X-Git-Url: https://git.donarmstrong.com/?a=blobdiff_plain;f=readcolumn.cpp;h=53a8c4263dddc4687acc5a64cf7c5b1d27aac647;hb=2c97dd48b8e27ee0a6a86c7a082f4c504c3357c6;hp=1d2920820076fd6c8597abedb2c59bf58f687daf;hpb=154e089bcd37c2c2f773e53ffb88a20170b27037;p=mothur.git diff --git a/readcolumn.cpp b/readcolumn.cpp index 1d29208..53a8c42 100644 --- a/readcolumn.cpp +++ b/readcolumn.cpp @@ -14,147 +14,141 @@ ReadColumnMatrix::ReadColumnMatrix(string df) : distFile(df){ - successOpen = openInputFile(distFile, fileHandle); + successOpen = m->openInputFile(distFile, fileHandle); + sim = false; } +/***********************************************************************/ + +ReadColumnMatrix::ReadColumnMatrix(string df, bool s) : distFile(df){ + + successOpen = m->openInputFile(distFile, fileHandle); + sim = s; +} /***********************************************************************/ -void ReadColumnMatrix::read(NameAssignment* nameMap){ +int ReadColumnMatrix::read(NameAssignment* nameMap){ try { - - string firstName, secondName; - float distance; - int nseqs = nameMap->size(); - list = new ListVector(nameMap->getListVector()); - - Progress* reading = new Progress("Reading matrix: ", nseqs * nseqs); + string firstName, secondName; + float distance; + int nseqs = nameMap->size(); + + list = new ListVector(nameMap->getListVector()); - int lt = 1; - int refRow = 0; //we'll keep track of one cell - Cell(refRow,refCol) - and see if it's transpose - int refCol = 0; //shows up later - Cell(refCol,refRow). If it does, then its a square matrix + Progress* reading = new Progress("Reading matrix: ", nseqs * nseqs); + + int lt = 1; + int refRow = 0; //we'll keep track of one cell - Cell(refRow,refCol) - and see if it's transpose + int refCol = 0; //shows up later - Cell(refCol,refRow). If it does, then its a square matrix + + //need to see if this is a square or a triangular matrix... - //need to see if this is a square or a triangular matrix... - while(fileHandle && lt == 1){ //let's assume it's a triangular matrix... - - fileHandle >> firstName >> secondName >> distance; // get the row and column names and distance + while(fileHandle && lt == 1){ //let's assume it's a triangular matrix... + - if(nameMap->count(firstName)==0){ - cerr << "AError: Sequence '" << firstName << "' was not found in the names file, please correct\n"; - } - if(nameMap->count(secondName)==0){ - cerr << "AError: Sequence '" << secondName << "' was not found in the names file, please correct\n"; - } - - if (distance == -1) { distance = 1000000; } - - if(distance < cutoff && nameMap->get(firstName) != nameMap->get(secondName)){ - if(nameMap->get(firstName) > nameMap->get(secondName)){ - PCell value(nameMap->get(firstName), nameMap->get(secondName), distance); + fileHandle >> firstName >> secondName >> distance; // get the row and column names and distance + + if (m->control_pressed) { fileHandle.close(); delete reading; return 0; } + + map::iterator itA = nameMap->find(firstName); + map::iterator itB = nameMap->find(secondName); - if(refRow == refCol){ // in other words, if we haven't loaded refRow and refCol... - refRow = nameMap->get(firstName); - refCol = nameMap->get(secondName); - D->addCell(value); - } - else if(refRow == nameMap->get(firstName) && refCol == nameMap->get(secondName)){ - lt = 0; - } - else{ - D->addCell(value); - } + if(itA == nameMap->end()){ m->mothurOut("AAError: Sequence '" + firstName + "' was not found in the names file, please correct\n"); exit(1); } + if(itB == nameMap->end()){ m->mothurOut("ABError: Sequence '" + secondName + "' was not found in the names file, please correct\n"); exit(1); } + + if (distance == -1) { distance = 1000000; } + else if (sim) { distance = 1.0 - distance; } //user has entered a sim matrix that we need to convert. + + if(distance < cutoff && itA != itB){ + if(itA->second > itB->second){ + PCell value(itA->second, itB->second, distance); + + if(refRow == refCol){ // in other words, if we haven't loaded refRow and refCol... + refRow = itA->second; + refCol = itB->second; + D->addCell(value); } - else if(nameMap->get(firstName) < nameMap->get(secondName)){ - PCell value(nameMap->get(secondName), nameMap->get(firstName), distance); - - if(refRow == refCol){ // in other words, if we haven't loaded refRow and refCol... - refRow = nameMap->get(firstName); - refCol = nameMap->get(secondName); - D->addCell(value); - } - else if(refRow == nameMap->get(secondName) && refCol == nameMap->get(firstName)){ - lt = 0; - } - else{ - D->addCell(value); - } + else if(refRow == itA->second && refCol == itB->second){ + lt = 0; + } + else{ + D->addCell(value); } - reading->update(nameMap->get(firstName) * nseqs); } - gobble(fileHandle); - } - - if(lt == 0){ // oops, it was square - fileHandle.close(); //let's start over - D->clear(); //let's start over - - openInputFile(distFile, fileHandle); //let's start over - - while(fileHandle){ - fileHandle >> firstName >> secondName >> distance; + else if(itA->second < itB->second){ + PCell value(itB->second, itA->second, distance); - if(nameMap->count(firstName)==0){ - cerr << "BError: Sequence '" << firstName << "' was not found in the names file, please correct\n"; + if(refRow == refCol){ // in other words, if we haven't loaded refRow and refCol... + refRow = itA->second; + refCol = itB->second; + D->addCell(value); } - if(nameMap->count(secondName)==0){ - cerr << "BError: Sequence '" << secondName << "' was not found in the names file, please correct\n"; + else if(refRow == itB->second && refCol == itA->second){ + lt = 0; } - - if (distance == -1) { distance = 1000000; } - - if(distance < cutoff && nameMap->get(firstName) > nameMap->get(secondName)){ - PCell value(nameMap->get(firstName), nameMap->get(secondName), distance); + else{ D->addCell(value); - reading->update(nameMap->get(firstName) * nseqs); } - - gobble(fileHandle); } + reading->update(itA->second * nseqs); } - // else if(lt == 0){ - // while(fileHandle){ - // fileHandle >> firstName >> secondName >> distance; - // - // if(nameMap->count(firstName)==0){ - // cerr << "CError: Sequence '" << firstName << "' was not found in the names file, please correct\n"; - // } - // if(nameMap->count(secondName)==0){ - // cerr << "CError: Sequence '" << secondName << "' was not found in the names file, please correct\n"; - // } - // if (distance == -1) { distance = 1000000; } - - // if(distance < cutoff && (*nameMap)[firstName].second < (*nameMap)[secondName].second){ - //// cout << (*nameMap)[secondName] << ' ' << (*nameMap)[firstName] << ' ' << distance << endl; - // D->addCell(Cell((*nameMap)[secondName].second, (*nameMap)[firstName].second, distance)); - // reading->update((*nameMap)[secondName].second * nseqs); - // } - // - // gobble(fileHandle); - // } - // } - reading->finish(); - fileHandle.close(); - - list->setLabel("0"); + m->gobble(fileHandle); + } + + if(lt == 0){ // oops, it was square + fileHandle.close(); //let's start over + D->clear(); //let's start over + + m->openInputFile(distFile, fileHandle); //let's start over + + while(fileHandle){ + fileHandle >> firstName >> secondName >> distance; + + if (m->control_pressed) { fileHandle.close(); delete reading; return 0; } + + map::iterator itA = nameMap->find(firstName); + map::iterator itB = nameMap->find(secondName); + + if(itA == nameMap->end()){ m->mothurOut("AAError: Sequence '" + firstName + "' was not found in the names file, please correct\n"); exit(1); } + if(itB == nameMap->end()){ m->mothurOut("ABError: Sequence '" + secondName + "' was not found in the names file, please correct\n"); exit(1); } + + if (distance == -1) { distance = 1000000; } + else if (sim) { distance = 1.0 - distance; } //user has entered a sim matrix that we need to convert. + + if(distance < cutoff && itA->second > itB->second){ + PCell value(itA->second, itB->second, distance); + D->addCell(value); + reading->update(itA->second * nseqs); + } + + m->gobble(fileHandle); + } + } + + if (m->control_pressed) { fileHandle.close(); delete reading; return 0; } + + reading->finish(); + fileHandle.close(); + + list->setLabel("0"); + + return 1; + } catch(exception& e) { - cout << "Standard Error: " << e.what() << " has occurred in the ReadColumnMatrix class Function read. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n"; + m->errorOut(e, "ReadColumnMatrix", "read"); exit(1); } - catch(...) { - cout << "An unknown error has occurred in the ReadColumnMatrix class function read. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n"; - exit(1); - } - } /***********************************************************************/ ReadColumnMatrix::~ReadColumnMatrix(){ - delete D; - delete list; + //delete D; + //delete list; }