]> git.donarmstrong.com Git - mothur.git/blobdiff - filterseqscommand.cpp
1.12.1
[mothur.git] / filterseqscommand.cpp
index a6290a6afe55a28001e5e229e518a7c03c6a1c28..7ab00a7822f7a7fc7ddd1ef0de00d2d24b8c51de 100644 (file)
@@ -28,7 +28,7 @@ FilterSeqsCommand::FilterSeqsCommand(string option)  {
                        OptionParser parser(option);
                        map<string,string> parameters = parser.getParameters();
                        
-                       ValidParameters validParameter;
+                       ValidParameters validParameter("filter.seqs");
                        map<string,string>::iterator it;
                        
                        //check to make sure all parameters are valid for command
@@ -72,11 +72,22 @@ FilterSeqsCommand::FilterSeqsCommand(string option)  {
                                                if (path == "") {       fastafileNames[i] = inputDir + fastafileNames[i];               }
                                        }
 
-                                       int ableToOpen;
                                        ifstream in;
-                                       ableToOpen = openInputFile(fastafileNames[i], in);
+                                       int ableToOpen = 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]);
+                                                       m->mothurOut("Unable to open " + fastafileNames[i] + ". Trying default " + tryPath); m->mothurOutEndLine();
+                                                       ableToOpen = openInputFile(tryPath, in, "noerror");
+                                                       fastafileNames[i] = tryPath;
+                                               }
+                                       }
+                                       in.close();
+                                       
                                        if (ableToOpen == 1) { 
-                                               m->mothurOut(fastafileNames[i] + " will be disregarded."); m->mothurOutEndLine(); 
+                                               m->mothurOut("Unable to open " + fastafileNames[i] + ". It will be disregarded."); m->mothurOutEndLine();
                                                //erase from file list
                                                fastafileNames.erase(fastafileNames.begin()+i);
                                                i--;
@@ -102,6 +113,9 @@ FilterSeqsCommand::FilterSeqsCommand(string option)  {
                        // ...at some point should added some additional type checking...
                        
                        string temp;
+                       hard = validParameter.validFile(parameters, "hard", true);                              if (hard == "not found") { hard = ""; }
+                       else if (hard == "not open") { hard = ""; abort = true; }       
+
                        temp = validParameter.validFile(parameters, "trump", false);                    if (temp == "not found") { temp = "*"; }
                        trump = temp[0];
                        
@@ -111,13 +125,13 @@ FilterSeqsCommand::FilterSeqsCommand(string option)  {
                        temp = validParameter.validFile(parameters, "processors", false);       if (temp == "not found"){       temp = "1";                             }
                        convert(temp, processors); 
                        
-                       hard = validParameter.validFile(parameters, "hard", true);                              if (hard == "not found") { hard = ""; }
-                       else if (hard == "not open") { abort = true; }  
-                       
-                       vertical = validParameter.validFile(parameters, "vertical", false);             if (vertical == "not found") { vertical = "T"; }
+                       vertical = validParameter.validFile(parameters, "vertical", false);             
+                       if (vertical == "not found") { 
+                               if ((hard == "") && (trump == '*') && (soft == 0)) { vertical = "T"; } //you have not given a hard file or set the trump char.
+                               else { vertical = "F";  }
+                       }
                        
                        numSeqs = 0;
-                       
                }
                
        }
@@ -138,11 +152,11 @@ void FilterSeqsCommand::help(){
                m->mothurOut("For example: fasta=abrecovery.fasta-amazon.fasta \n");
                m->mothurOut("The trump parameter .... The default is ...\n");
                m->mothurOut("The soft parameter .... The default is ....\n");
-               m->mothurOut("The hard parameter .... The default is ....\n");
-               m->mothurOut("The vertical parameter .... The default is T.\n");
+               m->mothurOut("The hard parameter allows you to enter a file containing the filter you want to use.\n");
+               m->mothurOut("The vertical parameter removes columns where all sequences contain a gap character. The default is T.\n");
                m->mothurOut("The filter.seqs command should be in the following format: \n");
-               m->mothurOut("filter.seqs(fasta=yourFastaFile, trump=yourTrump, soft=yourSoft, hard=yourHard, vertical=yourVertical) \n");
-               m->mothurOut("Example filter.seqs(fasta=abrecovery.fasta, trump=..., soft=..., hard=..., vertical=T).\n");
+               m->mothurOut("filter.seqs(fasta=yourFastaFile, trump=yourTrump) \n");
+               m->mothurOut("Example filter.seqs(fasta=abrecovery.fasta, trump=.).\n");
                m->mothurOut("Note: No spaces between parameter labels (i.e. fasta), '=' and parameters (i.e.yourFasta).\n\n");
                
        }
@@ -244,35 +258,26 @@ int FilterSeqsCommand::filterSequences() {
 #ifdef USE_MPI 
                                int pid, start, end, numSeqsPerProcessor, num; 
                                int tag = 2001;
-                               vector<long>MPIPos;
+                               vector<unsigned long int>MPIPos;
                                                
                                MPI_Status status; 
                                MPI_Comm_size(MPI_COMM_WORLD, &processors); //set processors to the number of mpi processes running
                                MPI_Comm_rank(MPI_COMM_WORLD, &pid); //find out who we are
-                               
+       cout << pid << "is in create filter " << endl;                  
                                MPI_File outMPI;
                                MPI_File tempMPI;
                                MPI_File inMPI;
                                int outMode=MPI_MODE_CREATE|MPI_MODE_WRONLY; 
                                int inMode=MPI_MODE_RDONLY; 
                                
-                               //char* outFilename = new char[filteredFasta.length()];
-                               //memcpy(outFilename, filteredFasta.c_str(), filteredFasta.length());
-                               
                                char outFilename[1024];
                                strcpy(outFilename, filteredFasta.c_str());
-
-                               //char* inFileName = new char[fastafileNames[s].length()];
-                               //memcpy(inFileName, fastafileNames[s].c_str(), fastafileNames[s].length());
-                               
+                       
                                char inFileName[1024];
                                strcpy(inFileName, fastafileNames[s].c_str());
                                
                                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);
-                               
-                               //delete inFileName;
-                               //delete outFilename;
 
                                if (m->control_pressed) {  MPI_File_close(&inMPI);  MPI_File_close(&outMPI);  return 0;  }
 
@@ -388,7 +393,7 @@ int FilterSeqsCommand::filterSequences() {
 }
 #ifdef USE_MPI
 /**************************************************************************************/
-int FilterSeqsCommand::driverMPIRun(int start, int num, MPI_File& inMPI, MPI_File& outMPI, vector<long>& MPIPos) {     
+int FilterSeqsCommand::driverMPIRun(int start, int num, MPI_File& inMPI, MPI_File& outMPI, vector<unsigned long int>& MPIPos) {        
        try {
                string outputString = "";
                int count = 0;
@@ -564,7 +569,7 @@ string FilterSeqsCommand::createFilter() {
 #ifdef USE_MPI 
                                int pid, numSeqsPerProcessor, num; 
                                int tag = 2001;
-                               vector<long> MPIPos;
+                               vector<unsigned long int> MPIPos;
                                
                                MPI_Status status; 
                                MPI_File inMPI; 
@@ -792,7 +797,7 @@ int FilterSeqsCommand::driverCreateFilter(Filters& F, string filename, linePair*
 }
 #ifdef USE_MPI
 /**************************************************************************************/
-int FilterSeqsCommand::MPICreateFilter(int start, int num, Filters& F, MPI_File& inMPI, vector<long>& MPIPos) {        
+int FilterSeqsCommand::MPICreateFilter(int start, int num, Filters& F, MPI_File& inMPI, vector<unsigned long int>& MPIPos) {   
        try {
                
                MPI_Status status; 
@@ -878,7 +883,7 @@ int FilterSeqsCommand::createProcessesCreateFilter(Filters& F, string filename)
 int FilterSeqsCommand::setLines(string filename) {
        try {
                
-               vector<long int> positions;
+               vector<unsigned long int> positions;
                bufferSizes.clear();
                
                ifstream inFASTA;
@@ -889,7 +894,7 @@ int FilterSeqsCommand::setLines(string filename) {
                        input = getline(inFASTA);
 
                        if (input.length() != 0) {
-                               if(input[0] == '>'){ long int pos = inFASTA.tellg(); positions.push_back(pos - input.length() - 1);     }
+                               if(input[0] == '>'){ unsigned long int pos = inFASTA.tellg(); positions.push_back(pos - input.length() - 1);    }
                        }
                }
                inFASTA.close();
@@ -897,7 +902,7 @@ int FilterSeqsCommand::setLines(string filename) {
                int numFastaSeqs = positions.size();
        
                FILE * pFile;
-               long size;
+               unsigned long int size;
                
                //get num bytes in file
                pFile = fopen (filename.c_str(),"rb");
@@ -914,12 +919,12 @@ int FilterSeqsCommand::setLines(string filename) {
                
                for (int i = 0; i < processors; i++) {
 
-                       long int startPos = positions[ i * numSeqsPerProcessor ];
+                       unsigned long int startPos = positions[ i * numSeqsPerProcessor ];
                        if(i == processors - 1){
                                numSeqsPerProcessor = numFastaSeqs - i * numSeqsPerProcessor;
                                bufferSizes.push_back(size - startPos);
                        }else{  
-                               long int myEnd = positions[ (i+1) * numSeqsPerProcessor ];
+                               unsigned long int myEnd = positions[ (i+1) * numSeqsPerProcessor ];
                                bufferSizes.push_back(myEnd-startPos);
                        }
                        lines.push_back(new linePair(startPos, numSeqsPerProcessor));