X-Git-Url: https://git.donarmstrong.com/?p=mothur.git;a=blobdiff_plain;f=chimeraccodecommand.cpp;h=d890db4087813cc3c4c4e453b55013fa555e7689;hp=2b46f2bcda9246b865759f832e070fbc27d8fe0d;hb=cf9987b67aa49777a4c91c2d21f96e58bf17aa82;hpb=52ea8f3a52c05331b9bd7519ae0b518bda233d05 diff --git a/chimeraccodecommand.cpp b/chimeraccodecommand.cpp index 2b46f2b..d890db4 100644 --- a/chimeraccodecommand.cpp +++ b/chimeraccodecommand.cpp @@ -9,25 +9,110 @@ #include "chimeraccodecommand.h" #include "ccode.h" - +#include "referencedb.h" +//********************************************************************************************************************** +vector ChimeraCcodeCommand::setParameters(){ + try { + CommandParameter ptemplate("reference", "InputTypes", "", "", "none", "none", "none","",false,true,true); parameters.push_back(ptemplate); + CommandParameter pfasta("fasta", "InputTypes", "", "", "none", "none", "none","chimera-mapinfo-accnos",false,true,true); parameters.push_back(pfasta); + CommandParameter pfilter("filter", "Boolean", "", "F", "", "", "","",false,false); parameters.push_back(pfilter); + CommandParameter pwindow("window", "Number", "", "0", "", "", "","",false,false); parameters.push_back(pwindow); + CommandParameter pnumwanted("numwanted", "Number", "", "20", "", "", "","",false,false); parameters.push_back(pnumwanted); + CommandParameter pmask("mask", "String", "", "", "", "", "","",false,false); parameters.push_back(pmask); + CommandParameter pprocessors("processors", "Number", "", "1", "", "", "","",false,false,true); parameters.push_back(pprocessors); + CommandParameter pinputdir("inputdir", "String", "", "", "", "", "","",false,false); parameters.push_back(pinputdir); + CommandParameter poutputdir("outputdir", "String", "", "", "", "", "","",false,false); parameters.push_back(poutputdir); + CommandParameter psave("save", "Boolean", "", "F", "", "", "","",false,false); parameters.push_back(psave); + + vector myArray; + for (int i = 0; i < parameters.size(); i++) { myArray.push_back(parameters[i].name); } + return myArray; + } + catch(exception& e) { + m->errorOut(e, "ChimeraCcodeCommand", "setParameters"); + exit(1); + } +} +//********************************************************************************************************************** +string ChimeraCcodeCommand::getHelpString(){ + try { + string helpString = ""; + helpString += "The chimera.ccode command reads a fastafile and referencefile and outputs potentially chimeric sequences.\n"; + helpString += "This command was created using the algorithms described in the 'Evaluating putative chimeric sequences from PCR-amplified products' paper by Juan M. Gonzalez, Johannes Zimmerman and Cesareo Saiz-Jimenez.\n"; + helpString += "The chimera.ccode command parameters are fasta, reference, filter, mask, processors, window and numwanted.\n"; + helpString += "The fasta parameter allows you to enter the fasta file containing your potentially chimeric sequences, and is required unless you have a valid current fasta file. \n"; + helpString += "You may enter multiple fasta files by separating their names with dashes. ie. fasta=abrecovery.fasta-amzon.fasta \n"; + helpString += "The reference parameter allows you to enter a reference file containing known non-chimeric sequences, and is required. \n"; + helpString += "The filter parameter allows you to specify if you would like to apply a vertical and 50% soft filter. \n"; + helpString += "The processors parameter allows you to specify how many processors you would like to use. The default is 1. \n"; +#ifdef USE_MPI + helpString += "When using MPI, the processors parameter is set to the number of MPI processes running. \n"; +#endif + helpString += "The mask parameter allows you to specify a file containing one sequence you wish to use as a mask for the your sequences. \n"; + helpString += "The window parameter allows you to specify the window size for searching for chimeras. \n"; + helpString += "The numwanted parameter allows you to specify how many sequences you would each query sequence compared with.\n"; + helpString += "If the save parameter is set to true the reference sequences will be saved in memory, to clear them later you can use the clear.memory command. Default=f."; + helpString += "The chimera.ccode command should be in the following format: \n"; + helpString += "chimera.ccode(fasta=yourFastaFile, reference=yourTemplate) \n"; + helpString += "Example: chimera.ccode(fasta=AD.align, reference=core_set_aligned.imputed.fasta) \n"; + helpString += "Note: No spaces between parameter labels (i.e. fasta), '=' and parameters (i.e.yourFastaFile).\n"; + return helpString; + } + catch(exception& e) { + m->errorOut(e, "ChimeraCcodeCommand", "getHelpString"); + exit(1); + } +} +//********************************************************************************************************************** +string ChimeraCcodeCommand::getOutputPattern(string type) { + try { + string pattern = ""; + + if (type == "chimera") { pattern = "[filename],[tag],ccode.chimeras-[filename],ccode.chimeras"; } + else if (type == "accnos") { pattern = "[filename],[tag],ccode.accnos-[filename],ccode.accnos"; } + else if (type == "mapinfo") { pattern = "[filename],mapinfo"; } + else { m->mothurOut("[ERROR]: No definition for type " + type + " output pattern.\n"); m->control_pressed = true; } + + return pattern; + } + catch(exception& e) { + m->errorOut(e, "ChimeraCcodeCommand", "getOutputPattern"); + exit(1); + } +} +//********************************************************************************************************************** +ChimeraCcodeCommand::ChimeraCcodeCommand(){ + try { + abort = true; calledHelp = true; + setParameters(); + vector tempOutNames; + outputTypes["chimera"] = tempOutNames; + outputTypes["mapinfo"] = tempOutNames; + outputTypes["accnos"] = tempOutNames; + + } + catch(exception& e) { + m->errorOut(e, "ChimeraCcodeCommand", "ChimeraCcodeCommand"); + exit(1); + } +} //*************************************************************************************************************** - ChimeraCcodeCommand::ChimeraCcodeCommand(string option) { try { - abort = false; + abort = false; calledHelp = false; + ReferenceDB* rdb = ReferenceDB::getInstance(); //allow user to run help - if(option == "help") { help(); abort = true; } + if(option == "help") { help(); abort = true; calledHelp = true; } + else if(option == "citation") { citation(); abort = true; calledHelp = true;} else { - //valid paramters for this command - string Array[] = {"fasta", "filter", "processors", "window", "template", "mask", "numwanted", "outputdir","inputdir", }; - vector myArray (Array, Array+(sizeof(Array)/sizeof(string))); + vector myArray = setParameters(); 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 @@ -35,64 +120,97 @@ 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("template"); + it = parameters.find("reference"); //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; } + if (path == "") { parameters["reference"] = inputDir + it->second; } } } //check for required parameters 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); + if (fastafile == "not found") { //if there is a current fasta file, use it + string filename = m->getFastaFile(); + if (filename != "") { fastaFileNames.push_back(filename); m->mothurOut("Using " + filename + " as input file for the fasta parameter."); m->mothurOutEndLine(); } + else { m->mothurOut("You have no current fastafile and the fasta parameter is required."); 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 = 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]; } + + bool ignore = false; + if (fastaFileNames[i] == "current") { + fastaFileNames[i] = m->getFastaFile(); + if (fastaFileNames[i] != "") { m->mothurOut("Using " + fastaFileNames[i] + " as input file for the fasta parameter where you had given current."); m->mothurOutEndLine(); } + else { + m->mothurOut("You have no current fastafile, ignoring current."); m->mothurOutEndLine(); ignore=true; + //erase from file list + fastaFileNames.erase(fastaFileNames.begin()+i); + i--; + } } - - 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); - in.close(); + if (!ignore) { + + 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"); - #ifdef USE_MPI - for (int j = 1; j < processors; j++) { - MPI_Send(&ableToOpen, 1, MPI_INT, j, 2001, MPI_COMM_WORLD); + //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; } - }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(); - //erase from file list - fastaFileNames.erase(fastaFileNames.begin()+i); - i--; + //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--; + }else { + m->setFastaFile(fastaFileNames[i]); + } } } @@ -101,42 +219,58 @@ 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 - } - - templatefile = validParameter.validFile(parameters, "template", true); - if (templatefile == "not open") { abort = true; } - else if (templatefile == "not found") { templatefile = ""; m->mothurOut("template is a required parameter for the chimera.ccode command."); m->mothurOutEndLine(); abort = true; } + outputDir = validParameter.validFile(parameters, "outputdir", false); if (outputDir == "not found"){ outputDir = ""; } maskfile = validParameter.validFile(parameters, "mask", false); 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); + temp = validParameter.validFile(parameters, "processors", false); if (temp == "not found"){ temp = m->getProcessors(); } + m->setProcessors(temp); + m->mothurConvert(temp, processors); temp = validParameter.validFile(parameters, "window", false); if (temp == "not found") { temp = "0"; } - convert(temp, window); + m->mothurConvert(temp, window); temp = validParameter.validFile(parameters, "numwanted", false); if (temp == "not found") { temp = "20"; } - convert(temp, numwanted); + m->mothurConvert(temp, numwanted); + + temp = validParameter.validFile(parameters, "save", false); if (temp == "not found"){ temp = "f"; } + save = m->isTrue(temp); + rdb->save = save; + if (save) { //clear out old references + rdb->clearMemory(); + } + + //this has to go after save so that if the user sets save=t and provides no reference we abort + templatefile = validParameter.validFile(parameters, "reference", true); + if (templatefile == "not found") { + //check for saved reference sequences + if (rdb->referenceSeqs.size() != 0) { + templatefile = "saved"; + }else { + m->mothurOut("[ERROR]: You don't have any saved reference sequences and the reference parameter is a required."); + m->mothurOutEndLine(); + abort = true; + } + }else if (templatefile == "not open") { abort = true; } + else { if (save) { rdb->setSavedReference(templatefile); } } + } } @@ -145,46 +279,11 @@ ChimeraCcodeCommand::ChimeraCcodeCommand(string option) { exit(1); } } -//********************************************************************************************************************** - -void ChimeraCcodeCommand::help(){ - try { - - m->mothurOut("The chimera.ccode command reads a fastafile and templatefile and outputs potentially chimeric sequences.\n"); - 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"); - #ifdef USE_MPI - m->mothurOut("When using MPI, the processors parameter is set to the number of MPI processes running. \n"); - #endif - m->mothurOut("The mask parameter allows you to specify a file containing one sequence you wish to use as a mask for the your sequences. \n"); - m->mothurOut("The window parameter allows you to specify the window size for searching for chimeras. \n"); - 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.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) { - m->errorOut(e, "ChimeraCcodeCommand", "help"); - exit(1); - } -} - //*************************************************************************************************************** - -ChimeraCcodeCommand::~ChimeraCcodeCommand(){ /* do nothing */ } - -//*************************************************************************************************************** - int ChimeraCcodeCommand::execute(){ try{ - if (abort == true) { return 0; } + if (abort == true) { if (calledHelp) { return 0; } return 2; } for (int s = 0; s < fastaFileNames.size(); s++) { @@ -201,27 +300,23 @@ 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"; - }else { - outputFileName = outputDir + getRootName(getSimpleName(fastaFileNames[s])) + "ccode.chimeras"; - accnosFileName = outputDir + getRootName(getSimpleName(fastaFileNames[s])) + "ccode.accnos"; - } - - string mapInfo = outputDir + getRootName(getSimpleName(fastaFileNames[s])) + "mapinfo"; - bool hasAccnos = true; - - if (m->control_pressed) { delete chimera; for (int j = 0; j < outputNames.size(); j++) { remove(outputNames[j].c_str()); } return 0; } + map variables; + variables["[filename]"] = outputDir + m->getRootName(m->getSimpleName(fastaFileNames[s])); + string mapInfo = getOutputFileName("mapinfo", variables); + if (maskfile != "") { variables["[tag]"] = maskfile; } + outputFileName = getOutputFileName("chimera", variables); + accnosFileName = getOutputFileName("accnos", variables); + + if (m->control_pressed) { delete chimera; for (int j = 0; j < outputNames.size(); j++) { m->mothurRemove(outputNames[j]); } outputTypes.clear(); return 0; } #ifdef USE_MPI - int pid, end, numSeqsPerProcessor; + int pid, numSeqsPerProcessor; int tag = 2001; - vector MPIPos; - MPIWroteAccnos = false; - + vector MPIPos; + MPI_Status status; MPI_Comm_rank(MPI_COMM_WORLD, &pid); //find out who we are MPI_Comm_size(MPI_COMM_WORLD, &processors); @@ -246,10 +341,10 @@ int ChimeraCcodeCommand::execute(){ 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()); } delete chimera; return 0; } + if (m->control_pressed) { MPI_File_close(&inMPI); MPI_File_close(&outMPI); MPI_File_close(&outMPIAccnos); for (int j = 0; j < outputNames.size(); j++) { m->mothurRemove(outputNames[j]); } outputTypes.clear(); delete chimera; return 0; } if (pid == 0) { //you are the root process - string outTemp = "For full window mapping info refer to " + mapInfo + "\n\n"; + string outTemp = "For full window mapping info refer to " + mapInfo + "\n"; //print header int length = outTemp.length(); @@ -259,7 +354,7 @@ 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++) { @@ -276,13 +371,8 @@ int ChimeraCcodeCommand::execute(){ //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()); } delete chimera; return 0; } + if (m->control_pressed) { MPI_File_close(&inMPI); MPI_File_close(&outMPI); MPI_File_close(&outMPIAccnos); m->mothurRemove(outputFileName); m->mothurRemove(accnosFileName); for (int j = 0; j < outputNames.size(); j++) { m->mothurRemove(outputNames[j]); } outputTypes.clear(); delete chimera; return 0; } - 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_Recv(&numSeqs, 1, MPI_INT, 0, tag, MPI_COMM_WORLD, &status); MPIPos.resize(numSeqs+1); @@ -297,9 +387,7 @@ int ChimeraCcodeCommand::execute(){ //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); for (int j = 0; j < outputNames.size(); j++) { remove(outputNames[j].c_str()); } delete chimera; return 0; } - - MPI_Send(&MPIWroteAccnos, 1, MPI_INT, 0, tag, MPI_COMM_WORLD); + if (m->control_pressed) { MPI_File_close(&inMPI); MPI_File_close(&outMPI); MPI_File_close(&outMPIAccnos); for (int j = 0; j < outputNames.size(); j++) { m->mothurRemove(outputNames[j]); } outputTypes.clear(); delete chimera; return 0; } } //close files @@ -308,110 +396,56 @@ int ChimeraCcodeCommand::execute(){ MPI_File_close(&outMPIAccnos); MPI_Barrier(MPI_COMM_WORLD); //make everyone wait - just in case - - //delete accnos file if blank - if (pid == 0) { - if (!MPIWroteAccnos) { - hasAccnos = false; - remove(accnosFileName.c_str()); - } - } #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(); + + //break up file - #if defined (__APPLE__) || (__MACH__) || (linux) || (__linux) + #if defined (__APPLE__) || (__MACH__) || (linux) || (__linux) || (__linux__) || (__unix__) || (__unix) + 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)])); + } + if(processors == 1){ - ifstream inFASTA; - openInputFile(fastaFileNames[s], inFASTA); - getNumSeqs(inFASTA, numSeqs); - inFASTA.close(); - - lines.push_back(new linePair(0, numSeqs)); - - driver(lines[0], outputFileName, fastaFileNames[s], accnosFileName); + + 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]; } lines.clear(); - delete chimera; - return 0; - } + if (m->control_pressed) { m->mothurRemove(outputFileName); m->mothurRemove(tempHeader); m->mothurRemove(accnosFileName); for (int j = 0; j < outputNames.size(); j++) { m->mothurRemove(outputNames[j]); } for (int i = 0; i < lines.size(); i++) { delete lines[i]; } outputTypes.clear(); 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(fastaFileNames[s], 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; - } - 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); + m->mothurRemove((outputFileName + toString(processIDS[i]) + ".temp")); } - vector nonBlankAccnosFiles; - //delete blank accnos files generated with multiple processes - for(int i=0;iappendFiles((accnosFileName + toString(processIDS[i]) + ".temp"), accnosFileName); + m->mothurRemove((accnosFileName + toString(processIDS[i]) + ".temp")); } - //append accnos files - if (nonBlankAccnosFiles.size() != 0) { - rename(nonBlankAccnosFiles[0].c_str(), accnosFileName.c_str()); - - for (int h=1; h < nonBlankAccnosFiles.size(); h++) { - appendFiles(nonBlankAccnosFiles[h], accnosFileName); - remove(nonBlankAccnosFiles[h].c_str()); - } - }else{ hasAccnos = false; } - 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()); } + m->mothurRemove(outputFileName); + m->mothurRemove(accnosFileName); + for (int j = 0; j < outputNames.size(); j++) { m->mothurRemove(outputNames[j]); } outputTypes.clear(); for (int i = 0; i < lines.size(); i++) { delete lines[i]; } lines.clear(); delete chimera; return 0; @@ -420,45 +454,38 @@ int ChimeraCcodeCommand::execute(){ } #else - ifstream inFASTA; - openInputFile(fastaFileNames[s], inFASTA); - getNumSeqs(inFASTA, numSeqs); - inFASTA.close(); - lines.push_back(new linePair(0, numSeqs)); + lines.push_back(new linePair(0, 1000)); + numSeqs = driver(lines[0], outputFileName, fastaFileNames[s], accnosFileName); - driver(lines[0], outputFileName, fastaFileNames[s], accnosFileName); + if (m->control_pressed) { m->mothurRemove(outputFileName); m->mothurRemove(tempHeader); m->mothurRemove(accnosFileName); for (int j = 0; j < outputNames.size(); j++) { m->mothurRemove(outputNames[j]); } for (int i = 0; i < lines.size(); i++) { delete lines[i]; } outputTypes.clear(); 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; - } - - //delete accnos file if its blank - if (isBlank(accnosFileName)) { remove(accnosFileName.c_str()); hasAccnos = false; } #endif - appendFiles(outputFileName, tempHeader); + m->appendFiles(outputFileName, tempHeader); - remove(outputFileName.c_str()); + m->mothurRemove(outputFileName); rename(tempHeader.c_str(), outputFileName.c_str()); #endif delete chimera; - outputNames.push_back(outputFileName); - outputNames.push_back(mapInfo); - if (hasAccnos) { outputNames.push_back(accnosFileName); } - + 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(); } + + //set accnos file as new current accnosfile + string current = ""; + itTypes = outputTypes.find("accnos"); + if (itTypes != outputTypes.end()) { + if ((itTypes->second).size() != 0) { current = (itTypes->second)[0]; m->setAccnosFile(current); } + } + m->mothurOutEndLine(); m->mothurOut("Output File Names: "); m->mothurOutEndLine(); for (int i = 0; i < outputNames.size(); i++) { m->mothurOut(outputNames[i]); m->mothurOutEndLine(); } @@ -474,24 +501,27 @@ 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 @@ -506,20 +536,28 @@ int ChimeraCcodeCommand::driver(linePair* line, string outputFName, string filen //print results chimera->print(out, out2); } + count++; } delete candidateSeq; + #if defined (__APPLE__) || (__MACH__) || (linux) || (__linux) || (__linux__) || (__unix__) || (__unix) + unsigned long long 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->mothurOutJustToScreen("Processing sequence: " + toString(count) + "\n"); } } //report progress - if((line->numSeqs) % 100 != 0){ m->mothurOut("Processing sequence: " + toString(line->numSeqs)); m->mothurOutEndLine(); } + if((count) % 100 != 0){ m->mothurOutJustToScreen("Processing sequence: " + toString(count) + "\n"); } out.close(); out2.close(); inFASTA.close(); - return 0; + return count; } catch(exception& e) { m->errorOut(e, "ChimeraCcodeCommand", "driver"); @@ -528,7 +566,7 @@ 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; @@ -551,7 +589,7 @@ 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 @@ -564,8 +602,7 @@ int ChimeraCcodeCommand::driverMPI(int start, int num, MPI_File& inMPI, MPI_File if (m->control_pressed) { delete candidateSeq; return 1; } //print results - bool isChimeric = chimera->print(outMPI, outAccMPI); - if (isChimeric) { MPIWroteAccnos = true; } + chimera->print(outMPI, outAccMPI); } } delete candidateSeq; @@ -590,9 +627,9 @@ int ChimeraCcodeCommand::driverMPI(int start, int num, MPI_File& inMPI, MPI_File int ChimeraCcodeCommand::createProcesses(string outputFileName, string filename, string accnos) { try { -#if defined (__APPLE__) || (__MACH__) || (linux) || (__linux) +#if defined (__APPLE__) || (__MACH__) || (linux) || (__linux) || (__linux__) || (__unix__) || (__unix) int process = 0; - // processIDS.resize(0); + int num = 0; //loop through and create all the processes you want while (process != processors) { @@ -602,9 +639,21 @@ 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("unable to spawn the necessary processes."); m->mothurOutEndLine(); 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); + } } //force parent to wait until all the processes are done @@ -613,7 +662,15 @@ 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(); m->mothurRemove(tempFile); + } + + return num; #endif } catch(exception& e) {