]> git.donarmstrong.com Git - mothur.git/blobdiff - chimeracheckcommand.cpp
working on chimera change to add trim feature, fixed bug in print of distance file...
[mothur.git] / chimeracheckcommand.cpp
index de6f438d81f7fcd62e111d97a02e87d63aad1e66..3104fac58c9b6c4ff9bc1dc4870fcc77c0541853 100644 (file)
@@ -9,8 +9,53 @@
 
 #include "chimeracheckcommand.h"
 
+//**********************************************************************************************************************
+vector<string> ChimeraCheckCommand::getValidParameters(){      
+       try {
+               string AlignArray[] =  {"fasta","processors","increment","template","ksize","svg", "name","outputdir","inputdir" };
+               vector<string> myArray (AlignArray, AlignArray+(sizeof(AlignArray)/sizeof(string)));
+               return myArray;
+       }
+       catch(exception& e) {
+               m->errorOut(e, "ChimeraCheckCommand", "getValidParameters");
+               exit(1);
+       }
+}
+//**********************************************************************************************************************
+vector<string> ChimeraCheckCommand::getRequiredParameters(){   
+       try {
+               string AlignArray[] =  {"template","fasta"};
+               vector<string> myArray (AlignArray, AlignArray+(sizeof(AlignArray)/sizeof(string)));
+               return myArray;
+       }
+       catch(exception& e) {
+               m->errorOut(e, "ChimeraCheckCommand", "getRequiredParameters");
+               exit(1);
+       }
+}
+//**********************************************************************************************************************
+vector<string> ChimeraCheckCommand::getRequiredFiles(){        
+       try {
+               vector<string> myArray;
+               return myArray;
+       }
+       catch(exception& e) {
+               m->errorOut(e, "ChimeraCheckCommand", "getRequiredFiles");
+               exit(1);
+       }
+}
+//**********************************************************************************************************************
+ChimeraCheckCommand::ChimeraCheckCommand(){    
+       try {
+               vector<string> tempOutNames;
+               outputTypes["chimera"] = tempOutNames;
+       }
+       catch(exception& e) {
+               m->errorOut(e, "ChimeraCheckCommand", "ChimeraCheckCommand");
+               exit(1);
+       }
+}
 //***************************************************************************************************************
-
 ChimeraCheckCommand::ChimeraCheckCommand(string option)  {
        try {
                abort = false;
@@ -34,6 +79,9 @@ ChimeraCheckCommand::ChimeraCheckCommand(string option)  {
                                if (validParameter.isValidParameter(it->first, myArray, it->second) != true) {  abort = true;  }
                        }
                        
+                       vector<string> tempOutNames;
+                       outputTypes["chimera"] = 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 = "";          }
@@ -41,7 +89,7 @@ ChimeraCheckCommand::ChimeraCheckCommand(string option)  {
                                it = parameters.find("template");
                                //user has given a template file
                                if(it != parameters.end()){ 
-                                       string path = hasPath(it->second);
+                                       string path = m->hasPath(it->second);
                                        //if the user has not given a path then, add inputdir. else leave path alone.
                                        if (path == "") {       parameters["template"] = inputDir + it->second;         }
                                }
@@ -51,12 +99,12 @@ ChimeraCheckCommand::ChimeraCheckCommand(string option)  {
                        fastafile = validParameter.validFile(parameters, "fasta", false);
                        if (fastafile == "not found") { fastafile = ""; m->mothurOut("fasta is a required parameter for the chimera.check command."); m->mothurOutEndLine(); abort = true;  }
                        else { 
-                               splitAtDash(fastafile, fastaFileNames);
+                               m->splitAtDash(fastafile, fastaFileNames);
                                
                                //go through files and make sure they are good, if not, then disregard them
                                for (int i = 0; i < fastaFileNames.size(); i++) {
                                        if (inputDir != "") {
-                                               string path = hasPath(fastaFileNames[i]);
+                                               string path = m->hasPath(fastaFileNames[i]);
                                                //if the user has not given a path then, add inputdir. else leave path alone.
                                                if (path == "") {       fastaFileNames[i] = inputDir + fastaFileNames[i];               }
                                        }
@@ -64,17 +112,32 @@ ChimeraCheckCommand::ChimeraCheckCommand(string option)  {
                                        int ableToOpen;
                                        ifstream in;
                                        
-                                       ableToOpen = openInputFile(fastaFileNames[i], in, "noerror");
+                                       ableToOpen = m->openInputFile(fastaFileNames[i], in, "noerror");
                                
                                        //if you can't open it, try default location
                                        if (ableToOpen == 1) {
                                                if (m->getDefaultPath() != "") { //default path is set
-                                                       string tryPath = m->getDefaultPath() + getSimpleName(fastaFileNames[i]);
+                                                       string tryPath = m->getDefaultPath() + m->getSimpleName(fastaFileNames[i]);
                                                        m->mothurOut("Unable to open " + fastaFileNames[i] + ". Trying default " + tryPath); m->mothurOutEndLine();
-                                                       ableToOpen = openInputFile(tryPath, in, "noerror");
+                                                       ifstream in2;
+                                                       ableToOpen = m->openInputFile(tryPath, in2, "noerror");
+                                                       in2.close();
                                                        fastaFileNames[i] = tryPath;
                                                }
                                        }
+                                       
+                                       //if you can't open it, try default location
+                                       if (ableToOpen == 1) {
+                                               if (m->getOutputDir() != "") { //default path is set
+                                                       string tryPath = m->getOutputDir() + m->getSimpleName(fastaFileNames[i]);
+                                                       m->mothurOut("Unable to open " + fastaFileNames[i] + ". Trying output directory " + tryPath); m->mothurOutEndLine();
+                                                       ifstream in2;
+                                                       ableToOpen = m->openInputFile(tryPath, in2, "noerror");
+                                                       in2.close();
+                                                       fastaFileNames[i] = tryPath;
+                                               }
+                                       }
+                                       
                                        in.close();
                                        
                                        if (ableToOpen == 1) { 
@@ -99,12 +162,12 @@ ChimeraCheckCommand::ChimeraCheckCommand(string option)  {
                        namefile = validParameter.validFile(parameters, "name", false);
                        if (namefile == "not found") { namefile = ""; }
                        else { 
-                               splitAtDash(namefile, nameFileNames);
+                               m->splitAtDash(namefile, nameFileNames);
                                
                                //go through files and make sure they are good, if not, then disregard them
                                for (int i = 0; i < nameFileNames.size(); i++) {
                                        if (inputDir != "") {
-                                               string path = hasPath(nameFileNames[i]);
+                                               string path = m->hasPath(nameFileNames[i]);
                                                //if the user has not given a path then, add inputdir. else leave path alone.
                                                if (path == "") {       nameFileNames[i] = inputDir + nameFileNames[i];         }
                                        }
@@ -112,17 +175,32 @@ ChimeraCheckCommand::ChimeraCheckCommand(string option)  {
                                        int ableToOpen;
                                        ifstream in;
                                        
-                                       ableToOpen = openInputFile(nameFileNames[i], in, "noerror");
+                                       ableToOpen = m->openInputFile(nameFileNames[i], in, "noerror");
                                
                                        //if you can't open it, try default location
                                        if (ableToOpen == 1) {
                                                if (m->getDefaultPath() != "") { //default path is set
-                                                       string tryPath = m->getDefaultPath() + getSimpleName(nameFileNames[i]);
+                                                       string tryPath = m->getDefaultPath() + m->getSimpleName(nameFileNames[i]);
                                                        m->mothurOut("Unable to open " + nameFileNames[i] + ". Trying default " + tryPath); m->mothurOutEndLine();
-                                                       ableToOpen = openInputFile(tryPath, in, "noerror");
+                                                       ifstream in2;
+                                                       ableToOpen = m->openInputFile(tryPath, in2, "noerror");
+                                                       in2.close();
                                                        nameFileNames[i] = tryPath;
                                                }
                                        }
+                                       
+                                       //if you can't open it, try default location
+                                       if (ableToOpen == 1) {
+                                               if (m->getOutputDir() != "") { //default path is set
+                                                       string tryPath = m->getOutputDir() + m->getSimpleName(nameFileNames[i]);
+                                                       m->mothurOut("Unable to open " + nameFileNames[i] + ". Trying output directory " + tryPath); m->mothurOutEndLine();
+                                                       ifstream in2;
+                                                       ableToOpen = m->openInputFile(tryPath, in2, "noerror");
+                                                       in2.close();
+                                                       nameFileNames[i] = tryPath;
+                                               }
+                                       }
+                                       
                                        in.close();
                                        
                                        if (ableToOpen == 1) { 
@@ -150,7 +228,7 @@ ChimeraCheckCommand::ChimeraCheckCommand(string option)  {
                        convert(temp, ksize);
                        
                        temp = validParameter.validFile(parameters, "svg", false);                              if (temp == "not found") { temp = "F"; }
-                       svg = isTrue(temp);
+                       svg = m->isTrue(temp);
                        if (nameFileNames.size() != 0) { svg = true; }
                        
                        temp = validParameter.validFile(parameters, "increment", false);                if (temp == "not found") { temp = "10"; }
@@ -217,12 +295,13 @@ int ChimeraCheckCommand::execute(){
 
                        if (m->control_pressed) { delete chimera;       return 0;       }
                        
-                       string outputFileName = outputDir + getRootName(getSimpleName(fastaFileNames[i]))  + "chimeracheck.chimeras";
-                       outputNames.push_back(outputFileName);
+                       if (outputDir == "") { outputDir = m->hasPath(fastaFileNames[i]);  }//if user entered a file with a path then preserve it
+                       string outputFileName = outputDir + m->getRootName(m->getSimpleName(fastaFileNames[i]))  + "chimeracheck.chimeras";
+                       outputNames.push_back(outputFileName); outputTypes["chimera"].push_back(outputFileName);
                        
                #ifdef USE_MPI
                
-                               int pid, end, numSeqsPerProcessor; 
+                               int pid, numSeqsPerProcessor; 
                                int tag = 2001;
                                vector<unsigned long int> MPIPos;
                                
@@ -245,10 +324,10 @@ int ChimeraCheckCommand::execute(){
                                MPI_File_open(MPI_COMM_WORLD, inFileName, inMode, MPI_INFO_NULL, &inMPI);  //comm, filename, mode, info, filepointer
                                MPI_File_open(MPI_COMM_WORLD, outFilename, outMode, MPI_INFO_NULL, &outMPI);
                                
-                               if (m->control_pressed) {  MPI_File_close(&inMPI);  MPI_File_close(&outMPI);  for (int j = 0; j < outputNames.size(); j++) {    remove(outputNames[j].c_str()); } delete chimera; return 0;  }
+                               if (m->control_pressed) {  MPI_File_close(&inMPI);  MPI_File_close(&outMPI);  for (int j = 0; j < outputNames.size(); j++) {    remove(outputNames[j].c_str()); } outputTypes.clear(); delete chimera; return 0;  }
                                
                                if (pid == 0) { //you are the root process 
-                                       MPIPos = setFilePosFasta(fastaFileNames[i], numSeqs); //fills MPIPos, returns numSeqs
+                                       MPIPos = m->setFilePosFasta(fastaFileNames[i], numSeqs); //fills MPIPos, returns numSeqs
                                        
                                        //send file positions to all processes
                                        for(int j = 1; j < processors; j++) { 
@@ -265,12 +344,12 @@ int ChimeraCheckCommand::execute(){
                                        //align your part
                                        driverMPI(startIndex, numSeqsPerProcessor, inMPI, outMPI, MPIPos);
                                        
-                                       if (m->control_pressed) {  MPI_File_close(&inMPI);  MPI_File_close(&outMPI);  for (int j = 0; j < outputNames.size(); j++) {    remove(outputNames[j].c_str()); }   delete chimera; return 0;  }
+                                       if (m->control_pressed) {  MPI_File_close(&inMPI);  MPI_File_close(&outMPI);  for (int j = 0; j < outputNames.size(); j++) {    remove(outputNames[j].c_str()); }   outputTypes.clear(); delete chimera; return 0;  }
                                        
                                        //wait on chidren
                                        for(int j = 1; j < processors; j++) { 
-                                               char buf[4];
-                                               MPI_Recv(buf, 4, MPI_CHAR, j, tag, MPI_COMM_WORLD, &status); 
+                                               char buf[5];
+                                               MPI_Recv(buf, 5, MPI_CHAR, j, tag, MPI_COMM_WORLD, &status); 
                                        }
                                }else{ //you are a child process
                                        MPI_Recv(&numSeqs, 1, MPI_INT, 0, tag, MPI_COMM_WORLD, &status);
@@ -285,12 +364,12 @@ int ChimeraCheckCommand::execute(){
                                        //align your part
                                        driverMPI(startIndex, numSeqsPerProcessor, inMPI, outMPI, MPIPos);
                                        
-                                       if (m->control_pressed) {  MPI_File_close(&inMPI);  MPI_File_close(&outMPI);   for (int j = 0; j < outputNames.size(); j++) {   remove(outputNames[j].c_str()); }  delete chimera; return 0;  }
+                                       if (m->control_pressed) {  MPI_File_close(&inMPI);  MPI_File_close(&outMPI);   for (int j = 0; j < outputNames.size(); j++) {   remove(outputNames[j].c_str()); }  outputTypes.clear(); delete chimera; return 0;  }
                                        
                                        //tell parent you are done.
-                                       char buf[4];
+                                       char buf[5];
                                        strcpy(buf, "done"); 
-                                       MPI_Send(buf, 4, MPI_CHAR, 0, tag, MPI_COMM_WORLD);
+                                       MPI_Send(buf, 5, MPI_CHAR, 0, tag, MPI_COMM_WORLD);
                                }
                                
                                //close files 
@@ -299,7 +378,7 @@ int ChimeraCheckCommand::execute(){
                                MPI_Barrier(MPI_COMM_WORLD); //make everyone wait - just in case
                #else
                        
-                       vector<unsigned long int> positions = divideFile(fastaFileNames[i], processors);
+                       vector<unsigned long int> positions = m->divideFile(fastaFileNames[i], processors);
                                
                        for (int s = 0; s < (positions.size()-1); s++) {
                                lines.push_back(new linePair(positions[s], positions[(s+1)]));
@@ -310,7 +389,7 @@ int ChimeraCheckCommand::execute(){
                                if(processors == 1){
                                        numSeqs = driver(lines[0], outputFileName, fastaFileNames[i]);
                                        
-                                       if (m->control_pressed) { for (int j = 0; j < outputNames.size(); j++) {        remove(outputNames[j].c_str()); } for (int j = 0; j < lines.size(); j++) {  delete lines[j];  }  lines.clear(); delete chimera; return 0; }
+                                       if (m->control_pressed) { for (int j = 0; j < outputNames.size(); j++) {        remove(outputNames[j].c_str()); } for (int j = 0; j < lines.size(); j++) {  delete lines[j];  } outputTypes.clear();  lines.clear(); delete chimera; return 0; }
                                                                        
                                }else{
                                        processIDS.resize(0);
@@ -321,12 +400,12 @@ int ChimeraCheckCommand::execute(){
                                                
                                        //append output files
                                        for(int j=1;j<processors;j++){
-                                               appendFiles((outputFileName + toString(processIDS[j]) + ".temp"), outputFileName);
+                                               m->appendFiles((outputFileName + toString(processIDS[j]) + ".temp"), outputFileName);
                                                remove((outputFileName + toString(processIDS[j]) + ".temp").c_str());
                                        }
                                        
                                        if (m->control_pressed) { 
-                                               for (int j = 0; j < outputNames.size(); j++) {  remove(outputNames[j].c_str()); } 
+                                               for (int j = 0; j < outputNames.size(); j++) {  remove(outputNames[j].c_str()); } outputTypes.clear();
                                                for (int j = 0; j < lines.size(); j++) {  delete lines[j];  }  lines.clear();
                                                delete chimera;
                                                return 0;
@@ -336,7 +415,7 @@ int ChimeraCheckCommand::execute(){
                        #else
                                numSeqs = driver(lines[0], outputFileName, fastaFileNames[i]);
                                
-                               if (m->control_pressed) { for (int j = 0; j < lines.size(); j++) {  delete lines[j];  }  lines.clear(); for (int j = 0; j < outputNames.size(); j++) {  remove(outputNames[j].c_str()); } delete chimera; return 0; }
+                               if (m->control_pressed) { for (int j = 0; j < lines.size(); j++) {  delete lines[j];  }  lines.clear(); for (int j = 0; j < outputNames.size(); j++) {  remove(outputNames[j].c_str()); } outputTypes.clear(); delete chimera; return 0; }
                        #endif
                #endif          
                        delete chimera;
@@ -365,12 +444,12 @@ int ChimeraCheckCommand::execute(){
 int ChimeraCheckCommand::driver(linePair* filePos, string outputFName, string filename){
        try {
                ofstream out;
-               openOutputFile(outputFName, out);
+               m->openOutputFile(outputFName, out);
                
                ofstream out2;
                
                ifstream inFASTA;
-               openInputFile(filename, inFASTA);
+               m->openInputFile(filename, inFASTA);
 
                inFASTA.seekg(filePos->start);
 
@@ -381,7 +460,7 @@ int ChimeraCheckCommand::driver(linePair* filePos, string outputFName, string fi
 
                        if (m->control_pressed) {       return 1;       }
                
-                       Sequence* candidateSeq = new Sequence(inFASTA);  gobble(inFASTA);
+                       Sequence* candidateSeq = new Sequence(inFASTA);  m->gobble(inFASTA);
                                
                        if (candidateSeq->getName() != "") { //incase there is a commented sequence at the end of a file
                                //find chimeras
@@ -394,8 +473,12 @@ int ChimeraCheckCommand::driver(linePair* filePos, string outputFName, string fi
                        }
                        delete candidateSeq;
                        
-                       unsigned long int pos = inFASTA.tellg();
-                       if ((pos == -1) || (pos >= filePos->end)) { break; }
+                       #if defined (__APPLE__) || (__MACH__) || (linux) || (__linux)
+                               unsigned long int pos = inFASTA.tellg();
+                               if ((pos == -1) || (pos >= filePos->end)) { break; }
+                       #else
+                               if (inFASTA.eof()) { break; }
+                       #endif
                        
                        //report progress
                        if((count) % 100 == 0){ m->mothurOut("Processing sequence: " + toString(count)); m->mothurOutEndLine();         }
@@ -437,7 +520,7 @@ int ChimeraCheckCommand::driverMPI(int start, int num, MPI_File& inMPI, MPI_File
                        istringstream iss (tempBuf,istringstream::in);
                        delete buf4;
 
-                       Sequence* candidateSeq = new Sequence(iss);  gobble(iss);
+                       Sequence* candidateSeq = new Sequence(iss);  m->gobble(iss);
                                
                        if (candidateSeq->getName() != "") { //incase there is a commented sequence at the end of a file
                                //find chimeras
@@ -484,12 +567,16 @@ int ChimeraCheckCommand::createProcesses(string outputFileName, string filename)
                                //pass numSeqs to parent
                                ofstream out;
                                string tempFile = outputFileName + toString(getpid()) + ".num.temp";
-                               openOutputFile(tempFile, out);
+                               m->openOutputFile(tempFile, out);
                                out << num << endl;
                                out.close();
                                
                                exit(0);
-                       }else { m->mothurOut("unable to spawn the necessary processes."); m->mothurOutEndLine(); exit(0); }
+                       }else { 
+                               m->mothurOut("[ERROR]: unable to spawn the necessary processes."); m->mothurOutEndLine(); 
+                               for (int i = 0; i < processIDS.size(); i++) { kill (processIDS[i], SIGINT); }
+                               exit(0);
+                       }
                }
                
                //force parent to wait until all the processes are done
@@ -501,7 +588,7 @@ int ChimeraCheckCommand::createProcesses(string outputFileName, string filename)
                for (int i = 0; i < processIDS.size(); i++) {
                        ifstream in;
                        string tempFile =  outputFileName + toString(processIDS[i]) + ".num.temp";
-                       openInputFile(tempFile, in);
+                       m->openInputFile(tempFile, in);
                        if (!in.eof()) { int tempNum = 0; in >> tempNum; num += tempNum; }
                        in.close(); remove(tempFile.c_str());
                }