]> git.donarmstrong.com Git - mothur.git/blobdiff - sequence.cpp
testing mpi
[mothur.git] / sequence.cpp
index 10a964848bd24f81a2f7658b0f1711427fbe3897..b73bfab7caefdb0a97e5f7cc8f3271ec3e51cb57 100644 (file)
 /***********************************************************************/
 
 Sequence::Sequence(){
+       m = MothurOut::getInstance();
        initialize();
 }
 
 /***********************************************************************/
 
 Sequence::Sequence(string newName, string sequence) {
-
-       initialize();   
-       name = newName;
-       
-       //setUnaligned removes any gap characters for us
-       setUnaligned(sequence);
-       setAligned(sequence);
-       
+       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();
+       int pid;
+       MPI_Comm_rank(MPI_COMM_WORLD, &pid); 
+       cout << pid << " after mothur instance " << &name << endl;
+               initialize();
+       cout << "after mothur initialize" << endl;
+               fastaString >> name;
+       cout << "after name "  << endl;
+               name = name.substr(1);
+               string sequence;
+               
+               //read comments
+               while ((name[0] == '#') && fastaString) { 
+                       while (fastaString)     {       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;
+                       }
+               }
+       cout << "after mothur comment" << endl; 
+               //read real sequence
+               while (fastaString)     {       char c = fastaString.get(); if (c == 10 || c == 13){    break;  }       } // get rest of line if there's any crap there
+       cout << "after mothur name" << endl;    
+               sequence = getSequenceString(fastaString);              
+       cout << "after mothur sequence" << endl;        
+               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);
-       
-       while (!fastaFile.eof())        {       char c = fastaFile.get(); if (c == 10 || c == 13){      break;  }       } // get rest of line if there's any crap there
-
-       char letter;
-       string sequence;
-       
-       while(fastaFile){
-               letter= fastaFile.get();
-               if(letter == '>'){
-                       fastaFile.putback(letter);
-                       break;
+       try {
+               m = MothurOut::getInstance();
+               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);
+                       
+                       if (fastaFile) {  
+                               fastaFile >> name;  
+                               name = name.substr(1);  
+                       }else { 
+                               name = "";
+                               break;
+                       }
                }
-               else if(isprint(letter)){
-                       letter = toupper(letter);
-                       if(letter == 'U'){letter = 'T';}
-                       sequence += letter;
+               
+               //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) {
+       try {
+               char letter;
+               string sequence = "";   
+               
+               while(fastaFile){
+                       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(ifstream& 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);
+       }
+}
+//********************************************************************************************************************
+string Sequence::getSequenceString(istringstream& fastaFile) {
+       try {
+               char letter;
+               string sequence = "";   
+               
+               while(fastaFile){
+                       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);
        }
-
-       setAligned(sequence);   
-       //setUnaligned removes any gap characters for us                                                
-       setUnaligned(sequence);                                                         
 }
-
 //********************************************************************************************************************
 
 void Sequence::initialize(){
@@ -109,6 +262,7 @@ void Sequence::setAligned(string sequence){
        //if the alignment starts or ends with a gap, replace it with a period to indicate missing data
        aligned = sequence;
        alignmentLength = aligned.length();
+       setUnaligned(sequence); 
 
        if(aligned[0] == '-'){
                for(int i=0;i<alignmentLength;i++){
@@ -291,6 +445,7 @@ void Sequence::reverseComplement(){
                else                                            {       temp += 'N';    }
        }
        unaligned = temp;
+       aligned = temp;
        
 }