size(); i++) { delete lines[i]; } lines.clear(); - delete chimera; - return 0; - } - - //delete accnos file if its blank - if (isBlank(accnosFileName)) { remove(accnosFileName.c_str()); hasAccnos = false; } - - }else{ - vector positions; - processIDS.resize(0); - - ifstream inFASTA; - openInputFile(fastafile, 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 i = 0; i < processors; i++) { - long int startPos = positions[ i * numSeqsPerProcessor ]; - if(i == processors - 1){ - numSeqsPerProcessor = numSeqs - i * numSeqsPerProcessor; + MPI_File_write_shared(outMPI, buf2, length, MPI_CHAR, &status); + delete buf2; + + MPIPos = m->setFilePosFasta(fastaFileNames[s], numSeqs); //fills MPIPos, returns numSeqs + + //send file positions to all processes + for(int i = 1; i < processors; i++) { + MPI_Send(&numSeqs, 1, MPI_INT, i, tag, MPI_COMM_WORLD); + MPI_Send(&MPIPos[0], (numSeqs+1), MPI_LONG, i, tag, MPI_COMM_WORLD); } - lines.push_back(new linePair(startPos, numSeqsPerProcessor)); - } - + + //figure out how many sequences you have to align + numSeqsPerProcessor = numSeqs / processors; + int startIndex = pid * numSeqsPerProcessor; + if(pid == (processors - 1)){ numSeqsPerProcessor = numSeqs - pid * numSeqsPerProcessor; } + - createProcesses(outputFileName, fastafile, accnosFileName); - - rename((outputFileName + toString(processIDS[0]) + ".temp").c_str(), outputFileName.c_str()); + //align your part + driverMPI(startIndex, numSeqsPerProcessor, inMPI, outMPI, outMPIAccnos, MPIPos); + + if (m->control_pressed) { MPI_File_close(&inMPI); MPI_File_close(&outMPI); MPI_File_close(&outMPIAccnos); remove(outputFileName.c_str()); remove(accnosFileName.c_str()); for (int j = 0; j < outputNames.size(); j++) { remove(outputNames[j].c_str()); } outputTypes.clear(); delete chimera; return 0; } + + }else{ //you are a child process + MPI_Recv(&numSeqs, 1, MPI_INT, 0, tag, MPI_COMM_WORLD, &status); + MPIPos.resize(numSeqs+1); + MPI_Recv(&MPIPos[0], (numSeqs+1), MPI_LONG, 0, tag, MPI_COMM_WORLD, &status); + + //figure out how many sequences you have to align + numSeqsPerProcessor = numSeqs / processors; + int startIndex = pid * numSeqsPerProcessor; + if(pid == (processors - 1)){ numSeqsPerProcessor = numSeqs - pid * numSeqsPerProcessor; } + - //append output files - for(int i=1;icontrol_pressed) { MPI_File_close(&inMPI); MPI_File_close(&outMPI); MPI_File_close(&outMPIAccnos); for (int j = 0; j < outputNames.size(); j++) { remove(outputNames[j].c_str()); } outputTypes.clear(); delete chimera; return 0; } } - vector nonBlankAccnosFiles; - //delete blank accnos files generated with multiple processes - for(int i=0;icontrol_pressed) { - remove(outputFileName.c_str()); - remove(accnosFileName.c_str()); - for (int i = 0; i < lines.size(); i++) { delete lines[i]; } lines.clear(); - delete chimera; - return 0; - } - - } - #else - ifstream inFASTA; - openInputFile(fastafile, inFASTA); - numSeqs=count(istreambuf_iterator(inFASTA),istreambuf_iterator(), '>'); - inFASTA.close(); - lines.push_back(new linePair(0, numSeqs)); + ofstream outHeader; + string tempHeader = outputDir + m->getRootName(m->getSimpleName(fastaFileNames[s])) + maskfile + "ccode.chimeras.tempHeader"; + m->openOutputFile(tempHeader, outHeader); - driver(lines[0], outputFileName, fastafile, accnosFileName); + outHeader << "For full window mapping info refer to " << mapInfo << endl << endl; + + outHeader.close(); - if (m->control_pressed) { - remove(outputFileName.c_str()); - remove(tempHeader.c_str()); - remove(accnosFileName.c_str()); - for (int i = 0; i < lines.size(); i++) { delete lines[i]; } lines.clear(); - delete chimera; - return 0; - } + 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)])); + } - //delete accnos file if its blank - if (isBlank(accnosFileName)) { remove(accnosFileName.c_str()); hasAccnos = false; } - #endif - - //m->mothurOut("Output File Names: "); - //if ((filter) && (method == "bellerophon")) { m->mothurOut( - //if (outputDir == "") { fastafile = getRootName(fastafile) + "filter.fasta"; } - // else { fastafile = outputDir + getRootName(getSimpleName(fastafile)) + "filter.fasta"; } - - appendFiles(outputFileName, tempHeader); - - remove(outputFileName.c_str()); - rename(tempHeader.c_str(), outputFileName.c_str()); - #endif + //break up file + #if defined (__APPLE__) || (__MACH__) || (linux) || (__linux) + if(processors == 1){ + + numSeqs = 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]; } outputTypes.clear(); lines.clear(); delete chimera; return 0; } + + }else{ + processIDS.resize(0); + + 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()); + } + + if (m->control_pressed) { + remove(outputFileName.c_str()); + remove(accnosFileName.c_str()); + for (int j = 0; j < outputNames.size(); j++) { remove(outputNames[j].c_str()); } outputTypes.clear(); + for (int i = 0; i < lines.size(); i++) { delete lines[i]; } lines.clear(); + delete chimera; + return 0; + } + + } + + #else + numSeqs = 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]; } outputTypes.clear(); lines.clear(); delete chimera; return 0; } + + #endif - delete chimera; + m->appendFiles(outputFileName, tempHeader); + remove(outputFileName.c_str()); + rename(tempHeader.c_str(), outputFileName.c_str()); + #endif + + delete chimera; + + outputNames.push_back(outputFileName); outputTypes["chimera"].push_back(outputFileName); + outputNames.push_back(mapInfo); outputTypes["mapinfo"].push_back(mapInfo); + outputNames.push_back(accnosFileName); outputTypes["accnos"].push_back(accnosFileName); + + for (int i = 0; i < lines.size(); i++) { delete lines[i]; } lines.clear(); + + m->mothurOutEndLine(); m->mothurOut("It took " + toString(time(NULL) - start) + " secs to check " + toString(numSeqs) + " sequences."); m->mothurOutEndLine(); + } m->mothurOutEndLine(); m->mothurOut("Output File Names: "); m->mothurOutEndLine(); - m->mothurOut(outputFileName); m->mothurOutEndLine(); - if (hasAccnos) { m->mothurOut(accnosFileName); m->mothurOutEndLine(); } + for (int i = 0; i < outputNames.size(); i++) { m->mothurOut(outputNames[i]); m->mothurOutEndLine(); } m->mothurOutEndLine(); - - for (int i = 0; i < lines.size(); i++) { delete lines[i]; } lines.clear(); - - m->mothurOutEndLine(); m->mothurOut("It took " + toString(time(NULL) - start) + " secs to check " + toString(numSeqs) + " sequences."); m->mothurOutEndLine(); return 0; @@ -440,24 +449,27 @@ 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 @@ -472,20 +484,28 @@ 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"); @@ -494,7 +514,7 @@ int ChimeraCcodeCommand::driver(linePair* line, string outputFName, string filen } //********************************************************************************************************************** #ifdef USE_MPI -int ChimeraCcodeCommand::driverMPI(int start, int num, MPI_File& inMPI, MPI_File& outMPI, MPI_File& outAccMPI, vector& MPIPos){ +int ChimeraCcodeCommand::driverMPI(int start, int num, MPI_File& inMPI, MPI_File& outMPI, MPI_File& outAccMPI, vector& MPIPos){ try { MPI_Status status; @@ -517,7 +537,7 @@ 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 @@ -531,7 +551,6 @@ int ChimeraCcodeCommand::driverMPI(int start, int num, MPI_File& inMPI, MPI_File //print results bool isChimeric = chimera->print(outMPI, outAccMPI); - if (isChimeric) { MPIWroteAccnos = true; } } } delete candidateSeq; @@ -558,7 +577,7 @@ 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) { @@ -568,9 +587,21 @@ 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("[ERROR]: unable to spawn the necessary processes."); m->mothurOutEndLine(); + for (int i = 0; i < processIDS.size(); i++) { kill (processIDS[i], SIGINT); } exit(0); - }else { m->mothurOut("unable to spawn the necessary processes."); m->mothurOutEndLine(); exit(0); } + } } //force parent to wait until all the processes are done @@ -579,7 +610,15 @@ 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) {