]> git.donarmstrong.com Git - mothur.git/blobdiff - rarefactcommand.cpp
added checks for ^C to quit command instead of program
[mothur.git] / rarefactcommand.cpp
index 41c38e86899f35d0dfc93e6f54a6f63b6f845254..566561e9e695c3f74f7ffb1ce0b1d56ee0e3bc20 100644 (file)
 //**********************************************************************************************************************
 
 
-RareFactCommand::RareFactCommand(string option){
+RareFactCommand::RareFactCommand(string option)  {
        try {
                globaldata = GlobalData::getInstance();
                abort = false;
                allLines = 1;
-               lines.clear();
                labels.clear();
                Estimators.clear();
-               
+                               
                //allow user to run help
-               if(option == "help") { validCalculator = new ValidCalculators(); help(); abort = true; }
+               if(option == "help") { validCalculator = new ValidCalculators(); help(); delete validCalculator; abort = true; }
                
                else {
                        //valid paramters for this command
-                       string Array[] =  {"iters","freq","line","label","calc","abund"};
+                       string Array[] =  {"iters","freq","label","calc","abund","outputdir","inputdir"};
                        vector<string> myArray (Array, Array+(sizeof(Array)/sizeof(string)));
                        
                        OptionParser parser(option);
@@ -49,18 +48,17 @@ RareFactCommand::RareFactCommand(string option){
                                if (validParameter.isValidParameter(it->first, myArray, it->second) != true) {  abort = true;  }
                        }
                        
+                       //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(globaldata->inputFileName); //if user entered a file with a path then preserve it  
+                       }
+
                        //make sure the user has already run the read.otu command
-                       if ((globaldata->getListFile() == "") && (globaldata->getRabundFile() == "") && (globaldata->getSabundFile() == "")) { mothurOut("You must read a list, sabund or rabund before you can use the rarefaction.single command."); mothurOutEndLine(); abort = true; }
+                       if ((globaldata->getSharedFile() == "") && (globaldata->getListFile() == "") && (globaldata->getRabundFile() == "") && (globaldata->getSabundFile() == "")) { m->mothurOut("You must read a list, sabund, rabund or shared file before you can use the rarefact.single command."); m->mothurOutEndLine(); abort = true; }
                        
                        //check for optional parameter and set defaults
                        // ...at some point should added some additional type checking...
-                       line = validParameter.validFile(parameters, "line", false);                             
-                       if (line == "not found") { line = "";  }
-                       else { 
-                               if(line != "all") {  splitAtDash(line, lines);  allLines = 0;  }
-                               else { allLines = 1;  }
-                       }
-                       
                        label = validParameter.validFile(parameters, "label", false);                   
                        if (label == "not found") { label = ""; }
                        else { 
@@ -68,13 +66,10 @@ RareFactCommand::RareFactCommand(string option){
                                else { allLines = 1;  }
                        }
                        
-                       //make sure user did not use both the line and label parameters
-                       if ((line != "") && (label != "")) { mothurOut("You cannot use both the line and label parameters at the same time. "); mothurOutEndLine(); abort = true; }
-                       //if the user has not specified any line or labels use the ones from read.otu
-                       else if((line == "") && (label == "")) {  
+                       //if the user has not specified any labels use the ones from read.otu
+                       if(label == "") {  
                                allLines = globaldata->allLines; 
                                labels = globaldata->labels; 
-                               lines = globaldata->lines;
                        }
                                
                        calc = validParameter.validFile(parameters, "calc", false);                     
@@ -93,48 +88,11 @@ RareFactCommand::RareFactCommand(string option){
                        
                        temp = validParameter.validFile(parameters, "iters", false);                    if (temp == "not found") { temp = "1000"; }
                        convert(temp, nIters); 
-                       
-                       if (abort == false) {
-                       
-                               string fileNameRoot = getRootName(globaldata->inputFileName);
-                               int i;
-                               validCalculator = new ValidCalculators();
-                               
-                               
-                               for (i=0; i<Estimators.size(); i++) {
-                                       if (validCalculator->isValidCalculator("rarefaction", Estimators[i]) == true) { 
-                                               if (Estimators[i] == "sobs") { 
-                                                       rDisplays.push_back(new RareDisplay(new Sobs(), new ThreeColumnFile(fileNameRoot+"rarefaction")));
-                                               }else if (Estimators[i] == "chao") { 
-                                                       rDisplays.push_back(new RareDisplay(new Chao1(), new ThreeColumnFile(fileNameRoot+"r_chao")));
-                                               }else if (Estimators[i] == "ace") { 
-                                                       if(abund < 5)
-                                                               abund = 10;
-                                                       rDisplays.push_back(new RareDisplay(new Ace(abund), new ThreeColumnFile(fileNameRoot+"r_ace")));
-                                               }else if (Estimators[i] == "jack") { 
-                                                       rDisplays.push_back(new RareDisplay(new Jackknife(), new ThreeColumnFile(fileNameRoot+"r_jack")));
-                                               }else if (Estimators[i] == "shannon") { 
-                                                       rDisplays.push_back(new RareDisplay(new Shannon(), new ThreeColumnFile(fileNameRoot+"r_shannon")));
-                                               }else if (Estimators[i] == "npshannon") { 
-                                                       rDisplays.push_back(new RareDisplay(new NPShannon(), new ThreeColumnFile(fileNameRoot+"r_npshannon")));
-                                               }else if (Estimators[i] == "simpson") { 
-                                                       rDisplays.push_back(new RareDisplay(new Simpson(), new ThreeColumnFile(fileNameRoot+"r_simpson")));
-                                               }else if (Estimators[i] == "bootstrap") { 
-                                                       rDisplays.push_back(new RareDisplay(new Bootstrap(), new ThreeColumnFile(fileNameRoot+"r_bootstrap")));
-                                               }else if (Estimators[i] == "coverage") { 
-                                                       rDisplays.push_back(new RareDisplay(new Coverage(), new ThreeColumnFile(fileNameRoot+"r_coverage")));
-                                               }else if (Estimators[i] == "nseqs") { 
-                                                       rDisplays.push_back(new RareDisplay(new NSeqs(), new ThreeColumnFile(fileNameRoot+"r_nseqs")));
-                                               }
-                                       }
-                               }
-                       }
-                               
                }
                
        }
        catch(exception& e) {
-               errorOut(e, "RareFactCommand", "RareFactCommand");
+               m->errorOut(e, "RareFactCommand", "RareFactCommand");
                exit(1);
        }
 }
@@ -142,33 +100,26 @@ RareFactCommand::RareFactCommand(string option){
 
 void RareFactCommand::help(){
        try {
-               mothurOut("The rarefaction.single command can only be executed after a successful read.otu WTIH ONE EXECEPTION.\n");
-               mothurOut("The rarefaction.single command can be executed after a successful cluster command.  It will use the .list file from the output of the cluster.\n");
-               mothurOut("The rarefaction.single command parameters are label, line, iters, freq, calc and abund.  No parameters are required, but you may not use \n");
-               mothurOut("both the line and label parameters at the same time. The rarefaction.single command should be in the following format: \n");
-               mothurOut("rarefaction.single(label=yourLabel, line=yourLines, iters=yourIters, freq=yourFreq, calc=yourEstimators).\n");
-               mothurOut("Example rarefaction.single(label=unique-.01-.03, line=0-5-10, iters=10000, freq=10, calc=sobs-rchao-race-rjack-rbootstrap-rshannon-rnpshannon-rsimpson).\n");
-               mothurOut("The default values for iters is 1000, freq is 100, and calc is rarefaction which calculates the rarefaction curve for the observed richness.\n");
+               m->mothurOut("The rarefaction.single command can only be executed after a successful read.otu WTIH ONE EXECEPTION.\n");
+               m->mothurOut("The rarefaction.single command can be executed after a successful cluster command.  It will use the .list file from the output of the cluster.\n");
+               m->mothurOut("The rarefaction.single command parameters are label, iters, freq, calc and abund.  No parameters are required. \n");
+               m->mothurOut("The rarefaction.single command should be in the following format: \n");
+               m->mothurOut("rarefaction.single(label=yourLabel, iters=yourIters, freq=yourFreq, calc=yourEstimators).\n");
+               m->mothurOut("Example rarefaction.single(label=unique-.01-.03, iters=10000, freq=10, calc=sobs-rchao-race-rjack-rbootstrap-rshannon-rnpshannon-rsimpson).\n");
+               m->mothurOut("The default values for iters is 1000, freq is 100, and calc is rarefaction which calculates the rarefaction curve for the observed richness.\n");
                validCalculator->printCalc("rarefaction", cout);
-               mothurOut("The label and line parameters are used to analyze specific lines in your input.\n");
-               mothurOut("Note: No spaces between parameter labels (i.e. freq), '=' and parameters (i.e.yourFreq).\n\n");
+               m->mothurOut("The label parameter is used to analyze specific labels in your input.\n");
+               m->mothurOut("Note: No spaces between parameter labels (i.e. freq), '=' and parameters (i.e.yourFreq).\n\n");
        }
        catch(exception& e) {
-               errorOut(e, "RareFactCommand", "help");
+               m->errorOut(e, "RareFactCommand", "help");
                exit(1);
        }
 }
 
 //**********************************************************************************************************************
 
-RareFactCommand::~RareFactCommand(){
-       if (abort == false) {
-               globaldata->gorder = NULL;
-               delete input;  globaldata->ginput = NULL;
-               delete read;
-               delete validCalculator;
-       }
-}
+RareFactCommand::~RareFactCommand(){}
 
 //**********************************************************************************************************************
 
@@ -177,92 +128,237 @@ int RareFactCommand::execute(){
        
                if (abort == true) { return 0; }
                
-               int count = 1;
+               vector<string> outputNames;
                
-               //if the users entered no valid calculators don't execute command
-               if (rDisplays.size() == 0) { return 0; }
-
-               read = new ReadOTUFile(globaldata->inputFileName);      
-               read->read(&*globaldata); 
-
-               order = globaldata->gorder;
-               string lastLabel = order->getLabel();
-               input = globaldata->ginput;
+               if ((globaldata->getFormat() != "sharedfile")) { inputFileNames.push_back(globaldata->inputFileName);  }
+               else {  inputFileNames = parseSharedFile(globaldata->getSharedFile());  globaldata->setFormat("rabund");  }
                
-               //if the users enters label "0.06" and there is no "0.06" in their file use the next lowest label.
-               set<string> processedLabels;
-               set<string> userLabels = labels;
-               set<int> userLines = lines;
-       
-               //as long as you are not at the end of the file or done wih the lines you want
-               while((order != NULL) && ((allLines == 1) || (userLabels.size() != 0) || (userLines.size() != 0))) {
+               if (m->control_pressed) { return 0; }
+               
+               for (int p = 0; p < inputFileNames.size(); p++) {
                        
-                       if(allLines == 1 || lines.count(count) == 1 || labels.count(order->getLabel()) == 1){
+                       string fileNameRoot = outputDir + getRootName(getSimpleName(inputFileNames[p]));
+                       globaldata->inputFileName = inputFileNames[p];
                        
-                               rCurve = new Rarefact(order, rDisplays);
-                               rCurve->getCurve(freq, nIters);
-                               delete rCurve;
+                       if (m->control_pressed) { return 0; }
                        
-                               mothurOut(order->getLabel()); mothurOutEndLine();
-                               processedLabels.insert(order->getLabel());
-                               userLabels.erase(order->getLabel());
-                               userLines.erase(count);
+                       if (inputFileNames.size() > 1) {
+                               m->mothurOutEndLine(); m->mothurOut("Processing group " + groups[p]); m->mothurOutEndLine(); m->mothurOutEndLine();
                        }
+                       int i;
+                       validCalculator = new ValidCalculators();
+                       
+                       
+                       for (i=0; i<Estimators.size(); i++) {
+                               if (validCalculator->isValidCalculator("rarefaction", Estimators[i]) == true) { 
+                                       if (Estimators[i] == "sobs") { 
+                                               rDisplays.push_back(new RareDisplay(new Sobs(), new ThreeColumnFile(fileNameRoot+"rarefaction")));
+                                               outputNames.push_back(fileNameRoot+"rarefaction");
+                                       }else if (Estimators[i] == "chao") { 
+                                               rDisplays.push_back(new RareDisplay(new Chao1(), new ThreeColumnFile(fileNameRoot+"r_chao")));
+                                               outputNames.push_back(fileNameRoot+"r_chao");
+                                       }else if (Estimators[i] == "ace") { 
+                                               if(abund < 5)
+                                                       abund = 10;
+                                               rDisplays.push_back(new RareDisplay(new Ace(abund), new ThreeColumnFile(fileNameRoot+"r_ace")));
+                                               outputNames.push_back(fileNameRoot+"r_ace");
+                                       }else if (Estimators[i] == "jack") { 
+                                               rDisplays.push_back(new RareDisplay(new Jackknife(), new ThreeColumnFile(fileNameRoot+"r_jack")));
+                                               outputNames.push_back(fileNameRoot+"r_jack");
+                                       }else if (Estimators[i] == "shannon") { 
+                                               rDisplays.push_back(new RareDisplay(new Shannon(), new ThreeColumnFile(fileNameRoot+"r_shannon")));
+                                               outputNames.push_back(fileNameRoot+"r_shannon");
+                                       }else if (Estimators[i] == "npshannon") { 
+                                               rDisplays.push_back(new RareDisplay(new NPShannon(), new ThreeColumnFile(fileNameRoot+"r_npshannon")));
+                                               outputNames.push_back(fileNameRoot+"r_npshannon");
+                                       }else if (Estimators[i] == "simpson") { 
+                                               rDisplays.push_back(new RareDisplay(new Simpson(), new ThreeColumnFile(fileNameRoot+"r_simpson")));
+                                               outputNames.push_back(fileNameRoot+"r_simpson");
+                                       }else if (Estimators[i] == "bootstrap") { 
+                                               rDisplays.push_back(new RareDisplay(new Bootstrap(), new ThreeColumnFile(fileNameRoot+"r_bootstrap")));
+                                               outputNames.push_back(fileNameRoot+"r_bootstrap");
+                                       }else if (Estimators[i] == "coverage") { 
+                                               rDisplays.push_back(new RareDisplay(new Coverage(), new ThreeColumnFile(fileNameRoot+"r_coverage")));
+                                               outputNames.push_back(fileNameRoot+"r_coverage");
+                                       }else if (Estimators[i] == "nseqs") { 
+                                               rDisplays.push_back(new RareDisplay(new NSeqs(), new ThreeColumnFile(fileNameRoot+"r_nseqs")));
+                                               outputNames.push_back(fileNameRoot+"r_nseqs");
+                                       }
+                               }
+                       }
+                       
                        
-                       if ((anyLabelsToProcess(order->getLabel(), userLabels, "") == true) && (processedLabels.count(lastLabel) != 1)) {
+                       //if the users entered no valid calculators don't execute command
+                       if (rDisplays.size() == 0) { for(int i=0;i<rDisplays.size();i++){       delete rDisplays[i];    } delete validCalculator; return 0; }
+                       
+                       read = new ReadOTUFile(globaldata->inputFileName);      
+                       read->read(&*globaldata); 
+                       
+                       order = globaldata->gorder;
+                       string lastLabel = order->getLabel();
+                       input = globaldata->ginput;
+                       
+                       //if the users enters label "0.06" and there is no "0.06" in their file use the next lowest label.
+                       set<string> processedLabels;
+                       set<string> userLabels = labels;
+                       
+                       if (m->control_pressed) { for(int i=0;i<rDisplays.size();i++){  delete rDisplays[i];    } delete validCalculator; delete read; delete input; globaldata->ginput = NULL; delete order; globaldata->gorder = NULL; for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); } return 0; }
+                       
+                       //as long as you are not at the end of the file or done wih the lines you want
+                       while((order != NULL) && ((allLines == 1) || (userLabels.size() != 0))) {
+                               
+                               if (m->control_pressed) { for(int i=0;i<rDisplays.size();i++){  delete rDisplays[i];    } delete validCalculator; delete read; delete input; globaldata->ginput = NULL; delete order; globaldata->gorder = NULL; for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); } return 0; }
+
+                               
+                               if(allLines == 1 || labels.count(order->getLabel()) == 1){
+                                       
+                                       m->mothurOut(order->getLabel()); m->mothurOutEndLine();
+                                       rCurve = new Rarefact(order, rDisplays);
+                                       rCurve->getCurve(freq, nIters);
+                                       delete rCurve;
+                                       
+                                       processedLabels.insert(order->getLabel());
+                                       userLabels.erase(order->getLabel());
+                               }
+                               
+                               if ((anyLabelsToProcess(order->getLabel(), userLabels, "") == true) && (processedLabels.count(lastLabel) != 1)) {
+                                       string saveLabel = order->getLabel();
+                                       
+                                       delete order;
+                                       order = (input->getOrderVector(lastLabel));
+                                       
+                                       m->mothurOut(order->getLabel()); m->mothurOutEndLine();
+                                       rCurve = new Rarefact(order, rDisplays);
+                                       rCurve->getCurve(freq, nIters);
+                                       delete rCurve;
+                                       
+                                       processedLabels.insert(order->getLabel());
+                                       userLabels.erase(order->getLabel());
+                                       
+                                       //restore real lastlabel to save below
+                                       order->setLabel(saveLabel);
+                               }
+                               
+                               lastLabel = order->getLabel();          
+                               
                                delete order;
+                               order = (input->getOrderVector());
+                       }
+                       
+                       if (m->control_pressed) { for(int i=0;i<rDisplays.size();i++){  delete rDisplays[i];    } delete validCalculator; delete read; delete input; globaldata->ginput = NULL; for (int i = 0; i < outputNames.size(); i++) {  remove(outputNames[i].c_str()); }  return 0; }
+
+                       //output error messages about any remaining user labels
+                       set<string>::iterator it;
+                       bool needToRun = false;
+                       for (it = userLabels.begin(); it != userLabels.end(); it++) {  
+                               m->mothurOut("Your file does not include the label " + *it);
+                               if (processedLabels.count(lastLabel) != 1) {
+                                       m->mothurOut(". I will use " + lastLabel + "."); m->mothurOutEndLine();
+                                       needToRun = true;
+                               }else {
+                                       m->mothurOut(". Please refer to " + lastLabel + "."); m->mothurOutEndLine();
+                               }
+                       }
+                       
+                       if (m->control_pressed) { for(int i=0;i<rDisplays.size();i++){  delete rDisplays[i];    } delete validCalculator; delete read; delete input; globaldata->ginput = NULL;  for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); } return 0; }
+
+                       //run last label if you need to
+                       if (needToRun == true)  {
+                               if (order != NULL) {    delete order;   }
                                order = (input->getOrderVector(lastLabel));
                                
+                               m->mothurOut(order->getLabel()); m->mothurOutEndLine();
                                rCurve = new Rarefact(order, rDisplays);
                                rCurve->getCurve(freq, nIters);
                                delete rCurve;
-                       
-                               mothurOut(order->getLabel()); mothurOutEndLine();
-                               processedLabels.insert(order->getLabel());
-                               userLabels.erase(order->getLabel());
+                               
+                               delete order;
                        }
                        
-                       lastLabel = order->getLabel();          
                        
-                       delete order;
-                       order = (input->getOrderVector());
-                       count++;
+                       for(int i=0;i<rDisplays.size();i++){    delete rDisplays[i];    }       
+                       rDisplays.clear();
+                       globaldata->gorder = NULL;
+                       delete input;  globaldata->ginput = NULL;
+                       delete read;
+                       delete validCalculator;
+                       
                }
                
-               //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);
-                       if (processedLabels.count(lastLabel) != 1) {
-                               mothurOut(". I will use " + lastLabel + "."); mothurOutEndLine();
-                               needToRun = true;
-                       }else {
-                               mothurOut(". Please refer to " + lastLabel + "."); mothurOutEndLine();
-                       }
-               }
+               if (m->control_pressed) {  for (int i = 0; i < outputNames.size(); i++) {       remove(outputNames[i].c_str()); } 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();
+
+               return 0;
+       }
+       catch(exception& e) {
+               m->errorOut(e, "RareFactCommand", "execute");
+               exit(1);
+       }
+}
+//**********************************************************************************************************************
+vector<string> RareFactCommand::parseSharedFile(string filename) {
+       try {
+               vector<string> filenames;
+               
+               map<string, ofstream*> filehandles;
+               map<string, ofstream*>::iterator it3;
                
-               //run last line if you need to
-               if (needToRun == true)  {
-                       delete order;
-                       order = (input->getOrderVector(lastLabel));
                                
-                       rCurve = new Rarefact(order, rDisplays);
-                       rCurve->getCurve(freq, nIters);
-                       delete rCurve;
+               //read first line
+               read = new ReadOTUFile(filename);       
+               read->read(&*globaldata); 
                        
-                       mothurOut(order->getLabel()); mothurOutEndLine();
-                       delete order;
+               input = globaldata->ginput;
+               vector<SharedRAbundVector*> lookup = input->getSharedRAbundVectors();
+               
+               string sharedFileRoot = getRootName(filename);
+               
+               //clears file before we start to write to it below
+               for (int i=0; i<lookup.size(); i++) {
+                       remove((sharedFileRoot + lookup[i]->getGroup() + ".rabund").c_str());
+                       filenames.push_back((sharedFileRoot + lookup[i]->getGroup() + ".rabund"));
                }
                
+               ofstream* temp;
+               for (int i=0; i<lookup.size(); i++) {
+                       temp = new ofstream;
+                       filehandles[lookup[i]->getGroup()] = temp;
+                       groups.push_back(lookup[i]->getGroup());
+               }
 
-               for(int i=0;i<rDisplays.size();i++){    delete rDisplays[i];    }       
-               return 0;
+               while(lookup[0] != NULL) {
+               
+                       for (int i = 0; i < lookup.size(); i++) {
+                               RAbundVector rav = lookup[i]->getRAbundVector();
+                               openOutputFileAppend(sharedFileRoot + lookup[i]->getGroup() + ".rabund", *(filehandles[lookup[i]->getGroup()]));
+                               rav.print(*(filehandles[lookup[i]->getGroup()]));
+                               (*(filehandles[lookup[i]->getGroup()])).close();
+                       }
+               
+                       for (int i = 0; i < lookup.size(); i++) {  delete lookup[i];  } 
+                       lookup = input->getSharedRAbundVectors();
+               }
+               
+               //free memory
+               for (it3 = filehandles.begin(); it3 != filehandles.end(); it3++) {
+                       delete it3->second;
+               }
+               delete read;
+               delete input;
+               globaldata->ginput = NULL;
+
+               return filenames;
        }
        catch(exception& e) {
-               errorOut(e, "RareFactCommand", "execute");
+               m->errorOut(e, "RareFactCommand", "parseSharedFile");
                exit(1);
        }
 }
-
 //**********************************************************************************************************************
+
+
+