X-Git-Url: https://git.donarmstrong.com/?a=blobdiff_plain;f=hclustercommand.cpp;h=12f1f9e0b33904639cca446f0b0d698e9f64f625;hb=e51cb7e261265800fa2d2831b6516e33ebc7d78b;hp=5c08320bbed66718d9fa3fb335a36af6cfaf3a44;hpb=191ae1be0679d5cf4eda950b3b1bf26fb7dd503d;p=mothur.git diff --git a/hclustercommand.cpp b/hclustercommand.cpp index 5c08320..12f1f9e 100644 --- a/hclustercommand.cpp +++ b/hclustercommand.cpp @@ -10,52 +10,59 @@ #include "hclustercommand.h" //********************************************************************************************************************** -vector HClusterCommand::getValidParameters(){ +vector HClusterCommand::setParameters(){ try { - string Array[] = {"cutoff","hard","precision","method","phylip","column","name","sorted","showabund","timing","outputdir","inputdir"}; - vector myArray (Array, Array+(sizeof(Array)/sizeof(string))); + CommandParameter pphylip("phylip", "InputTypes", "", "", "PhylipColumn", "PhylipColumn", "none",false,false); parameters.push_back(pphylip); + CommandParameter pname("name", "InputTypes", "", "", "none", "none", "ColumnName",false,false); parameters.push_back(pname); + CommandParameter pcolumn("column", "InputTypes", "", "", "PhylipColumn", "PhylipColumn", "ColumnName",false,false); parameters.push_back(pcolumn); + CommandParameter pcutoff("cutoff", "Number", "", "10", "", "", "",false,false); parameters.push_back(pcutoff); + CommandParameter pprecision("precision", "Number", "", "100", "", "", "",false,false); parameters.push_back(pprecision); + CommandParameter pmethod("method", "Multiple", "furthest-nearest-average-weighted", "average", "", "", "",false,false); parameters.push_back(pmethod); + CommandParameter phard("hard", "Boolean", "", "T", "", "", "",false,false); parameters.push_back(phard); + CommandParameter psorted("sorted", "Boolean", "", "F", "", "", "",false,false); parameters.push_back(psorted); + CommandParameter pshowabund("showabund", "Boolean", "", "T", "", "", "",false,false); parameters.push_back(pshowabund); + CommandParameter ptiming("timing", "Boolean", "", "F", "", "", "",false,false); parameters.push_back(ptiming); + 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, "HClusterCommand", "getValidParameters"); - exit(1); - } -} -//********************************************************************************************************************** -HClusterCommand::HClusterCommand(){ - try { - abort = true; - //initialize outputTypes - vector tempOutNames; - outputTypes["list"] = tempOutNames; - outputTypes["rabund"] = tempOutNames; - outputTypes["sabund"] = tempOutNames; - } - catch(exception& e) { - m->errorOut(e, "HClusterCommand", "HClusterCommand"); + m->errorOut(e, "HClusterCommand", "setParameters"); exit(1); } } //********************************************************************************************************************** -vector HClusterCommand::getRequiredParameters(){ +string HClusterCommand::getHelpString(){ try { - string Array[] = {"phylip","column","or"}; - vector myArray (Array, Array+(sizeof(Array)/sizeof(string))); - return myArray; + string helpString = ""; + helpString += "The hcluster command parameter options are cutoff, precision, method, phylip, column, name, showabund, timing and sorted. Phylip or column and name are required, unless you have valid current files.\n"; + helpString += "The phylip and column parameter allow you to enter your distance file, and sorted indicates whether your column distance file is already sorted. \n"; + helpString += "The name parameter allows you to enter your name file and is required if your distance file is in column format. \n"; + helpString += "The hcluster command should be in the following format: \n"; + helpString += "hcluster(column=youDistanceFile, name=yourNameFile, method=yourMethod, cutoff=yourCutoff, precision=yourPrecision) \n"; + helpString += "The acceptable hcluster methods are furthest, nearest, weighted and average.\n"; + return helpString; } catch(exception& e) { - m->errorOut(e, "HClusterCommand", "getRequiredParameters"); + m->errorOut(e, "HClusterCommand", "getHelpString"); exit(1); } } //********************************************************************************************************************** -vector HClusterCommand::getRequiredFiles(){ +HClusterCommand::HClusterCommand(){ try { - vector myArray; - return myArray; + abort = true; calledHelp = true; + setParameters(); + vector tempOutNames; + outputTypes["list"] = tempOutNames; + outputTypes["rabund"] = tempOutNames; + outputTypes["sabund"] = tempOutNames; } catch(exception& e) { - m->errorOut(e, "HClusterCommand", "getRequiredFiles"); + m->errorOut(e, "HClusterCommand", "HClusterCommand"); exit(1); } } @@ -63,16 +70,13 @@ vector HClusterCommand::getRequiredFiles(){ //This function checks to make sure the cluster command has no errors and then clusters based on the method chosen. HClusterCommand::HClusterCommand(string option) { try{ - globaldata = GlobalData::getInstance(); - abort = false; + abort = false; calledHelp = false; //allow user to run help - if(option == "help") { help(); abort = true; } + if(option == "help") { help(); abort = true; calledHelp = true; } else { - //valid paramters for this command - string Array[] = {"cutoff","hard","precision","method","phylip","column","name","sorted","showabund","timing","outputdir","inputdir"}; - vector myArray (Array, Array+(sizeof(Array)/sizeof(string))); + vector myArray = setParameters(); OptionParser parser(option); map parameters = parser.getParameters(); @@ -93,8 +97,6 @@ HClusterCommand::HClusterCommand(string option) { 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 = ""; } @@ -143,11 +145,31 @@ HClusterCommand::HClusterCommand(string option) { if (namefile == "not open") { abort = true; } else if (namefile == "not found") { namefile = ""; } - if ((phylipfile == "") && (columnfile == "")) { m->mothurOut("When executing a hcluster command you must enter a phylip or a column."); m->mothurOutEndLine(); abort = true; } + if ((phylipfile == "") && (columnfile == "")) { + //is there are current file available for either of these? + //give priority to column, then phylip + columnfile = m->getColumnFile(); + if (columnfile != "") { m->mothurOut("Using " + columnfile + " as input file for the column parameter."); m->mothurOutEndLine(); } + else { + phylipfile = m->getPhylipFile(); + if (phylipfile != "") { m->mothurOut("Using " + phylipfile + " as input file for the phylip parameter."); m->mothurOutEndLine(); } + else { + m->mothurOut("No valid current files. You must provide a phylip or column file before you can use the hcluster command."); 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; } + if (namefile == "") { + namefile = m->getNameFile(); + if (namefile != "") { m->mothurOut("Using " + namefile + " as input file for the name parameter."); m->mothurOutEndLine(); } + else { + m->mothurOut("You need to provide a namefile if you are going to use the column format."); m->mothurOutEndLine(); + abort = true; + } + } } //check for optional parameter and set defaults @@ -160,7 +182,7 @@ HClusterCommand::HClusterCommand(string option) { length = temp.length(); convert(temp, precision); - temp = validParameter.validFile(parameters, "hard", false); if (temp == "not found") { temp = "F"; } + temp = validParameter.validFile(parameters, "hard", false); if (temp == "not found") { temp = "T"; } hard = m->isTrue(temp); temp = validParameter.validFile(parameters, "cutoff", false); @@ -169,10 +191,10 @@ HClusterCommand::HClusterCommand(string option) { cutoff += (5 / (precision * 10.0)); method = validParameter.validFile(parameters, "method", false); - if (method == "not found") { method = "furthest"; } + if (method == "not found") { method = "average"; } - if ((method == "furthest") || (method == "nearest") || (method == "average")) { } - else { m->mothurOut("Not a valid clustering method. Valid clustering algorithms are furthest, nearest or average."); m->mothurOutEndLine(); abort = true; } + if ((method == "furthest") || (method == "nearest") || (method == "average") || (method == "weighted")) { } + else { m->mothurOut("Not a valid clustering method. Valid clustering algorithms are furthest, nearest, average or weighted."); m->mothurOutEndLine(); abort = true; } showabund = validParameter.validFile(parameters, "showabund", false); if (showabund == "not found") { showabund = "T"; } @@ -192,6 +214,7 @@ HClusterCommand::HClusterCommand(string option) { if (method == "furthest") { tag = "fn"; } else if (method == "nearest") { tag = "nn"; } + else if (method == "weighted") { tag = "wn"; } else { tag = "an"; } m->openOutputFile(fileroot+ tag + ".sabund", sabundFile); @@ -212,45 +235,23 @@ HClusterCommand::HClusterCommand(string option) { //********************************************************************************************************************** -void HClusterCommand::help(){ - try { - 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) { - m->errorOut(e, "HClusterCommand", "help"); - exit(1); - } -} - -//********************************************************************************************************************** - -HClusterCommand::~HClusterCommand(){} - -//********************************************************************************************************************** - int HClusterCommand::execute(){ try { - if (abort == true) { return 0; } + if (abort == true) { if (calledHelp) { return 0; } return 2; } + NameAssignment* nameMap = NULL; if(namefile != ""){ - globaldata->nameMap = new NameAssignment(namefile); - globaldata->nameMap->readMap(); - }else{ - globaldata->nameMap = NULL; - } + nameMap = new NameAssignment(namefile); + nameMap->readMap(); + } time_t estart = time(NULL); if (!sorted) { read = new ReadCluster(distfile, cutoff, outputDir, true); read->setFormat(format); - read->read(globaldata->nameMap); + read->read(nameMap); if (m->control_pressed) { delete read; @@ -266,7 +267,7 @@ int HClusterCommand::execute(){ list = read->getListVector(); delete read; }else { - list = new ListVector(globaldata->nameMap->getListVector()); + list = new ListVector(nameMap->getListVector()); } if (m->control_pressed) { @@ -295,7 +296,7 @@ int HClusterCommand::execute(){ print_start = true; start = time(NULL); - cluster = new HCluster(rabund, list, method, distfile, globaldata->nameMap, cutoff); + cluster = new HCluster(rabund, list, method, distfile, nameMap, cutoff); vector seqs; seqs.resize(1); // to start loop if (m->control_pressed) { @@ -374,18 +375,7 @@ int HClusterCommand::execute(){ else if(rndPreviousDistgListVector; globaldata->gListVector = NULL; - - //saves .list file so you can do the collect, rarefaction and summary commands without doing a read.list - if (globaldata->getFormat() == "phylip") { globaldata->setPhylipFile(""); } - else if (globaldata->getFormat() == "column") { globaldata->setColumnFile(""); } - - globaldata->setListFile(fileroot+ tag + ".list"); - globaldata->setNameFile(""); - globaldata->setFormat("list"); - + sabundFile.close(); rabundFile.close(); listFile.close(); @@ -395,7 +385,26 @@ int HClusterCommand::execute(){ for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); } outputTypes.clear(); return 0; } - + + //set list file as new current listfile + string current = ""; + itTypes = outputTypes.find("list"); + if (itTypes != outputTypes.end()) { + if ((itTypes->second).size() != 0) { current = (itTypes->second)[0]; m->setListFile(current); } + } + + //set rabund file as new current rabundfile + itTypes = outputTypes.find("rabund"); + if (itTypes != outputTypes.end()) { + if ((itTypes->second).size() != 0) { current = (itTypes->second)[0]; m->setRabundFile(current); } + } + + //set sabund file as new current sabundfile + itTypes = outputTypes.find("sabund"); + if (itTypes != outputTypes.end()) { + if ((itTypes->second).size() != 0) { current = (itTypes->second)[0]; m->setSabundFile(current); } + } + m->mothurOutEndLine(); m->mothurOut("Output File Names: "); m->mothurOutEndLine();