X-Git-Url: https://git.donarmstrong.com/?p=mothur.git;a=blobdiff_plain;f=distancecommand.cpp;h=9243df3873a9687d43da9da940b32da1ce9a5091;hp=a3cebe27ac3a0dec63f8b21b3d1a446864350d28;hb=cf9987b67aa49777a4c91c2d21f96e58bf17aa82;hpb=5694c92fbf646fe01abc87bde2af59e14a9a56b6 diff --git a/distancecommand.cpp b/distancecommand.cpp index a3cebe2..9243df3 100644 --- a/distancecommand.cpp +++ b/distancecommand.cpp @@ -8,76 +8,98 @@ */ #include "distancecommand.h" -#include "ignoregaps.h" -#include "eachgapdist.h" -#include "eachgapignore.h" -#include "onegapdist.h" -#include "onegapignore.h" - //********************************************************************************************************************** -vector DistanceCommand::getValidParameters(){ +vector DistanceCommand::setParameters(){ try { - string Array[] = {"fasta","oldfasta","column", "output", "calc", "countends", "cutoff", "processors", "outputdir","inputdir","compress"}; - vector myArray (Array, Array+(sizeof(Array)/sizeof(string))); + CommandParameter pcolumn("column", "InputTypes", "", "", "none", "none", "OldFastaColumn","column",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","phylip-column",false,true, true); parameters.push_back(pfasta); + CommandParameter poutput("output", "Multiple", "column-lt-square-phylip", "column", "", "", "","phylip-column",false,false, true); 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); + CommandParameter pprocessors("processors", "Number", "", "1", "", "", "","",false,false, true); parameters.push_back(pprocessors); + CommandParameter pcutoff("cutoff", "Number", "", "1.0", "", "", "","",false,false, true); parameters.push_back(pcutoff); + CommandParameter pinputdir("inputdir", "String", "", "", "", "", "","",false,false); parameters.push_back(pinputdir); + CommandParameter poutputdir("outputdir", "String", "", "", "", "", "","",false,false); parameters.push_back(poutputdir); + + vector myArray; + for (int i = 0; i < parameters.size(); i++) { myArray.push_back(parameters[i].name); } return myArray; } catch(exception& e) { - m->errorOut(e, "DistanceCommand", "getValidParameters"); + m->errorOut(e, "DistanceCommand", "setParameters"); exit(1); } } //********************************************************************************************************************** -DistanceCommand::DistanceCommand(){ +string DistanceCommand::getHelpString(){ try { - abort = true; - //initialize outputTypes - vector tempOutNames; - outputTypes["phylip"] = tempOutNames; - outputTypes["column"] = tempOutNames; + string helpString = ""; + helpString += "The dist.seqs command reads a file containing sequences and creates a distance file.\n"; + helpString += "The dist.seqs command parameters are fasta, oldfasta, column, calc, countends, output, compress, cutoff and processors. \n"; + helpString += "The fasta parameter is required, unless you have a valid current fasta file.\n"; + helpString += "The oldfasta and column parameters allow you to append the distances calculated to the column file.\n"; + helpString += "The calc parameter allows you to specify the method of calculating the distances. Your options are: nogaps, onegap or eachgap. The default is onegap.\n"; + helpString += "The countends parameter allows you to specify whether to include terminal gaps in distance. Your options are: T or F. The default is T.\n"; + helpString += "The cutoff parameter allows you to specify maximum distance to keep. The default is 1.0.\n"; + helpString += "The output parameter allows you to specify format of your distance matrix. Options are column, lt, and square. The default is column.\n"; + helpString += "The processors parameter allows you to specify number of processors to use. The default is 1.\n"; + helpString += "The compress parameter allows you to indicate that you want the resulting distance file compressed. The default is false.\n"; + helpString += "The dist.seqs command should be in the following format: \n"; + helpString += "dist.seqs(fasta=yourFastaFile, calc=yourCalc, countends=yourEnds, cutoff= yourCutOff, processors=yourProcessors) \n"; + helpString += "Example dist.seqs(fasta=amazon.fasta, calc=eachgap, countends=F, cutoff= 2.0, processors=3).\n"; + helpString += "Note: No spaces between parameter labels (i.e. calc), '=' and parameters (i.e.yourCalc).\n"; + return helpString; } catch(exception& e) { - m->errorOut(e, "DistanceCommand", "DistanceCommand"); + m->errorOut(e, "DistanceCommand", "getHelpString"); exit(1); } } //********************************************************************************************************************** -vector DistanceCommand::getRequiredParameters(){ - try { - string Array[] = {"fasta"}; - vector myArray (Array, Array+(sizeof(Array)/sizeof(string))); - return myArray; - } - catch(exception& e) { - m->errorOut(e, "DistanceCommand", "getRequiredParameters"); - exit(1); - } +string DistanceCommand::getOutputPattern(string type) { + try { + string pattern = ""; + + if (type == "phylip") { pattern = "[filename],[outputtag],dist"; } + else if (type == "column") { pattern = "[filename],dist"; } + else { m->mothurOut("[ERROR]: No definition for type " + type + " output pattern.\n"); m->control_pressed = true; } + + return pattern; + } + catch(exception& e) { + m->errorOut(e, "DistanceCommand", "getOutputPattern"); + exit(1); + } } //********************************************************************************************************************** -vector DistanceCommand::getRequiredFiles(){ +DistanceCommand::DistanceCommand(){ try { - vector myArray; - return myArray; + abort = true; calledHelp = true; + setParameters(); + vector tempOutNames; + outputTypes["phylip"] = tempOutNames; + outputTypes["column"] = tempOutNames; } catch(exception& e) { - m->errorOut(e, "DistanceCommand", "getRequiredFiles"); + m->errorOut(e, "DistanceCommand", "DistanceCommand"); exit(1); } } //********************************************************************************************************************** DistanceCommand::DistanceCommand(string option) { try { - abort = false; + abort = false; calledHelp = false; Estimators.clear(); //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","oldfasta","column", "output", "calc", "countends", "cutoff", "processors", "outputdir","inputdir","compress"}; - - vector myArray (Array, Array+(sizeof(Array)/sizeof(string))); + vector myArray = setParameters(); OptionParser parser(option); map parameters = parser.getParameters(); @@ -127,13 +149,21 @@ DistanceCommand::DistanceCommand(string option) { //check for required parameters fastafile = validParameter.validFile(parameters, "fasta", true); - if (fastafile == "not found") { m->mothurOut("fasta is a required parameter for the dist.seqs command."); m->mothurOutEndLine(); abort = true; } - else if (fastafile == "not open") { abort = true; } + if (fastafile == "not found") { + fastafile = m->getFastaFile(); + if (fastafile != "") { m->mothurOut("Using " + fastafile + " as input file for the fasta parameter."); m->mothurOutEndLine(); + ifstream inFASTA; + m->openInputFile(fastafile, inFASTA); + alignDB = SequenceDB(inFASTA); + inFASTA.close(); + }else { m->mothurOut("You have no current fastafile and the fasta parameter is required."); m->mothurOutEndLine(); abort = true; } + }else if (fastafile == "not open") { abort = true; } else{ ifstream inFASTA; m->openInputFile(fastafile, inFASTA); alignDB = SequenceDB(inFASTA); inFASTA.close(); + m->setFastaFile(fastafile); } oldfastafile = validParameter.validFile(parameters, "oldfasta", true); @@ -143,6 +173,7 @@ DistanceCommand::DistanceCommand(string option) { column = validParameter.validFile(parameters, "column", true); if (column == "not found") { column = ""; } else if (column == "not open") { abort = true; } + else { m->setColumnFile(column); } //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"){ @@ -164,41 +195,23 @@ 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 = "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, "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; } if ((column != "") && (oldfastafile != "") && (output != "column")) { m->mothurOut("You have provided column and oldfasta, indicating you want to append distances to your column file. Your output must be in column format to do so."); m->mothurOutEndLine(); abort=true; } if ((output != "column") && (output != "lt") && (output != "square")) { m->mothurOut(output + " is not a valid output form. Options are column, lt and square. I will use column."); m->mothurOutEndLine(); output = "column"; } - - ValidCalculators validCalculator; - - if (m->isTrue(countends) == true) { - for (int i=0; imothurOut("The dist.seqs command reads a file containing sequences and creates a distance file.\n"); - m->mothurOut("The dist.seqs command parameters are fasta, oldfasta, column, calc, countends, output, compress, cutoff and processors. \n"); - m->mothurOut("The fasta parameter is required.\n"); - m->mothurOut("The oldfasta and column parameters allow you to append the distances calculated to the column file.\n"); - m->mothurOut("The calc parameter allows you to specify the method of calculating the distances. Your options are: nogaps, onegap or eachgap. The default is onegap.\n"); - m->mothurOut("The countends parameter allows you to specify whether to include terminal gaps in distance. Your options are: T or F. The default is T.\n"); - m->mothurOut("The cutoff parameter allows you to specify maximum distance to keep. The default is 1.0.\n"); - m->mothurOut("The output parameter allows you to specify format of your distance matrix. Options are column, lt, and square. The default is column.\n"); - m->mothurOut("The processors parameter allows you to specify number of processors to use. The default is 1.\n"); - m->mothurOut("The compress parameter allows you to indicate that you want the resulting distance file compressed. The default is false.\n"); - m->mothurOut("The dist.seqs command should be in the following format: \n"); - m->mothurOut("dist.seqs(fasta=yourFastaFile, calc=yourCalc, countends=yourEnds, cutoff= yourCutOff, processors=yourProcessors) \n"); - m->mothurOut("Example dist.seqs(fasta=amazon.fasta, calc=eachgap, countends=F, cutoff= 2.0, processors=3).\n"); - m->mothurOut("Note: No spaces between parameter labels (i.e. calc), '=' and parameters (i.e.yourCalc).\n\n"); - } - catch(exception& e) { - m->errorOut(e, "DistanceCommand", "help"); - exit(1); - } -} //********************************************************************************************************************** int DistanceCommand::execute(){ try { - if (abort == true) { return 0; } + if (abort == true) { if (calledHelp) { return 0; } return 2; } int startTime = time(NULL); @@ -257,15 +241,20 @@ int DistanceCommand::execute(){ int numSeqs = alignDB.getNumSeqs(); cutoff += 0.005; + if (!alignDB.sameLength()) { m->mothurOut("[ERROR]: your sequences are not the same length, aborting."); m->mothurOutEndLine(); return 0; } + string outputFile; - + + map variables; + variables["[filename]"] = outputDir + m->getRootName(m->getSimpleName(fastafile)); if (output == "lt") { //does the user want lower triangle phylip formatted file - outputFile = outputDir + m->getRootName(m->getSimpleName(fastafile)) + "phylip.dist"; - remove(outputFile.c_str()); outputTypes["phylip"].push_back(outputFile); + variables["[outputtag]"] = "phylip"; + outputFile = getOutputFileName("phylip", variables); + 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 = getOutputFileName("column", variables); outputTypes["column"].push_back(outputFile); //so we don't accidentally overwrite @@ -274,10 +263,11 @@ int DistanceCommand::execute(){ rename(column.c_str(), tempcolumn.c_str()); } - remove(outputFile.c_str()); + m->mothurRemove(outputFile); }else { //assume square - outputFile = outputDir + m->getRootName(m->getSimpleName(fastafile)) + "square.dist"; - remove(outputFile.c_str()); + variables["[outputtag]"] = "square"; + outputFile = getOutputFileName("phylip", variables); + m->mothurRemove(outputFile); outputTypes["phylip"].push_back(outputFile); } @@ -320,11 +310,11 @@ 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); @@ -333,7 +323,7 @@ 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"); @@ -348,12 +338,12 @@ 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; @@ -370,9 +360,9 @@ 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); @@ -398,11 +388,11 @@ 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); @@ -411,14 +401,14 @@ 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; @@ -448,14 +438,14 @@ int DistanceCommand::execute(){ createProcesses(outputFile); } - #else + //#else //ifstream inFASTA; - if (output != "square") { driver(0, numSeqs, outputFile, cutoff); } - else { driver(0, numSeqs, outputFile, "square"); } - #endif + //if (output != "square") { driver(0, numSeqs, outputFile, cutoff); } + //else { driver(0, numSeqs, outputFile, "square"); } + //#endif #endif - if (m->control_pressed) { outputTypes.clear(); delete distCalculator; remove(outputFile.c_str()); return 0; } + if (m->control_pressed) { outputTypes.clear(); m->mothurRemove(outputFile); return 0; } #ifdef USE_MPI MPI_Comm_rank(MPI_COMM_WORLD, &pid); @@ -478,17 +468,17 @@ int DistanceCommand::execute(){ if (outputFile == column) { string tempcolumn = column + ".old"; m->appendFiles(tempcolumn, outputFile); - remove(tempcolumn.c_str()); + m->mothurRemove(tempcolumn); }else{ m->appendFiles(outputFile, column); - remove(outputFile.c_str()); + m->mothurRemove(outputFile); outputFile = column; } if (outputDir != "") { string newOutputName = outputDir + m->getSimpleName(outputFile); rename(outputFile.c_str(), newOutputName.c_str()); - remove(outputFile.c_str()); + m->mothurRemove(outputFile); outputFile = newOutputName; } } @@ -498,12 +488,23 @@ int DistanceCommand::execute(){ } #endif - if (m->control_pressed) { outputTypes.clear(); delete distCalculator; remove(outputFile.c_str()); return 0; } + if (m->control_pressed) { outputTypes.clear(); m->mothurRemove(outputFile); return 0; } - delete distCalculator; + //set phylip file as new current phylipfile + string current = ""; + itTypes = outputTypes.find("phylip"); + if (itTypes != outputTypes.end()) { + if ((itTypes->second).size() != 0) { current = (itTypes->second)[0]; m->setPhylipFile(current); } + } + + //set column file as new current columnfile + itTypes = outputTypes.find("column"); + if (itTypes != outputTypes.end()) { + if ((itTypes->second).size() != 0) { current = (itTypes->second)[0]; m->setColumnFile(current); } + } m->mothurOutEndLine(); - m->mothurOut("Output File Name: "); m->mothurOutEndLine(); + m->mothurOut("Output File Names: "); m->mothurOutEndLine(); m->mothurOut(outputFile); m->mothurOutEndLine(); m->mothurOutEndLine(); m->mothurOut("It took " + toString(time(NULL) - startTime) + " to calculate the distances for " + toString(numSeqs) + " sequences."); m->mothurOutEndLine(); @@ -527,7 +528,7 @@ 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(); @@ -560,25 +561,83 @@ void DistanceCommand::createProcesses(string filename) { int temp = processIDS[i]; wait(&temp); } +#else + ////////////////////////////////////////////////////////////////////////////////////////////////////// + //Windows version shared memory, so be careful when passing variables through the distanceData struct. + //Above fork() will clone, so memory is separate, but that's not the case with windows, + //that's why the distance calculator was moved inside of the driver to make separate copies. + ////////////////////////////////////////////////////////////////////////////////////////////////////// + + vector pDataArray; //[processors-1]; + DWORD dwThreadIdArray[processors-1]; + HANDLE hThreadArray[processors-1]; + + //Create processor-1 worker threads. + for( int i=0; icount != (pDataArray[i]->endLine-pDataArray[i]->startLine)) { + m->mothurOut("[ERROR]: process " + toString(i) + " only processed " + toString(pDataArray[i]->count) + " of " + toString(pDataArray[i]->endLine-pDataArray[i]->startLine) + " sequences assigned to it, quitting. \n"); m->control_pressed = true; + } + CloseHandle(hThreadArray[i]); + delete pDataArray[i]; + } +#endif //append and remove temp files for (int i=0;iappendFiles((filename + toString(processIDS[i]) + ".temp"), filename); - remove((filename + toString(processIDS[i]) + ".temp").c_str()); + m->mothurRemove((filename + toString(processIDS[i]) + ".temp")); } -#endif + } catch(exception& e) { m->errorOut(e, "DistanceCommand", "createProcesses"); exit(1); } } - /**************************************************************************************************/ /////// need to fix to work with calcs and sequencedb int DistanceCommand::driver(int startLine, int endLine, string dFileName, float cutoff){ try { - + ValidCalculators validCalculator; + Dist* distCalculator; + if (m->isTrue(countends) == true) { + for (int i=0; icontrol_pressed) { outFile.close(); return 0; } + if (m->control_pressed) { delete distCalculator; outFile.close(); return 0; } //if there was a column file given and we are appending, we don't want to calculate the distances that are already in the column file //the alignDB contains the new sequences and then the old, so if i an oldsequence and j is an old sequence then break out of this loop @@ -616,13 +675,14 @@ int DistanceCommand::driver(int startLine, int endLine, string dFileName, float if (output == "lt") { outFile << endl; } if(i % 100 == 0){ - m->mothurOut(toString(i) + "\t" + toString(time(NULL) - startTime)); m->mothurOutEndLine(); + m->mothurOutJustToScreen(toString(i) + "\t" + toString(time(NULL) - startTime)+"\n"); } } - m->mothurOut(toString(endLine-1) + "\t" + toString(time(NULL) - startTime)); m->mothurOutEndLine(); + m->mothurOutJustToScreen(toString(endLine-1) + "\t" + toString(time(NULL) - startTime)+"\n"); outFile.close(); + delete distCalculator; return 1; } @@ -635,7 +695,26 @@ int DistanceCommand::driver(int startLine, int endLine, string dFileName, float /////// need to fix to work with calcs and sequencedb int DistanceCommand::driver(int startLine, int endLine, string dFileName, string square){ try { - + ValidCalculators validCalculator; + Dist* distCalculator; + if (m->isTrue(countends) == true) { + for (int i=0; icontrol_pressed) { outFile.close(); return 0; } + if (m->control_pressed) { delete distCalculator; outFile.close(); return 0; } distCalculator->calcDist(alignDB.get(i), alignDB.get(j)); double dist = distCalculator->getDist(); @@ -666,13 +745,14 @@ int DistanceCommand::driver(int startLine, int endLine, string dFileName, string outFile << endl; if(i % 100 == 0){ - m->mothurOut(toString(i) + "\t" + toString(time(NULL) - startTime)); m->mothurOutEndLine(); + m->mothurOutJustToScreen(toString(i) + "\t" + toString(time(NULL) - startTime)+"\n"); } } - m->mothurOut(toString(endLine-1) + "\t" + toString(time(NULL) - startTime)); m->mothurOutEndLine(); + m->mothurOutJustToScreen(toString(endLine-1) + "\t" + toString(time(NULL) - startTime)+"\n"); outFile.close(); + delete distCalculator; return 1; } @@ -686,6 +766,28 @@ int DistanceCommand::driver(int startLine, int endLine, string dFileName, string /////// need to fix to work with calcs and sequencedb int DistanceCommand::driverMPI(int startLine, int endLine, MPI_File& outMPI, float cutoff){ try { + + ValidCalculators validCalculator; + Dist* distCalculator; + if (m->isTrue(countends) == true) { + for (int i=0; icontrol_pressed) { return 0; } + if (m->control_pressed) { delete distCalculator; return 0; } //if there was a column file given and we are appending, we don't want to calculate the distances that are already in the column file //the alignDB contains the new sequences and then the old, so if i an oldsequence and j is an old sequence then break out of this loop @@ -710,8 +812,7 @@ int DistanceCommand::driverMPI(int startLine, int endLine, MPI_File& outMPI, flo } if(i % 100 == 0){ - //m->mothurOut(toString(i) + "\t" + toString(time(NULL) - startTime)); m->mothurOutEndLine(); - cout << i << '\t' << (time(NULL) - startTime) << endl; + m->mothurOutJustToScreen(toString(i) + "\t" + toString(time(NULL) - startTime)+"\n"); } @@ -727,8 +828,8 @@ int DistanceCommand::driverMPI(int startLine, int endLine, MPI_File& outMPI, flo } - //m->mothurOut(toString(endLine-1) + "\t" + toString(time(NULL) - startTime)); m->mothurOutEndLine(); - cout << (endLine-1) << '\t' << (time(NULL) - startTime) << endl; + m->mothurOutJustToScreen(toString(endLine-1) + "\t" + toString(time(NULL) - startTime)+"\n"); + delete distCalculator; return 1; } catch(exception& e) { @@ -738,8 +839,29 @@ 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; + if (m->isTrue(countends) == true) { + for (int i=0; icontrol_pressed) { return 0; } + if (m->control_pressed) { delete distCalculator; return 0; } distCalculator->calcDist(alignDB.get(i), alignDB.get(j)); double dist = distCalculator->getDist(); @@ -783,9 +905,7 @@ int DistanceCommand::driverMPI(int startLine, int endLine, string file, unsigned if(i % 100 == 0){ - //m->mothurOut(toString(i) + "\t" + toString(time(NULL) - startTime)); m->mothurOutEndLine(); - cout << i << '\t' << (time(NULL) - startTime) << endl; - } + m->mothurOutJustToScreen(toString(i) + "\t" + toString(time(NULL) - startTime)+"\n"); } //send results to parent @@ -799,9 +919,10 @@ int DistanceCommand::driverMPI(int startLine, int endLine, string file, unsigned delete buf; } - //m->mothurOut(toString(endLine-1) + "\t" + toString(time(NULL) - startTime)); m->mothurOutEndLine(); - cout << (endLine-1) << '\t' << (time(NULL) - startTime) << endl; + m->mothurOutJustToScreen(toString(endLine-1) + "\t" + toString(time(NULL) - startTime)+"\n"); + MPI_File_close(&outMPI); + delete distCalculator; return 1; } @@ -812,8 +933,28 @@ 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; + if (m->isTrue(countends) == true) { + for (int i=0; icontrol_pressed) { return 0; } + if (m->control_pressed) { delete distCalculator; return 0; } distCalculator->calcDist(alignDB.get(i), alignDB.get(j)); double dist = distCalculator->getDist(); @@ -857,8 +998,7 @@ int DistanceCommand::driverMPI(int startLine, int endLine, string file, unsigned if(i % 100 == 0){ - //m->mothurOut(toString(i) + "\t" + toString(time(NULL) - startTime)); m->mothurOutEndLine(); - cout << i << '\t' << (time(NULL) - startTime) << endl; + m->mothurOutJustToScreen(toString(i) + "\t" + toString(time(NULL) - startTime)+"\n"); } @@ -873,10 +1013,10 @@ int DistanceCommand::driverMPI(int startLine, int endLine, string file, unsigned delete buf; } - //m->mothurOut(toString(endLine-1) + "\t" + toString(time(NULL) - startTime)); m->mothurOutEndLine(); - cout << (endLine-1) << '\t' << (time(NULL) - startTime) << endl; - MPI_File_close(&outMPI); + m->mothurOutJustToScreen(toString(endLine-1) + "\t" + toString(time(NULL) - startTime)+"\n"); + MPI_File_close(&outMPI); + delete distCalculator; return 1; } catch(exception& e) { @@ -893,7 +1033,7 @@ 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 @@ -928,7 +1068,7 @@ int DistanceCommand::convertMatrix(string outputFile) { //m->openInputFile(outfile, in); while(!in.eof()) { - if (m->control_pressed) { in.close(); remove(outfile.c_str()); out.close(); return 0; } + if (m->control_pressed) { in.close(); m->mothurRemove(outfile); out.close(); return 0; } in >> first >> second >> dist; m->gobble(in); @@ -963,7 +1103,7 @@ int DistanceCommand::convertMatrix(string outputFile) { in.close(); out.close(); - remove(outfile.c_str()); + m->mothurRemove(outfile); return 1; @@ -973,7 +1113,7 @@ int DistanceCommand::convertMatrix(string outputFile) { exit(1); } } -/************************************************************************************************** +************************************************************************************************** int DistanceCommand::convertToLowerTriangle(string outputFile) { try{ @@ -981,7 +1121,7 @@ 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 @@ -1018,7 +1158,7 @@ int DistanceCommand::convertToLowerTriangle(string outputFile) { //m->openInputFile(outfile, in); while(!in.eof()) { - if (m->control_pressed) { in.close(); remove(outfile.c_str()); out.close(); return 0; } + if (m->control_pressed) { in.close(); m->mothurRemove(outfile); out.close(); return 0; } in >> first >> second >> dist; m->gobble(in); @@ -1057,7 +1197,7 @@ int DistanceCommand::convertToLowerTriangle(string outputFile) { in.close(); out.close(); - remove(outfile.c_str()); + m->mothurRemove(outfile); return 1; @@ -1067,7 +1207,7 @@ 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. @@ -1128,7 +1268,7 @@ bool DistanceCommand::sanityCheck() { string name1, name2; float dist; while (!inDist.eof()) { - if (m->control_pressed) { inDist.close(); outDist.close(); remove(outputFile.c_str()); return good; } + if (m->control_pressed) { inDist.close(); outDist.close(); m->mothurRemove(outputFile); return good; } inDist >> name1 >> name2 >> dist; m->gobble(inDist); @@ -1145,10 +1285,10 @@ bool DistanceCommand::sanityCheck() { outDist.close(); if (good) { - remove(column.c_str()); + m->mothurRemove(column); rename(outputFile.c_str(), column.c_str()); }else{ - remove(outputFile.c_str()); //temp file is bad because file mismatch above + m->mothurRemove(outputFile); //temp file is bad because file mismatch above } return good;