X-Git-Url: https://git.donarmstrong.com/?a=blobdiff_plain;f=hclustercommand.cpp;h=e236d86cf2934a349500bbd2aacf2737d3f351b7;hb=8bc3e5b38c2317a1715f53be22fa96455868c281;hp=506ba544045c7c14e8944ef02d2b99a6e061acf3;hpb=b5a791c81d432082bf38755a08b33863f255341d;p=mothur.git diff --git a/hclustercommand.cpp b/hclustercommand.cpp index 506ba54..e236d86 100644 --- a/hclustercommand.cpp +++ b/hclustercommand.cpp @@ -9,9 +9,58 @@ #include "hclustercommand.h" +//********************************************************************************************************************** +vector HClusterCommand::getValidParameters(){ + try { + string Array[] = {"cutoff","hard","precision","method","phylip","column","name","sorted","showabund","timing","outputdir","inputdir"}; + vector myArray (Array, Array+(sizeof(Array)/sizeof(string))); + return myArray; + } + catch(exception& e) { + m->errorOut(e, "HClusterCommand", "getValidParameters"); + exit(1); + } +} +//********************************************************************************************************************** +HClusterCommand::HClusterCommand(){ + try { + //initialize outputTypes + vector tempOutNames; + outputTypes["list"] = tempOutNames; + outputTypes["rabund"] = tempOutNames; + outputTypes["sabund"] = tempOutNames; + } + catch(exception& e) { + m->errorOut(e, "HClusterCommand", "HClusterCommand"); + exit(1); + } +} +//********************************************************************************************************************** +vector HClusterCommand::getRequiredParameters(){ + try { + string Array[] = {"phylip","column","or"}; + vector myArray (Array, Array+(sizeof(Array)/sizeof(string))); + return myArray; + } + catch(exception& e) { + m->errorOut(e, "HClusterCommand", "getRequiredParameters"); + exit(1); + } +} +//********************************************************************************************************************** +vector HClusterCommand::getRequiredFiles(){ + try { + vector myArray; + return myArray; + } + catch(exception& e) { + m->errorOut(e, "HClusterCommand", "getRequiredFiles"); + exit(1); + } +} //********************************************************************************************************************** //This function checks to make sure the cluster command has no errors and then clusters based on the method chosen. -HClusterCommand::HClusterCommand(string option){ +HClusterCommand::HClusterCommand(string option) { try{ globaldata = GlobalData::getInstance(); abort = false; @@ -21,23 +70,63 @@ HClusterCommand::HClusterCommand(string option){ else { //valid paramters for this command - string Array[] = {"cutoff","precision","method","phylip","column","name","sorted","showabund","timing"}; + string Array[] = {"cutoff","hard","precision","method","phylip","column","name","sorted","showabund","timing","outputdir","inputdir"}; vector myArray (Array, Array+(sizeof(Array)/sizeof(string))); OptionParser parser(option); map parameters = parser.getParameters(); ValidParameters validParameter; + map::iterator it; //check to make sure all parameters are valid for command - for (map::iterator it = parameters.begin(); it != parameters.end(); it++) { + for (it = parameters.begin(); it != parameters.end(); it++) { if (validParameter.isValidParameter(it->first, myArray, it->second) != true) { abort = true; } } + //initialize outputTypes + vector tempOutNames; + outputTypes["list"] = tempOutNames; + outputTypes["rabund"] = tempOutNames; + outputTypes["sabund"] = tempOutNames; + globaldata->newRead(); + //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("phylip"); + //user has given a template file + if(it != parameters.end()){ + path = m->hasPath(it->second); + //if the user has not given a path then, add inputdir. else leave path alone. + if (path == "") { parameters["phylip"] = inputDir + it->second; } + } + + it = parameters.find("column"); + //user has given a template file + if(it != parameters.end()){ + path = m->hasPath(it->second); + //if the user has not given a path then, add inputdir. else leave path alone. + if (path == "") { parameters["column"] = inputDir + it->second; } + } + + it = parameters.find("name"); + //user has given a template file + if(it != parameters.end()){ + path = m->hasPath(it->second); + //if the user has not given a path then, add inputdir. else leave path alone. + if (path == "") { parameters["name"] = inputDir + it->second; } + } + } + + //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 = ""; } + //check for required parameters phylipfile = validParameter.validFile(parameters, "phylip", true); if (phylipfile == "not open") { abort = true; } @@ -53,8 +142,8 @@ HClusterCommand::HClusterCommand(string option){ if (namefile == "not open") { abort = true; } else if (namefile == "not found") { namefile = ""; } - if ((phylipfile == "") && (columnfile == "")) { mothurOut("When executing a hcluster command you must enter a phylip or a column."); mothurOutEndLine(); abort = true; } - else if ((phylipfile != "") && (columnfile != "")) { mothurOut("When executing a hcluster command you must enter ONLY ONE of the following: phylip or column."); mothurOutEndLine(); abort = true; } + if ((phylipfile == "") && (columnfile == "")) { m->mothurOut("When executing a hcluster command you must enter a phylip or a column."); m->mothurOutEndLine(); abort = true; } + else if ((phylipfile != "") && (columnfile != "")) { m->mothurOut("When executing a hcluster command you must enter ONLY ONE of the following: phylip or column."); m->mothurOutEndLine(); abort = true; } if (columnfile != "") { if (namefile == "") { cout << "You need to provide a namefile if you are going to use the column format." << endl; abort = true; } @@ -70,44 +159,52 @@ HClusterCommand::HClusterCommand(string option){ length = temp.length(); convert(temp, precision); + temp = validParameter.validFile(parameters, "hard", false); if (temp == "not found") { temp = "F"; } + hard = m->isTrue(temp); + temp = validParameter.validFile(parameters, "cutoff", false); if (temp == "not found") { temp = "10"; } convert(temp, cutoff); - cutoff += (5 / (precision * 10.0)); + cutoff += (5 / (precision * 10.0)); method = validParameter.validFile(parameters, "method", false); if (method == "not found") { method = "furthest"; } if ((method == "furthest") || (method == "nearest") || (method == "average")) { } - else { mothurOut("Not a valid clustering method. Valid clustering algorithms are furthest, nearest or average."); mothurOutEndLine(); abort = true; } + else { m->mothurOut("Not a valid clustering method. Valid clustering algorithms are furthest, nearest or average."); m->mothurOutEndLine(); abort = true; } showabund = validParameter.validFile(parameters, "showabund", false); if (showabund == "not found") { showabund = "T"; } sort = validParameter.validFile(parameters, "sorted", false); if (sort == "not found") { sort = "F"; } - sorted = isTrue(sort); + sorted = m->isTrue(sort); timing = validParameter.validFile(parameters, "timing", false); if (timing == "not found") { timing = "F"; } if (abort == false) { - - fileroot = getRootName(distfile); + + if (outputDir == "") { outputDir += m->hasPath(distfile); } + fileroot = outputDir + m->getRootName(m->getSimpleName(distfile)); if (method == "furthest") { tag = "fn"; } else if (method == "nearest") { tag = "nn"; } else { tag = "an"; } - openOutputFile(fileroot+ tag + ".sabund", sabundFile); - openOutputFile(fileroot+ tag + ".rabund", rabundFile); - openOutputFile(fileroot+ tag + ".list", listFile); + m->openOutputFile(fileroot+ tag + ".sabund", sabundFile); + m->openOutputFile(fileroot+ tag + ".rabund", rabundFile); + m->openOutputFile(fileroot+ tag + ".list", listFile); + + outputNames.push_back(fileroot+ tag + ".sabund"); outputTypes["sabund"].push_back(fileroot+ tag + ".sabund"); + outputNames.push_back(fileroot+ tag + ".rabund"); outputTypes["rabund"].push_back(fileroot+ tag + ".rabund"); + outputNames.push_back(fileroot+ tag + ".list"); outputTypes["list"].push_back(fileroot+ tag + ".list"); } } } catch(exception& e) { - errorOut(e, "HClusterCommand", "HClusterCommand"); + m->errorOut(e, "HClusterCommand", "HClusterCommand"); exit(1); } } @@ -116,15 +213,15 @@ HClusterCommand::HClusterCommand(string option){ void HClusterCommand::help(){ try { - mothurOut("The hcluster command parameter options are cutoff, precision, method, phylip, column, name, showabund, timing and sorted. Phylip or column and name are required.\n"); - mothurOut("The phylip and column parameter allow you to enter your distance file, and sorted indicates whether your column distance file is already sorted. \n"); - mothurOut("The name parameter allows you to enter your name file and is required if your distance file is in column format. \n"); - mothurOut("The hcluster command should be in the following format: \n"); - mothurOut("hcluster(column=youDistanceFile, name=yourNameFile, method=yourMethod, cutoff=yourCutoff, precision=yourPrecision) \n"); - mothurOut("The acceptable hcluster methods are furthest and nearest, but we hope to add average in the future.\n\n"); + m->mothurOut("The hcluster command parameter options are cutoff, precision, method, phylip, column, name, showabund, timing and sorted. Phylip or column and name are required.\n"); + m->mothurOut("The phylip and column parameter allow you to enter your distance file, and sorted indicates whether your column distance file is already sorted. \n"); + m->mothurOut("The name parameter allows you to enter your name file and is required if your distance file is in column format. \n"); + m->mothurOut("The hcluster command should be in the following format: \n"); + m->mothurOut("hcluster(column=youDistanceFile, name=yourNameFile, method=yourMethod, cutoff=yourCutoff, precision=yourPrecision) \n"); + m->mothurOut("The acceptable hcluster methods are furthest, nearest and average.\n\n"); } catch(exception& e) { - errorOut(e, "HClusterCommand", "help"); + m->errorOut(e, "HClusterCommand", "help"); exit(1); } } @@ -150,9 +247,19 @@ int HClusterCommand::execute(){ time_t estart = time(NULL); if (!sorted) { - read = new ReadCluster(distfile, cutoff); + read = new ReadCluster(distfile, cutoff, outputDir, true); read->setFormat(format); read->read(globaldata->nameMap); + + if (m->control_pressed) { + delete read; + sabundFile.close(); + rabundFile.close(); + listFile.close(); + for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); } outputTypes.clear(); + return 0; + } + distfile = read->getOutputFile(); list = read->getListVector(); @@ -160,15 +267,23 @@ int HClusterCommand::execute(){ }else { list = new ListVector(globaldata->nameMap->getListVector()); } - - mothurOut("It took " + toString(time(NULL) - estart) + " seconds to sort. "); mothurOutEndLine(); + + if (m->control_pressed) { + sabundFile.close(); + rabundFile.close(); + listFile.close(); + for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); } outputTypes.clear(); + return 0; + } + + m->mothurOut("It took " + toString(time(NULL) - estart) + " seconds to sort. "); m->mothurOutEndLine(); estart = time(NULL); //list vector made by read contains all sequence names if(list != NULL){ rabund = new RAbundVector(list->getRAbundVector()); }else{ - mothurOut("Error: no list vector!"); mothurOutEndLine(); return 0; + m->mothurOut("Error: no list vector!"); m->mothurOutEndLine(); return 0; } float previousDist = 0.00000; @@ -178,42 +293,63 @@ int HClusterCommand::execute(){ print_start = true; start = time(NULL); - - ifstream in; - openInputFile(distfile, in); - string firstName, secondName; - float distance; - - cluster = new HCluster(rabund, list, method); + + cluster = new HCluster(rabund, list, method, distfile, globaldata->nameMap, cutoff); vector seqs; seqs.resize(1); // to start loop + if (m->control_pressed) { + delete cluster; + sabundFile.close(); + rabundFile.close(); + listFile.close(); + for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); } outputTypes.clear(); + return 0; + } + + while (seqs.size() != 0){ - seqs = cluster->getSeqs(in, globaldata->nameMap, cutoff); - - for (int i = 0; i < seqs.size(); i++) { //-1 means skip me + seqs = cluster->getSeqs(); + + if (m->control_pressed) { + delete cluster; + sabundFile.close(); + rabundFile.close(); + listFile.close(); + for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); } outputTypes.clear(); + return 0; + } - if (print_start && isTrue(timing)) { - mothurOut("Clustering (" + tag + ") dist " + toString(distance) + "/" - + toString(roundDist(distance, precision)) - + "\t(precision: " + toString(precision) + ")"); - cout.flush(); - print_start = false; - } + for (int i = 0; i < seqs.size(); i++) { //-1 means skip me - if (seqs[i].seq1 != seqs[i].seq2) { cluster->update(seqs[i].seq1, seqs[i].seq2, seqs[i].dist); - float rndDist = roundDist(seqs[i].dist, precision); - + if (m->control_pressed) { + delete cluster; + sabundFile.close(); + rabundFile.close(); + listFile.close(); + for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); } outputTypes.clear(); + return 0; + } + + + float rndDist; + if (hard) { + rndDist = m->ceilDist(seqs[i].dist, precision); + }else{ + rndDist = m->roundDist(seqs[i].dist, precision); + } + + if((previousDist <= 0.0000) && (seqs[i].dist != previousDist)){ printData("unique"); } else if((rndDist != rndPreviousDist)){ printData(toString(rndPreviousDist, length-1)); } - + previousDist = seqs[i].dist; rndPreviousDist = rndDist; oldRAbund = *rabund; @@ -221,16 +357,16 @@ int HClusterCommand::execute(){ } } } - - in.close(); - if (print_start && isTrue(timing)) { - //mothurOut("Clustering (" + tag + ") for distance " + toString(previousDist) + "/" + toString(rndPreviousDist) - //+ "\t(precision: " + toString(precision) + ", Nodes: " + toString(matrix->getNNodes()) + ")"); - cout.flush(); - print_start = false; + if (m->control_pressed) { + delete cluster; + sabundFile.close(); + rabundFile.close(); + listFile.close(); + for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); } outputTypes.clear(); + return 0; } - + if(previousDist <= 0.0000){ printData("unique"); } @@ -253,13 +389,24 @@ int HClusterCommand::execute(){ rabundFile.close(); listFile.close(); delete cluster; - - mothurOut("It took " + toString(time(NULL) - estart) + " seconds to cluster. "); mothurOutEndLine(); + + if (m->control_pressed) { + for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); } outputTypes.clear(); + return 0; + } + + + m->mothurOutEndLine(); + m->mothurOut("Output File Names: "); m->mothurOutEndLine(); + for (int i = 0; i < outputNames.size(); i++) { m->mothurOut(outputNames[i]); m->mothurOutEndLine(); } + m->mothurOutEndLine(); + + m->mothurOut("It took " + toString(time(NULL) - estart) + " seconds to cluster. "); m->mothurOutEndLine(); return 0; } catch(exception& e) { - errorOut(e, "HClusterCommand", "execute"); + m->errorOut(e, "HClusterCommand", "execute"); exit(1); } } @@ -268,16 +415,16 @@ int HClusterCommand::execute(){ void HClusterCommand::printData(string label){ try { - if (isTrue(timing)) { - mothurOut("\tTime: " + toString(time(NULL) - start) + "\tsecs for " + toString(oldRAbund.getNumBins()) - + "\tclusters. Updates: " + toString(loops)); mothurOutEndLine(); + if (m->isTrue(timing)) { + m->mothurOut("\tTime: " + toString(time(NULL) - start) + "\tsecs for " + toString(oldRAbund.getNumBins()) + + "\tclusters. Updates: " + toString(loops)); m->mothurOutEndLine(); } print_start = true; loops = 0; start = time(NULL); oldRAbund.setLabel(label); - if (isTrue(showabund)) { + if (m->isTrue(showabund)) { oldRAbund.getSAbundVector().print(cout); } oldRAbund.print(rabundFile); @@ -287,7 +434,7 @@ void HClusterCommand::printData(string label){ oldList.print(listFile); } catch(exception& e) { - errorOut(e, "HClusterCommand", "printData"); + m->errorOut(e, "HClusterCommand", "printData"); exit(1); }