worked on hcluster. made .single command run using a sharedfile. and various other...

[mothur.git] / readdistcommand.cpp

diff --git a/readdistcommand.cpp b/readdistcommand.cpp
index 719740cf7589e121029197bb46e30bbdd62e6bda..7e81d7f1a3efe2535f1d9d11715bb2c517a508ec 100644 (file)
--- a/readdistcommand.cpp
+++ b/readdistcommand.cpp
@@ -72,7 +72,7 @@ ReadDistCommand::ReadDistCommand(string option){
                        if (columnfile != "") {
                                if (namefile == "") {  cout << "You need to provide a namefile if you are going to use the column format." << endl; abort = true; }
                        }
-               
+                       
                        //check for optional parameter and set defaults
                        // ...at some point should added some additional type checking...
                        //get user cutoff and precision or use defaults
@@ -93,6 +93,7 @@ ReadDistCommand::ReadDistCommand(string option){
                                else if (format == "matrix") { 
                                        groupMap = new GroupMap(groupfile);
                                        groupMap->readMap();
+       
                                        if (globaldata->gGroupmap != NULL) { delete globaldata->gGroupmap;  }
                                        globaldata->gGroupmap = groupMap;
                                }
@@ -163,6 +164,34 @@ int ReadDistCommand::execute(){
                        openInputFile(distFileName, in);
                        matrix = new FullMatrix(in); //reads the matrix file
                        in.close();
+                       
+                       //if files don't match...
+                       if (matrix->getNumSeqs() < groupMap->getNumSeqs()) {  
+                               mothurOut("Your distance file contains " + toString(matrix->getNumSeqs()) + " sequences, and your group file contains " + toString(groupMap->getNumSeqs()) + " sequences.");  mothurOutEndLine();                               
+                               //create new group file
+                               string newGroupFile = getRootName(groupfile) + "editted.groups";
+                               ofstream outGroups;
+                               openOutputFile(newGroupFile, outGroups);
+                               
+                               for (int i = 0; i < matrix->getNumSeqs(); i++) {
+                                       Names temp = matrix->getRowInfo(i);
+                                       outGroups << temp.seqName << '\t' << temp.groupName << endl;
+                               }
+                               outGroups.close();
+                               
+                               mothurOut(newGroupFile + " is a new group file containing only the sequence that are in your distance file. I will read this file instead."); mothurOutEndLine();
+                               
+                               //read new groupfile
+                               delete groupMap; groupMap = NULL;
+                               groupfile = newGroupFile;
+                               globaldata->setGroupFile(groupfile); 
+                               
+                               groupMap = new GroupMap(groupfile);
+                               groupMap->readMap();
+                               
+                               globaldata->gGroupmap = groupMap;
+                       }
+                       
                        //memory leak prevention
                        if (globaldata->gMatrix != NULL) { delete globaldata->gMatrix;  }
                        globaldata->gMatrix = matrix; //save matrix for coverage commands
@@ -177,6 +206,7 @@ int ReadDistCommand::execute(){
                        if (globaldata->gSparseMatrix != NULL) { delete globaldata->gSparseMatrix;  }
                        globaldata->gSparseMatrix = read->getMatrix();
                        numDists = globaldata->gSparseMatrix->getNNodes();
+       cout << "matrix contains " << numDists << " distances." << endl;
                        
       int lines = cutoff / (1.0/precision);
       vector<float> dist_cutoff(lines+1,0);