]> git.donarmstrong.com Git - mothur.git/blobdiff - chimerauchimecommand.cpp
fixes while testing 1.33.0
[mothur.git] / chimerauchimecommand.cpp
index cb155eb2e53c3b8ba47c4ed36aa495efaef4460a..7a48cf70d092f7d05c8b3bc3656bd66bf17c02d8 100644 (file)
@@ -564,14 +564,12 @@ ChimeraUchimeCommand::ChimeraUchimeCommand(string option)  {
             
             
                        temp = validParameter.validFile(parameters, "dereplicate", false);      
-                       if (temp == "not found") { 
-                               if (groupfile != "")    {  temp = "false";                                      }
-                               else                    {  temp = "true";       }
-                       }
+                       if (temp == "not found") { temp = "false";                      }
                        dups = m->isTrue(temp);
 
                        
                        if (hasName && (templatefile != "self")) { m->mothurOut("You have provided a namefile and the reference parameter is not set to self. I am not sure what reference you are trying to use, aborting."); m->mothurOutEndLine(); abort=true; }
+            if (hasCount && (templatefile != "self")) { m->mothurOut("You have provided a countfile and the reference parameter is not set to self. I am not sure what reference you are trying to use, aborting."); m->mothurOutEndLine(); abort=true; }
                        if (hasGroup && (templatefile != "self")) { m->mothurOut("You have provided a group file and the reference parameter is not set to self. I am not sure what reference you are trying to use, aborting."); m->mothurOutEndLine(); abort=true; }
                        
                        //look for uchime exe
@@ -677,7 +675,7 @@ int ChimeraUchimeCommand::execute(){
                 int error;
                 if (hasCount) {
                     CountTable ct;
-                    ct.readTable(nameFile);
+                    ct.readTable(nameFile, true, false);
                     for(map<string, string>::iterator it = seqs.begin(); it != seqs.end(); it++) {
                         int num = ct.getNumSeqs(it->first);
                         if (num == 0) { error = 1; }
@@ -728,7 +726,7 @@ int ChimeraUchimeCommand::execute(){
                                if(processors == 1)     {       totalSeqs = driverGroups(outputFileName, newFasta, accnosFileName, alnsFileName, newCountFile, 0, groups.size(), groups);
                     
                     if (hasCount && dups) {
-                        CountTable c; c.readTable(nameFile);
+                        CountTable c; c.readTable(nameFile, true, false);
                         if (!m->isBlank(newCountFile)) {
                             ifstream in2;
                             m->openInputFile(newCountFile, in2);
@@ -758,7 +756,7 @@ int ChimeraUchimeCommand::execute(){
                     
                     if (hasCount) {
                         set<string> doNotRemove;
-                        CountTable c; c.readTable(newCountFile);
+                        CountTable c; c.readTable(newCountFile, true, true);
                         vector<string> namesInTable = c.getNamesOfSeqs();
                         for (int i = 0; i < namesInTable.size(); i++) {
                             int temp = c.getNumSeqs(namesInTable[i]);
@@ -774,6 +772,7 @@ int ChimeraUchimeCommand::execute(){
                         }
                         out2.close();
                         c.printTable(newCountFile);
+                        outputNames.push_back(newCountFile); outputTypes["count"].push_back(newCountFile);
                     }
                 }
                 
@@ -819,6 +818,11 @@ int ChimeraUchimeCommand::execute(){
                if (itTypes != outputTypes.end()) {
                        if ((itTypes->second).size() != 0) { current = (itTypes->second)[0]; m->setAccnosFile(current); }
                }
+        
+        itTypes = outputTypes.find("count");
+               if (itTypes != outputTypes.end()) {
+                       if ((itTypes->second).size() != 0) { current = (itTypes->second)[0]; m->setCountTableFile(current); }
+               }
                
                m->mothurOutEndLine();
                m->mothurOut("Output File Names: "); m->mothurOutEndLine();
@@ -839,10 +843,6 @@ int ChimeraUchimeCommand::deconvoluteResults(map<string, string>& uniqueNames, s
                map<string, string>::iterator itUnique;
                int total = 0;
                
-               //edit accnos file
-               ifstream in2; 
-               m->openInputFile(accnosFileName, in2);
-               
                ofstream out2;
                m->openOutputFile(accnosFileName+".temp", out2);
                
@@ -852,27 +852,32 @@ int ChimeraUchimeCommand::deconvoluteResults(map<string, string>& uniqueNames, s
                set<string> chimerasInFile;
                set<string>::iterator itChimeras;
 
-               
-               while (!in2.eof()) {
-                       if (m->control_pressed) { in2.close(); out2.close(); m->mothurRemove(outputFileName); m->mothurRemove((accnosFileName+".temp")); return 0; }
-                       
-                       in2 >> name; m->gobble(in2);
-                       
-                       //find unique name
-                       itUnique = uniqueNames.find(name);
-                       
-                       if (itUnique == uniqueNames.end()) { m->mothurOut("[ERROR]: trouble parsing accnos results. Cannot find " + name + "."); m->mothurOutEndLine(); m->control_pressed = true; }
-                       else {
-                               itChimeras = chimerasInFile.find((itUnique->second));
-                               
-                               if (itChimeras == chimerasInFile.end()) {
-                                       out2 << itUnique->second << endl;
-                                       chimerasInFile.insert((itUnique->second));
-                                       total++;
-                               }
-                       }
-               }
-               in2.close();
+        if (!m->isBlank(accnosFileName)) {
+            //edit accnos file
+            ifstream in2;
+            m->openInputFile(accnosFileName, in2);
+            
+            while (!in2.eof()) {
+                if (m->control_pressed) { in2.close(); out2.close(); m->mothurRemove(outputFileName); m->mothurRemove((accnosFileName+".temp")); return 0; }
+                
+                in2 >> name; m->gobble(in2);
+                
+                //find unique name
+                itUnique = uniqueNames.find(name);
+                
+                if (itUnique == uniqueNames.end()) { m->mothurOut("[ERROR]: trouble parsing accnos results. Cannot find " + name + "."); m->mothurOutEndLine(); m->control_pressed = true; }
+                else {
+                    itChimeras = chimerasInFile.find((itUnique->second));
+                    
+                    if (itChimeras == chimerasInFile.end()) {
+                        out2 << itUnique->second << endl;
+                        chimerasInFile.insert((itUnique->second));
+                        total++;
+                    }
+                }
+            }
+            in2.close();
+        }
                out2.close();
                
                m->mothurRemove(accnosFileName);
@@ -1181,6 +1186,7 @@ int ChimeraUchimeCommand::driverGroups(string outputFName, string filename, stri
                int totalSeqs = 0;
                int numChimeras = 0;
         
+        
         ofstream outCountList;
         if (hasCount && dups) { m->openOutputFile(countlist, outCountList); }
         
@@ -1790,21 +1796,23 @@ int ChimeraUchimeCommand::createProcessesGroups(string outputFName, string filen
                int num = 0;
         
         CountTable newCount;
-        if (hasCount && dups) { newCount.readTable(nameFile); }
+        if (hasCount && dups) { newCount.readTable(nameFile, true, false); }
                
                //sanity check
                if (groups.size() < processors) { processors = groups.size(); }
                
                //divide the groups between the processors
                vector<linePair> lines;
-               int numGroupsPerProcessor = groups.size() / processors;
-               for (int i = 0; i < processors; i++) {
-                       int startIndex =  i * numGroupsPerProcessor;
-                       int endIndex = (i+1) * numGroupsPerProcessor;
-                       if(i == (processors - 1)){      endIndex = groups.size();       }
-                       lines.push_back(linePair(startIndex, endIndex));
-               }
-               
+        int remainingPairs = groups.size();
+        int startIndex = 0;
+        for (int remainingProcessors = processors; remainingProcessors > 0; remainingProcessors--) {
+            int numPairs = remainingPairs; //case for last processor
+            if (remainingProcessors != 1) { numPairs = ceil(remainingPairs / remainingProcessors); }
+            lines.push_back(linePair(startIndex, (startIndex+numPairs))); //startIndex, endIndex
+            startIndex = startIndex + numPairs;
+            remainingPairs = remainingPairs - numPairs;
+        }
+
 #if defined (__APPLE__) || (__MACH__) || (linux) || (__linux) || (__linux__) || (__unix__) || (__unix)         
                                
                //loop through and create all the processes you want