X-Git-Url: https://git.donarmstrong.com/?a=blobdiff_plain;f=sequence.cpp;h=b73bfab7caefdb0a97e5f7cc8f3271ec3e51cb57;hb=17a6a53298a907c005fa93fb82af9e533adcda09;hp=752e081e367fc6d572b5622e0b4eb6003a28f9bb;hpb=648ec37228eb16075ace911dd5a5773cdfe683da;p=mothur.git diff --git a/sequence.cpp b/sequence.cpp index 752e081..b73bfab 100644 --- a/sequence.cpp +++ b/sequence.cpp @@ -12,52 +12,108 @@ /***********************************************************************/ 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(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); +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; - if (fastaFile) { - fastaFile >> name; - name = name.substr(1); - }else { - name = ""; - break; + //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); } - - //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 + 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){ + 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; + } + } + + //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); - sequence = getSequenceString(fastaFile); - - setAligned(sequence); - //setUnaligned removes any gap characters for us - setUnaligned(sequence); + 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 +137,7 @@ string Sequence::getSequenceString(ifstream& fastaFile) { return sequence; } catch(exception& e) { - errorOut(e, "Sequence", "getSequenceString"); + m->errorOut(e, "Sequence", "getSequenceString"); exit(1); } } @@ -103,11 +159,58 @@ 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){ + 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(){