X-Git-Url: https://git.donarmstrong.com/?a=blobdiff_plain;f=chimeraperseuscommand.cpp;h=ef3058d8cda30d8781e9a0057c9e93d77feeb1b9;hb=HEAD;hp=842a65ee22fd54c730143cbed94f91688dba8006;hpb=037b7fccc64a5c7d5d5c23a949273a912160a400;p=mothur.git

diff --git a/chimeraperseuscommand.cpp b/chimeraperseuscommand.cpp
index 842a65e..ef3058d 100644
--- a/chimeraperseuscommand.cpp
+++ b/chimeraperseuscommand.cpp
@@ -469,10 +469,7 @@ ChimeraPerseusCommand::ChimeraPerseusCommand(string option)  {
 			m->mothurConvert(temp, beta);
             
 			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);
 		}
 	}
@@ -519,7 +516,7 @@ int ChimeraPerseusCommand::execute(){
             
             if (hasCount) {
                 CountTable* ct = new CountTable();
-                ct->readTable(nameFile);
+                ct->readTable(nameFile, true, false);
                 
                 if (ct->hasGroupInfo()) {
                     cparser = new SequenceCountParser(fastaFileNames[s], *ct);
@@ -537,7 +534,7 @@ int ChimeraPerseusCommand::execute(){
                     
                     if(processors == 1)	{	numSeqs = driverGroups(outputFileName, accnosFileName, newCountFile, 0, groups.size(), groups);
                         if (dups) {
-                            CountTable c; c.readTable(nameFile);
+                            CountTable c; c.readTable(nameFile, true, false);
                             if (!m->isBlank(newCountFile)) {
                                 ifstream in2;
                                 m->openInputFile(newCountFile, in2);
@@ -562,7 +559,7 @@ int ChimeraPerseusCommand::execute(){
                         numChimeras = deconvoluteResults(uniqueNames, outputFileName, accnosFileName);
                     }else {
                         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]);
@@ -1049,10 +1046,10 @@ int ChimeraPerseusCommand::driver(string chimeraFileName, vector<seqData>& seque
 			}
 	
 			//report progress
-			if((i+1) % 100 == 0){ 	m->mothurOut("Processing sequence: " + toString(i+1) + "\n");		}
+			if((i+1) % 100 == 0){ 	m->mothurOutJustToScreen("Processing sequence: " + toString(i+1) + "\n");		}
 		}
 		
-		if((numSeqs) % 100 != 0){ 	m->mothurOut("Processing sequence: " + toString(numSeqs) + "\n");		}
+		if((numSeqs) % 100 != 0){ 	m->mothurOutJustToScreen("Processing sequence: " + toString(numSeqs) + "\n");		}
 		
 		chimeraFile.close();
 		accnosFile.close();
@@ -1073,36 +1070,39 @@ int ChimeraPerseusCommand::createProcessesGroups(string outputFName, string accn
 		int num = 0;
 		
         CountTable newCount;
-        if (hasCount && dups) { newCount.readTable(name); }
+        if (hasCount && dups) { newCount.readTable(name, 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
 		while (process != processors) {
-			int pid = fork();
+			pid_t pid = fork();
 			
 			if (pid > 0) {
 				processIDS.push_back(pid);  //create map from line number to pid so you can append files in correct order later
 				process++;
 			}else if (pid == 0){
-				num = driverGroups(outputFName + toString(getpid()) + ".temp", accnos + toString(getpid()) + ".temp", accnos + ".byCount." + toString(getpid()) + ".temp", lines[process].start, lines[process].end, groups);
+				num = driverGroups(outputFName + toString(m->mothurGetpid(process)) + ".temp", accnos + toString(m->mothurGetpid(process)) + ".temp", accnos + ".byCount." + toString(m->mothurGetpid(process)) + ".temp", lines[process].start, lines[process].end, groups);
 				
 				//pass numSeqs to parent
 				ofstream out;
-				string tempFile = outputFName + toString(getpid()) + ".num.temp";
+				string tempFile = outputFName + toString(m->mothurGetpid(process)) + ".num.temp";
 				m->openOutputFile(tempFile, out);
 				out << num << endl;
 				out.close();