]> git.donarmstrong.com Git - mothur.git/blobdiff - otuhierarchycommand.cpp
fixes while testing 1.33.0
[mothur.git] / otuhierarchycommand.cpp
index bc87211e0bde9a4581119a91592683e34262d006..5d521e7ea437906a36f4dd39fd203e1b23dc3bb0 100644 (file)
@@ -8,18 +8,84 @@
  */
 
 #include "otuhierarchycommand.h"
+#include "inputdata.h"
 
 //**********************************************************************************************************************
-OtuHierarchyCommand::OtuHierarchyCommand(string option){
+vector<string> OtuHierarchyCommand::setParameters(){   
        try {
-               abort = false;
+               CommandParameter poutput("output", "Multiple", "name-number", "name", "", "", "","",false,false); parameters.push_back(poutput);
+               CommandParameter plist("list", "InputTypes", "", "", "none", "none", "none","otuheirarchy",false,true,true); parameters.push_back(plist);
+               CommandParameter plabel("label", "String", "", "", "", "", "","",false,false); parameters.push_back(plabel);
+               CommandParameter pinputdir("inputdir", "String", "", "", "", "", "","",false,false); parameters.push_back(pinputdir);
+               CommandParameter poutputdir("outputdir", "String", "", "", "", "", "","",false,false); parameters.push_back(poutputdir);
+               
+               vector<string> myArray;
+               for (int i = 0; i < parameters.size(); i++) {   myArray.push_back(parameters[i].name);          }
+               return myArray;
+       }
+       catch(exception& e) {
+               m->errorOut(e, "OtuHierarchyCommand", "setParameters");
+               exit(1);
+       }
+}
+//**********************************************************************************************************************
+string OtuHierarchyCommand::getHelpString(){   
+       try {
+               string helpString = "";
+               helpString += "The otu.hierarchy command is used to see how otus relate at two distances. \n";
+               helpString += "The otu.hierarchy command parameters are list, label and output.  list and label parameters are required. \n";
+               helpString += "The output parameter allows you to output the names of the sequence in the OTUs or the OTU numbers. Options are name and number, default is name. \n";
+               helpString += "The otu.hierarchy command should be in the following format: \n";
+               helpString += "otu.hierarchy(list=yourListFile, label=yourLabels).\n";
+               helpString += "Example otu.hierarchy(list=amazon.fn.list, label=0.01-0.03).\n";
+               helpString += "The otu.hierarchy command outputs a .otu.hierarchy file which is described on the wiki.\n";
+               helpString += "Note: No spaces between parameter labels (i.e. list), '=' and parameters (i.e.yourListFile).\n";
+               return helpString;
+       }
+       catch(exception& e) {
+               m->errorOut(e, "OtuHierarchyCommand", "getHelpString");
+               exit(1);
+       }
+}
+//**********************************************************************************************************************
+string OtuHierarchyCommand::getOutputPattern(string type) {
+    try {
+        string pattern = "";
+        
+        if (type == "otuheirarchy") {  pattern = "[filename],[distance1],[tag],[distance2],otu.hierarchy"; } 
+        else { m->mothurOut("[ERROR]: No definition for type " + type + " output pattern.\n"); m->control_pressed = true;  }
+        
+        return pattern;
+    }
+    catch(exception& e) {
+        m->errorOut(e, "OtuHierarchyCommand", "getOutputPattern");
+        exit(1);
+    }
+}
+//**********************************************************************************************************************
+OtuHierarchyCommand::OtuHierarchyCommand(){    
+       try {
+               abort = true; calledHelp = true; 
+               setParameters();
+               vector<string> tempOutNames;
+               outputTypes["otuheirarchy"] = tempOutNames;
+       }
+       catch(exception& e) {
+               m->errorOut(e, "OtuHierarchyCommand", "OtuHierarchyCommand");
+               exit(1);
+       }
+}
+//**********************************************************************************************************************
+OtuHierarchyCommand::OtuHierarchyCommand(string option) {
+       try {
+               abort = false; calledHelp = false;   
+               
                //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[] =  {"list","label","outputdir","inputdir"};
-                       vector<string> myArray (Array, Array+(sizeof(Array)/sizeof(string)));
+                       vector<string> myArray = setParameters();
                        
                        OptionParser parser(option);
                        map<string,string> parameters = parser.getParameters();
@@ -32,6 +98,10 @@ OtuHierarchyCommand::OtuHierarchyCommand(string option){
                                if (validParameter.isValidParameter(it->first, myArray, it->second) != true) {  abort = true;  }
                        }
                        
+                       //initialize outputTypes
+                       vector<string> tempOutNames;
+                       outputTypes["otuheirarchy"] = 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 = "";          }
@@ -40,124 +110,122 @@ OtuHierarchyCommand::OtuHierarchyCommand(string option){
                                it = parameters.find("list");
                                //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["list"] = inputDir + it->second;             }
                                }
                        }
 
                        listFile = validParameter.validFile(parameters, "list", true);
-                       if (listFile == "not found") { mothurOut("list is a required parameter for the otu.hierarchy command."); mothurOutEndLine(); abort = true; }
-                       else if (listFile == "not open") { abort = true; }      
+                       if (listFile == "not found") { 
+                               listFile = m->getListFile(); 
+                               if (listFile != "") {  m->mothurOut("Using " + listFile + " as input file for the list parameter."); m->mothurOutEndLine(); }
+                               else { 
+                                       m->mothurOut("No valid current list file. You must provide a list file."); m->mothurOutEndLine(); 
+                                       abort = true;
+                               }
+                       }else if (listFile == "not open") { abort = true; }     
+                       else { m->setListFile(listFile); }
                        
                        //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(listFile); //if user entered a file with a path then preserve it   
+                               outputDir += m->hasPath(listFile); //if user entered a file with a path then preserve it        
                        }
                        
                        //check for optional parameter and set defaults
                        // ...at some point should added some additional type checking...
                        label = validParameter.validFile(parameters, "label", false);                   
-                       if (label == "not found") { mothurOut("label is a required parameter for the otu.hierarchy command."); mothurOutEndLine(); abort = true; }
+                       if (label == "not found") { m->mothurOut("label is a required parameter for the otu.hierarchy command."); m->mothurOutEndLine(); abort = true; }
                        else { 
-                               splitAtDash(label, labels);
-                               if (labels.size() != 2) { mothurOut("You must provide 2 labels."); mothurOutEndLine(); abort = true; }
-                       }                               
+                               m->splitAtDash(label, mylabels);
+                               if (mylabels.size() != 2) { m->mothurOut("You must provide 2 labels."); m->mothurOutEndLine(); abort = true; }
+                       }       
+                       
+                       output = validParameter.validFile(parameters, "output", false);                 if (output == "not found") { output = "name"; }
+                       
+                       if ((output != "name") && (output != "number")) { m->mothurOut("output options are name and number. I will use name."); m->mothurOutEndLine(); output = "name"; }
                }
                
        }
        catch(exception& e) {
-               errorOut(e, "OtuHierarchyCommand", "OtuHierarchyCommand");
+               m->errorOut(e, "OtuHierarchyCommand", "OtuHierarchyCommand");
                exit(1);
        }                       
 }
 //**********************************************************************************************************************
 
-void OtuHierarchyCommand::help(){
-       try {
-               mothurOut("The otu.hierarchy command is used to see how otus relate at two distances. \n");
-               mothurOut("The otu.hierarchy command parameters are list and label.  Both parameters are required. \n");
-               mothurOut("The otu.hierarchy command should be in the following format: \n");
-               mothurOut("otu.hierarchy(list=yourListFile, label=yourLabels).\n");
-               mothurOut("Example otu.hierarchy(list=amazon.fn.list, label=0.01-0.03).\n");
-               mothurOut("The otu.hierarchy command outputs a .otu.hierarchy file which is described on the wiki.\n");
-               mothurOut("Note: No spaces between parameter labels (i.e. list), '=' and parameters (i.e.yourListFile).\n\n");
-       }
-       catch(exception& e) {
-               errorOut(e, "OtuHierarchyCommand", "help");
-               exit(1);
-       }
-}
-
-//**********************************************************************************************************************
-
-OtuHierarchyCommand::~OtuHierarchyCommand(){}
-
-//**********************************************************************************************************************
-
 int OtuHierarchyCommand::execute(){
        try {
                
-               if (abort == true) { return 0; }
+               if (abort == true) { if (calledHelp) { return 0; }  return 2;   }
                
                //get listvectors that correspond to labels requested, (or use smart distancing to get closest listvector)
-               vector<ListVector> lists = getListVectors();
+               vector< vector<string> > lists = getListVectors();
+               
+               if (m->control_pressed) { outputTypes.clear(); return 0; }
                
                //determine which is little and which is big, putting little first
-               if (lists.size() == 2) {
+               if (lists.size() == 4) {
                        //if big is first swap them
-                       if (lists[0].getNumBins() < lists[1].getNumBins()) {
-                               reverse(lists.begin(), lists.end());
+                       if (lists[0].size() < lists[2].size()) {
+                               vector< vector<string> > tempLists;
+                tempLists.push_back(lists[2]);
+                tempLists.push_back(lists[3]);
+                tempLists.push_back(lists[0]);
+                tempLists.push_back(lists[1]);
+                lists = tempLists;
+                string tempLabel = list2Label;
+                list2Label = list1Label;
+                list1Label = tempLabel;
                        }
                }else{
-                       mothurOut("error getting listvectors, unable to read 2 different vectors, check your label inputs."); mothurOutEndLine(); return 0;
+                       m->mothurOut("error getting listvectors, unable to read 2 different vectors, check your label inputs."); m->mothurOutEndLine(); return 0;
                }
                
                //map sequences to bin number in the "little" otu
-               map<string, int> littleBins; 
-               for (int i = 0; i < lists[0].getNumBins(); i++) {
-                       string names = lists[0].get(i); 
-                       
-                       //parse bin
-                       while (names.find_first_of(',') != -1) { 
-                               string name = names.substr(0,names.find_first_of(','));
-                               names = names.substr(names.find_first_of(',')+1, names.length());
-                               littleBins[name] = i;  
-                       }
-                       
-                       //get last name
-                       littleBins[names] = i;
-               }
+               map<string, int> littleBins;
+        vector<string> binLabels0 = lists[0];
+               for (int i = 0; i < lists[0].size(); i++) {
+               
+                       if (m->control_pressed) {  return 0; }
+                       string bin = lists[1][i];
+            vector<string> names; m->splitAtComma(bin, names);
+                       for (int j = 0; j < names.size(); j++) { littleBins[names[j]] = i; }
+        }
                
                ofstream out;
-               string outputFileName = outputDir + getRootName(getSimpleName(listFile)) + lists[0].getLabel() + "-" + lists[1].getLabel() + ".otu.hierarchy";
-               openOutputFile(outputFileName, out);
+        map<string, string> variables; 
+        variables["[filename]"] = outputDir + m->getRootName(m->getSimpleName(listFile));
+        variables["[distance1]"] = list1Label;
+        variables["[tag]"] = "-"; 
+        variables["[distance2]"] = list2Label;
+               string outputFileName = getOutputFileName("otuheirarchy",variables);
+               m->openOutputFile(outputFileName, out);
                
                //go through each bin in "big" otu and output the bins in "little" otu which created it
-               for (int i = 0; i < lists[1].getNumBins(); i++) {
+        vector<string> binLabels1 = lists[2];
+               for (int i = 0; i < lists[2].size(); i++) {
                
-                       string names = lists[1].get(i);
+                       if (m->control_pressed) { outputTypes.clear(); out.close(); m->mothurRemove(outputFileName); return 0; }
+                       
+                       string binnames = lists[3][i];
+            vector<string> names; m->splitAtComma(binnames, names);
                        
                        //output column 1
-                       out << names << '\t';
+                       if (output == "name")   {   out << binnames << '\t';    }
+                       else                                    {       out << binLabels1[i] << '\t';           }
                        
                        map<int, int> bins; //bin numbers in little that are in this bin in big
                        map<int, int>::iterator it;
                        
                        //parse bin
-                       while (names.find_first_of(',') != -1) { 
-                               string name = names.substr(0,names.find_first_of(','));
-                               names = names.substr(names.find_first_of(',')+1, names.length());
-                               bins[littleBins[name]] = littleBins[name];  
-                       }
-                       
-                       //get last name
-                       bins[littleBins[names]] = littleBins[names]; 
+                       for (int j = 0; j < names.size(); j++) { bins[littleBins[names[j]]] = littleBins[names[j]];   }
                        
                        string col2 = "";
                        for (it = bins.begin(); it != bins.end(); it++) {
-                               col2 += lists[0].get(it->first) + "\t";
+                               if (output == "name")   {   col2 += lists[1][it->first] + "\t"; }
+                               else                                    {       col2 += binLabels0[it->first] + "\t";           }
                        }
                        
                        //output column 2
@@ -166,127 +234,131 @@ int OtuHierarchyCommand::execute(){
                
                out.close();
                
+               if (m->control_pressed) { outputTypes.clear(); m->mothurRemove(outputFileName); return 0; }
+               
+               m->mothurOutEndLine();
+               m->mothurOut("Output File Names: "); m->mothurOutEndLine();
+               m->mothurOut(outputFileName); m->mothurOutEndLine();    outputNames.push_back(outputFileName); outputTypes["otuheirarchy"].push_back(outputFileName); 
+               m->mothurOutEndLine();
+               
                return 0;
        }
        catch(exception& e) {
-               errorOut(e, "OtuHierarchyCommand", "execute");
+               m->errorOut(e, "OtuHierarchyCommand", "execute");
                exit(1);
        }
 }
 
 //**********************************************************************************************************************
 //returns a vector of listVectors where "little" vector is first
-vector<ListVector> OtuHierarchyCommand::getListVectors() {
+vector< vector<string> > OtuHierarchyCommand::getListVectors() { //return value [0] -> otulabelsFirstLabel [1] -> binsFirstLabel [2] -> otulabelsSecondLabel [3] -> binsSecondLabel
        try {
-               
-               int pos; //to use in smart distancing, position of last read in file
-               int lastPos;
-               vector<ListVector> lists;
+               vector< vector<string> > lists;
+        
+        int count = 0;
+        for (set<string>::iterator it = mylabels.begin(); it != mylabels.end(); it++) {
+            string realLabel;
+            vector< vector<string> > thisList = getListVector(*it, realLabel);
+            
+            if (m->control_pressed) {  return lists; }
+            
+            for (int i = 0; i < thisList.size(); i++) { lists.push_back(thisList[i]); }
+            
+            if (count == 0) {  list1Label = realLabel; count++; }
+            else {  list2Label = realLabel; }
+        }
+        
+        return lists;
+       }
+       catch(exception& e) {
+               m->errorOut(e, "OtuHierarchyCommand", "getListVectors");
+               exit(1);
+       }
+}
+//**********************************************************************************************************************
+vector< vector<string> > OtuHierarchyCommand::getListVector(string label, string& realLabel){ //return value [0] -> otulabels [1] -> bins
+       try {
+        vector< vector<string> > myList;
+        
+               InputData input(listFile, "list");
+               ListVector* list = input.getListVector();
+               string lastLabel = list->getLabel();
                
                //if the users enters label "0.06" and there is no "0.06" in their file use the next lowest label.
+               set<string> labels; labels.insert(label);
                set<string> processedLabels;
                set<string> userLabels = labels;
-
-               //open file
-               ifstream in;
-               openInputFile(listFile, in);
-               
-               //get first list vector in file
-               ListVector* list = NULL;
-               string lastLabel = "";
-               if (!in.eof())  {
-                       pos = in.tellg();
-                       lastPos = pos;
-                       list = new ListVector(in);  
-                       gobble(in);
-                       lastLabel = list->getLabel();
-               }
                
-               while ((list != NULL) && (userLabels.size() != 0)) {
+               //as long as you are not at the end of the file or done wih the lines you want
+               while((list != NULL) && (userLabels.size() != 0)) {
+                       if (m->control_pressed) {  return myList;  }
                        
-                       //is this a listvector that we want?
                        if(labels.count(list->getLabel()) == 1){
-                               
-                               //make copy of listvector
-                               ListVector temp(*list);
-                               lists.push_back(temp);
-                       
                                processedLabels.insert(list->getLabel());
                                userLabels.erase(list->getLabel());
+                               break;
                        }
-               
-                       //you have a label the user want that is smaller than this label and the last label has not already been processed 
-                       if ((anyLabelsToProcess(list->getLabel(), userLabels, "") == true) && (processedLabels.count(lastLabel) != 1)) {
+                       
+                       if ((m->anyLabelsToProcess(list->getLabel(), userLabels, "") == true) && (processedLabels.count(lastLabel) != 1)) {
                                string saveLabel = list->getLabel();
-                               int savePos = in.tellg();
                                
-                               //get smart distance line
                                delete list;
-                               in.seekg(lastPos);
-                               if (!in.eof())  {       
-                                       list = new ListVector(in);  
-                               }else { list = NULL; }
+                               list = input.getListVector(lastLabel);
                                
-                               //make copy of listvector
-                               ListVector temp(*list);
-                               lists.push_back(temp);
-                                       
                                processedLabels.insert(list->getLabel());
-                               userLabels.erase(list->getLabel());                                     
-                                                                               
+                               userLabels.erase(list->getLabel());
+                               
                                //restore real lastlabel to save below
-                               list->setLabel(saveLabel);
-                               in.seekg(savePos);
+                               //list->setLabel(saveLabel);
+                               break;
                        }
                        
                        lastLabel = list->getLabel();
-                       lastPos = pos;
                        
-                       //get next line
+                       //get next line to process
+                       //prevent memory leak
                        delete list;
-                       if (!in.eof())  {       
-                               pos = in.tellg();
-                               list = new ListVector(in);  
-                               gobble(in);
-                       }else { list = NULL; }
+                       list = input.getListVector();
                }
                
-                                               
+               
+               if (m->control_pressed) {  return myList;  }
                
                //output error messages about any remaining user labels
                set<string>::iterator it;
                bool needToRun = false;
-               for (it = userLabels.begin(); it != userLabels.end(); it++) {  
-                       mothurOut("Your file does not include the label " + *it); 
+               for (it = userLabels.begin(); it != userLabels.end(); it++) {
+                       m->mothurOut("Your file does not include the label " + *it);
                        if (processedLabels.count(lastLabel) != 1) {
-                               mothurOut(". I will use " + lastLabel + "."); mothurOutEndLine();
+                               m->mothurOut(". I will use " + lastLabel + "."); m->mothurOutEndLine();
                                needToRun = true;
                        }else {
-                               mothurOut(". Please refer to " + lastLabel + "."); mothurOutEndLine();
+                               m->mothurOut(". Please refer to " + lastLabel + "."); m->mothurOutEndLine();
                        }
                }
                
                //run last label if you need to
                if (needToRun == true)  {
-                       if (list != NULL) {     delete list;    }
-                       
-                       in.seekg(lastPos);
-                       if (!in.eof())  {       
-                               list = new ListVector(in); 
-                               
-                               //make copy of listvector
-                               ListVector temp(*list);
-                               lists.push_back(temp);
-                               
-                               delete list;
-                       }
+                       delete list;
+                       list = input.getListVector(lastLabel);
                }
                
-               
-               return lists;
+        //at this point the list vector has the right distance
+        myList.push_back(list->getLabels());
+        vector<string> bins;
+        for (int i = 0; i < list->getNumBins(); i++) {
+            if (m->control_pressed) {  return myList;  }
+            bins.push_back(list->get(i));
+        }
+        myList.push_back(bins);
+        realLabel = list->getLabel();
+        
+        delete list;
+        
+               return myList;
        }
        catch(exception& e) {
-               errorOut(e, "OtuHierarchyCommand", "getListVectors");
+               m->errorOut(e, "OtuHierarchyCommand", "getListVector");
                exit(1);
        }
 }