X-Git-Url: https://git.donarmstrong.com/?a=blobdiff_plain;f=distancecommand.cpp;h=16407d59b9819183edb70748d0b3a0ca1ff9acda;hb=e10c72304ee071c0c40e0218a06d89dc4731cbc2;hp=f7dff2e3cb3fb5e81cc8d17af0bcd8d303a5ab6a;hpb=fc1ed1ae1b022719176910ab71993bd6535810ad;p=mothur.git diff --git a/distancecommand.cpp b/distancecommand.cpp index f7dff2e..16407d5 100644 --- a/distancecommand.cpp +++ b/distancecommand.cpp @@ -15,7 +15,7 @@ vector DistanceCommand::setParameters(){ CommandParameter pcolumn("column", "InputTypes", "", "", "none", "none", "OldFastaColumn",false,false); parameters.push_back(pcolumn); CommandParameter poldfasta("oldfasta", "InputTypes", "", "", "none", "none", "OldFastaColumn",false,false); parameters.push_back(poldfasta); CommandParameter pfasta("fasta", "InputTypes", "", "", "none", "none", "none",false,true); parameters.push_back(pfasta); - CommandParameter poutput("output", "Multiple", "column-lt-square", "column", "", "", "",false,false); parameters.push_back(poutput); + CommandParameter poutput("output", "Multiple", "column-lt-square-phylip", "column", "", "", "",false,false); parameters.push_back(poutput); CommandParameter pcalc("calc", "Multiple", "nogaps-eachgap-onegap", "onegap", "", "", "",false,false); parameters.push_back(pcalc); CommandParameter pcountends("countends", "Boolean", "", "T", "", "", "",false,false); parameters.push_back(pcountends); CommandParameter pcompress("compress", "Boolean", "", "F", "", "", "",false,false); parameters.push_back(pcompress); @@ -59,6 +59,27 @@ string DistanceCommand::getHelpString(){ } } //********************************************************************************************************************** +string DistanceCommand::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 == "phylip") { outputFileName = "dist"; } + else if (type == "column") { outputFileName = "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, "DistanceCommand", "getOutputFileNameTag"); + exit(1); + } +} +//********************************************************************************************************************** DistanceCommand::DistanceCommand(){ try { abort = true; calledHelp = true; @@ -179,16 +200,17 @@ DistanceCommand::DistanceCommand(string option) { convert(temp, countends); temp = validParameter.validFile(parameters, "cutoff", false); if(temp == "not found"){ temp = "1.0"; } - convert(temp, cutoff); + m->mothurConvert(temp, cutoff); 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, "compress", false); if(temp == "not found"){ temp = "F"; } convert(temp, compress); output = validParameter.validFile(parameters, "output", false); if(output == "not found"){ output = "column"; } + if (output == "phylip") { output = "lt"; } if (((column != "") && (oldfastafile == "")) || ((column == "") && (oldfastafile != ""))) { m->mothurOut("If you provide column or oldfasta, you must provide both."); m->mothurOutEndLine(); abort=true; } @@ -229,12 +251,12 @@ int DistanceCommand::execute(){ string outputFile; if (output == "lt") { //does the user want lower triangle phylip formatted file - outputFile = outputDir + m->getRootName(m->getSimpleName(fastafile)) + "phylip.dist"; + outputFile = outputDir + m->getRootName(m->getSimpleName(fastafile)) + "phylip." + getOutputFileNameTag("phylip"); m->mothurRemove(outputFile); outputTypes["phylip"].push_back(outputFile); //output numSeqs to phylip formatted dist file }else if (output == "column") { //user wants column format - outputFile = outputDir + m->getRootName(m->getSimpleName(fastafile)) + "dist"; + outputFile = outputDir + m->getRootName(m->getSimpleName(fastafile)) + getOutputFileNameTag("column"); outputTypes["column"].push_back(outputFile); //so we don't accidentally overwrite @@ -245,7 +267,7 @@ int DistanceCommand::execute(){ m->mothurRemove(outputFile); }else { //assume square - outputFile = outputDir + m->getRootName(m->getSimpleName(fastafile)) + "square.dist"; + outputFile = outputDir + m->getRootName(m->getSimpleName(fastafile)) + "square." + getOutputFileNameTag("phylip"); m->mothurRemove(outputFile); outputTypes["phylip"].push_back(outputFile); } @@ -289,11 +311,11 @@ int DistanceCommand::execute(){ driverMPI(start, end, outMPI, cutoff); - if (m->control_pressed) { outputTypes.clear(); MPI_File_close(&outMPI); delete distCalculator; return 0; } + if (m->control_pressed) { outputTypes.clear(); MPI_File_close(&outMPI); return 0; } //wait on chidren for(int i = 1; i < processors; i++) { - if (m->control_pressed) { outputTypes.clear(); MPI_File_close(&outMPI); delete distCalculator; return 0; } + if (m->control_pressed) { outputTypes.clear(); MPI_File_close(&outMPI); return 0; } char buf[5]; MPI_Recv(buf, 5, MPI_CHAR, i, tag, MPI_COMM_WORLD, &status); @@ -302,7 +324,7 @@ int DistanceCommand::execute(){ //do your part driverMPI(start, end, outMPI, cutoff); - if (m->control_pressed) { outputTypes.clear(); MPI_File_close(&outMPI); delete distCalculator; return 0; } + if (m->control_pressed) { outputTypes.clear(); MPI_File_close(&outMPI); return 0; } char buf[5]; strcpy(buf, "done"); @@ -317,12 +339,12 @@ int DistanceCommand::execute(){ //do your part string outputMyPart; - unsigned long int mySize; + unsigned long long mySize; if (output != "square"){ driverMPI(start, end, outputFile, mySize); } else { driverMPI(start, end, outputFile, mySize, output); } - if (m->control_pressed) { outputTypes.clear(); delete distCalculator; return 0; } + if (m->control_pressed) { outputTypes.clear(); return 0; } int amode=MPI_MODE_APPEND|MPI_MODE_WRONLY|MPI_MODE_CREATE; // MPI_File outMPI; @@ -339,9 +361,9 @@ int DistanceCommand::execute(){ //wait on chidren for(int b = 1; b < processors; b++) { - unsigned long int fileSize; + unsigned long long fileSize; - if (m->control_pressed) { outputTypes.clear(); MPI_File_close(&outMPI); delete distCalculator; return 0; } + if (m->control_pressed) { outputTypes.clear(); MPI_File_close(&outMPI); return 0; } MPI_Recv(&fileSize, 1, MPI_LONG, b, tag, MPI_COMM_WORLD, &status); @@ -367,11 +389,11 @@ int DistanceCommand::execute(){ MPI_File_close(&outMPI); }else { //you are a child process //do your part - unsigned long int size; + unsigned long long size; if (output != "square"){ driverMPI(start, end, (outputFile + toString(pid) + ".temp"), size); } else { driverMPI(start, end, (outputFile + toString(pid) + ".temp"), size, output); } - if (m->control_pressed) { delete distCalculator; return 0; } + if (m->control_pressed) { return 0; } //tell parent you are done. MPI_Send(&size, 1, MPI_LONG, 0, tag, MPI_COMM_WORLD); @@ -380,14 +402,14 @@ int DistanceCommand::execute(){ MPI_Barrier(MPI_COMM_WORLD); //make everyone wait - just in case #else - //#if defined (__APPLE__) || (__MACH__) || (linux) || (__linux) + //#if defined (__APPLE__) || (__MACH__) || (linux) || (__linux) || (__linux__) || (__unix__) || (__unix) //if you don't need to fork anything if(processors == 1){ if (output != "square") { driver(0, numSeqs, outputFile, cutoff); } else { driver(0, numSeqs, outputFile, "square"); } }else{ //you have multiple processors - unsigned long int numDists = 0; + unsigned long long numDists = 0; if (output == "square") { numDists = numSeqs * numSeqs; @@ -507,7 +529,7 @@ int DistanceCommand::execute(){ /**************************************************************************************************/ void DistanceCommand::createProcesses(string filename) { try { -#if defined (__APPLE__) || (__MACH__) || (linux) || (__linux) +#if defined (__APPLE__) || (__MACH__) || (linux) || (__linux) || (__linux__) || (__unix__) || (__unix) int process = 1; processIDS.clear(); @@ -817,7 +839,7 @@ int DistanceCommand::driverMPI(int startLine, int endLine, MPI_File& outMPI, flo } /**************************************************************************************************/ /////// need to fix to work with calcs and sequencedb -int DistanceCommand::driverMPI(int startLine, int endLine, string file, unsigned long int& size){ +int DistanceCommand::driverMPI(int startLine, int endLine, string file, unsigned long long& size){ try { ValidCalculators validCalculator; Dist* distCalculator; @@ -913,7 +935,7 @@ int DistanceCommand::driverMPI(int startLine, int endLine, string file, unsigned } /**************************************************************************************************/ /////// need to fix to work with calcs and sequencedb -int DistanceCommand::driverMPI(int startLine, int endLine, string file, unsigned long int& size, string square){ +int DistanceCommand::driverMPI(int startLine, int endLine, string file, unsigned long long& size, string square){ try { ValidCalculators validCalculator; Dist* distCalculator; @@ -1014,7 +1036,7 @@ int DistanceCommand::convertMatrix(string outputFile) { string outfile = m->getRootName(outputFile) + "sorted.dist.temp"; //use the unix sort - #if defined (__APPLE__) || (__MACH__) || (linux) || (__linux) + #if defined (__APPLE__) || (__MACH__) || (linux) || (__linux) || (__linux__) || (__unix__) || (__unix) string command = "sort -n " + outputFile + " -o " + outfile; system(command.c_str()); #else //sort using windows sort @@ -1094,7 +1116,7 @@ int DistanceCommand::convertMatrix(string outputFile) { exit(1); } } -/************************************************************************************************** +************************************************************************************************** int DistanceCommand::convertToLowerTriangle(string outputFile) { try{ @@ -1102,7 +1124,7 @@ int DistanceCommand::convertToLowerTriangle(string outputFile) { string outfile = m->getRootName(outputFile) + "sorted.dist.temp"; //use the unix sort - #if defined (__APPLE__) || (__MACH__) || (linux) || (__linux) + #if defined (__APPLE__) || (__MACH__) || (linux) || (__linux) || (__linux__) || (__unix__) || (__unix) string command = "sort -n " + outputFile + " -o " + outfile; system(command.c_str()); #else //sort using windows sort @@ -1188,7 +1210,7 @@ int DistanceCommand::convertToLowerTriangle(string outputFile) { exit(1); } } -/**************************************************************************************************/ +**************************************************************************************************/ //its okay if the column file does not contain all the names in the fasta file, since some distance may have been above a cutoff, //but no sequences can be in the column file that are not in oldfasta. also, if a distance is above the cutoff given then remove it. //also check to make sure the 2 files have the same alignment length.