]> git.donarmstrong.com Git - mothur.git/blobdiff - sequence.cpp
added MPI code, broke up chimera.seqs into 5 separated commands, added parse.sff...
[mothur.git] / sequence.cpp
index 752e081e367fc6d572b5622e0b4eb6003a28f9bb..19adf796b5d1d26f3a681144c6c801cdb4b4f63d 100644 (file)
 /***********************************************************************/
 
 Sequence::Sequence(){
+       m = MothurOut::getInstance();
        initialize();
 }
 
 /***********************************************************************/
 
 Sequence::Sequence(string newName, string sequence) {
-
-       initialize();   
-       name = newName;
+       try {
+               m = MothurOut::getInstance();
+               initialize();   
+               name = newName;
+               
+               //setUnaligned removes any gap characters for us
+               setUnaligned(sequence);
+               setAligned(sequence);
+       }
+       catch(exception& e) {
+               m->errorOut(e, "Sequence", "Sequence");
+               exit(1);
+       }                       
+}
+//********************************************************************************************************************
+//this function will jump over commented out sequences, but if the last sequence in a file is commented out it makes a blank seq
+Sequence::Sequence(istringstream& fastaString){
+       try {
+               m = MothurOut::getInstance();
        
-       //setUnaligned removes any gap characters for us
-       setUnaligned(sequence);
-       setAligned(sequence);
+               initialize();
+               fastaString >> name;
+               name = name.substr(1);
+               string sequence;
        
+               //read comments
+               while ((name[0] == '#') && fastaString) { 
+                       while (!fastaString.eof())      {       char c = fastaString.get(); if (c == 10 || c == 13){    break;  }       } // get rest of line if there's any crap there
+                       sequence = getCommentString(fastaString);
+                       
+                       if (fastaString) {  
+                               fastaString >> name;  
+                               name = name.substr(1);  
+                       }else { 
+                               name = "";
+                               break;
+                       }
+               }
+               
+               while (!fastaString.eof())      {       char c = fastaString.get();  if (c == 10 || c == 13){   break;  }       } // get rest of line if there's any crap there
+               
+               sequence = getSequenceString(fastaString);              
+               setAligned(sequence);   
+               //setUnaligned removes any gap characters for us                                                
+               setUnaligned(sequence);         
+       }
+       catch(exception& e) {
+               m->errorOut(e, "Sequence", "Sequence");
+               exit(1);
+       }                                                               
 }
+
 //********************************************************************************************************************
 //this function will jump over commented out sequences, but if the last sequence in a file is commented out it makes a blank seq
 Sequence::Sequence(ifstream& fastaFile){
-
-       initialize();
-       fastaFile >> name;
-       name = name.substr(1);
-       string sequence;
-       
-       //read comments
-       while ((name[0] == '#') && fastaFile) { 
-           while (!fastaFile.eof())    {       char c = fastaFile.get(); if (c == 10 || c == 13){      break;  }       } // get rest of line if there's any crap there
-               sequence = getCommentString(fastaFile);
+       try {
+               m = MothurOut::getInstance();
+               initialize();
+               fastaFile >> name;
+               name = name.substr(1);
+               string sequence;
                
-               if (fastaFile) {  
-                       fastaFile >> name;  
-                       name = name.substr(1);  
-               }else { 
-                       name = "";
-                       break;
+               //read comments
+               while ((name[0] == '#') && fastaFile) { 
+                       while (!fastaFile.eof())        {       char c = fastaFile.get(); if (c == 10 || c == 13){      break;  }       } // get rest of line if there's any crap there
+                       sequence = getCommentString(fastaFile);
+                       
+                       if (fastaFile) {  
+                               fastaFile >> name;  
+                               name = name.substr(1);  
+                       }else { 
+                               name = "";
+                               break;
+                       }
                }
-       }
-       
-       //read real sequence
-       while (!fastaFile.eof())        {       char c = fastaFile.get(); if (c == 10 || c == 13){      break;  }       } // get rest of line if there's any crap there
                
-       sequence = getSequenceString(fastaFile);                
-       
-       setAligned(sequence);   
-       //setUnaligned removes any gap characters for us                                                
-       setUnaligned(sequence);                                                         
+               //read real sequence
+               while (!fastaFile.eof())        {       char c = fastaFile.get(); if (c == 10 || c == 13){      break;  }       } // get rest of line if there's any crap there
+               
+               sequence = getSequenceString(fastaFile);                
+               
+               setAligned(sequence);   
+               //setUnaligned removes any gap characters for us                                                
+               setUnaligned(sequence); 
+       }
+       catch(exception& e) {
+               m->errorOut(e, "Sequence", "Sequence");
+               exit(1);
+       }                                                       
 }
 //********************************************************************************************************************
 string Sequence::getSequenceString(ifstream& fastaFile) {
@@ -81,7 +131,7 @@ string Sequence::getSequenceString(ifstream& fastaFile) {
                return sequence;
        }
        catch(exception& e) {
-               errorOut(e, "Sequence", "getSequenceString");
+               m->errorOut(e, "Sequence", "getSequenceString");
                exit(1);
        }
 }
@@ -103,11 +153,59 @@ string Sequence::getCommentString(ifstream& fastaFile) {
                return sequence;
        }
        catch(exception& e) {
-               errorOut(e, "Sequence", "getCommentString");
+               m->errorOut(e, "Sequence", "getCommentString");
+               exit(1);
+       }
+}
+//********************************************************************************************************************
+string Sequence::getSequenceString(istringstream& fastaFile) {
+       try {
+               char letter;
+               string sequence = "";   
+               
+               while(!fastaFile.eof()){
+                       letter= fastaFile.get();
+       
+                       if(letter == '>'){
+                               fastaFile.putback(letter);
+                               break;
+                       }
+                       else if(isprint(letter)){
+                               letter = toupper(letter);
+                               if(letter == 'U'){letter = 'T';}
+                               sequence += letter;
+                       }
+               }
+               
+               return sequence;
+       }
+       catch(exception& e) {
+               m->errorOut(e, "Sequence", "getSequenceString");
+               exit(1);
+       }
+}
+//********************************************************************************************************************
+//comment can contain '>' so we need to account for that
+string Sequence::getCommentString(istringstream& fastaFile) {
+       try {
+               char letter;
+               string sequence = "";
+               
+               while(fastaFile){
+                       letter=fastaFile.get();
+                       if((letter == '\r') || (letter == '\n')){  
+                               gobble(fastaFile);  //in case its a \r\n situation
+                               break;
+                       }
+               }
+               
+               return sequence;
+       }
+       catch(exception& e) {
+               m->errorOut(e, "Sequence", "getCommentString");
                exit(1);
        }
 }
-
 //********************************************************************************************************************
 
 void Sequence::initialize(){
@@ -345,5 +443,65 @@ void Sequence::reverseComplement(){
        aligned = temp;
        
 }
-
+#ifdef USE_MPI 
 //********************************************************************************************************************
+int Sequence::MPISend(int receiver) {
+       try {
+               //send name - string
+               int length = name.length();
+               char buf[name.length()];
+               strcpy(buf, name.c_str()); 
+               
+               MPI_Send(&length, 1, MPI_INT, receiver, 2001, MPI_COMM_WORLD); 
+
+               MPI_Send(&buf, length, MPI_CHAR, receiver, 2001, MPI_COMM_WORLD);
+       
+               //send aligned - string
+               length = aligned.length();
+               char buf2[aligned.length()];
+               strcpy(buf2, aligned.c_str()); 
+       
+               MPI_Send(&length, 1, MPI_INT, receiver, 2001, MPI_COMM_WORLD); 
+       
+               MPI_Send(&buf2, length, MPI_CHAR, receiver, 2001, MPI_COMM_WORLD);
+       
+               return 0;
+
+       }
+       catch(exception& e) {
+               m->errorOut(e, "Sequence", "MPISend");
+               exit(1);
+       }
+}
+/**************************************************************************************************/
+int Sequence::MPIRecv(int sender) {
+       try {
+               MPI_Status status;
+       
+               //receive name - string
+               int length;
+               MPI_Recv(&length, 1, MPI_INT, sender, 2001, MPI_COMM_WORLD, &status);
+       
+               char buf[length];
+               MPI_Recv(&buf, length, MPI_CHAR, sender, 2001, MPI_COMM_WORLD, &status);
+               name = buf;
+               
+               //receive aligned - string
+               MPI_Recv(&length, 1, MPI_INT, sender, 2001, MPI_COMM_WORLD, &status);
+       
+               char buf2[length];
+               MPI_Recv(&buf2, length, MPI_CHAR, sender, 2001, MPI_COMM_WORLD, &status);
+               aligned = buf2;
+               
+               setAligned(aligned);
+               
+               return 0;
+
+       }
+       catch(exception& e) {
+               m->errorOut(e, "Sequence", "MPIRecv");
+               exit(1);
+       }
+}
+#endif
+/**************************************************************************************************/