X-Git-Url: https://git.donarmstrong.com/?p=mothur.git;a=blobdiff_plain;f=readcolumn.cpp;h=e14125cc6d94f563e37d5ea1cea8235a5c8df04b;hp=f61a40c483b69f7418ca4d2702ca995f4180f3e2;hb=d1c97b8c04bb75faca1e76ffad60b37a4d789d3d;hpb=fdc1f6eaf544f695fc1511f24bddd7e6069c33ba diff --git a/readcolumn.cpp b/readcolumn.cpp index f61a40c..e14125c 100644 --- a/readcolumn.cpp +++ b/readcolumn.cpp @@ -14,9 +14,17 @@ 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; +} /***********************************************************************/ @@ -26,7 +34,7 @@ int ReadColumnMatrix::read(NameAssignment* nameMap){ string firstName, secondName; float distance; int nseqs = nameMap->size(); - + DMatrix->resize(nseqs); list = new ListVector(nameMap->getListVector()); Progress* reading = new Progress("Reading matrix: ", nseqs * nseqs); @@ -40,91 +48,90 @@ int ReadColumnMatrix::read(NameAssignment* nameMap){ while(fileHandle && lt == 1){ //let's assume it's a triangular matrix... - fileHandle >> firstName >> secondName >> distance; // get the row and column names and distance + fileHandle >> firstName; m->gobble(fileHandle); + fileHandle >> secondName; m->gobble(fileHandle); + fileHandle >> distance; // get the row and column names and distance + + if (m->debug) { cout << firstName << '\t' << secondName << '\t' << distance << endl; } 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()){ - cerr << "AAError: Sequence '" << firstName << "' was not found in the names file, please correct\n"; exit(1); - } - if(itB == nameMap->end()){ - cerr << "ABError: Sequence '" << secondName << "' was not found in the names file, please correct\n"; exit(1); - } + + 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 (globaldata->sim) { distance = 1.0 - distance; } //user has entered a sim matrix that we need to convert. + 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); - + PDistCell value(itA->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); + DMatrix->addCell(itB->second, value); } else if(refRow == itA->second && refCol == itB->second){ lt = 0; } else{ - D->addCell(value); + DMatrix->addCell(itB->second, value); } } else if(itA->second < itB->second){ - PCell value(itB->second, itA->second, distance); + PDistCell value(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); + DMatrix->addCell(itA->second, value); } else if(refRow == itB->second && refCol == itA->second){ lt = 0; } else{ - D->addCell(value); + DMatrix->addCell(itA->second, value); } } reading->update(itA->second * nseqs); } - gobble(fileHandle); + m->gobble(fileHandle); } if(lt == 0){ // oops, it was square fileHandle.close(); //let's start over - D->clear(); //let's start over + DMatrix->clear(); //let's start over - openInputFile(distFile, fileHandle); //let's start over + m->openInputFile(distFile, fileHandle); //let's start over while(fileHandle){ - fileHandle >> firstName >> secondName >> distance; + fileHandle >> firstName; m->gobble(fileHandle); + fileHandle >> secondName; m->gobble(fileHandle); + fileHandle >> 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(itA == nameMap->end()){ - cerr << "BError: Sequence '" << firstName << "' was not found in the names file, please correct\n"; - } - if(itB == nameMap->end()){ - cerr << "BError: Sequence '" << secondName << "' was not found in the names file, please correct\n"; - } + 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 (globaldata->sim) { distance = 1.0 - distance; } //user has entered a sim matrix that we need to convert. + 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); + PDistCell value(itA->second, distance); + DMatrix->addCell(itB->second, value); reading->update(itA->second * nseqs); } - gobble(fileHandle); + m->gobble(fileHandle); } } @@ -143,12 +150,130 @@ int ReadColumnMatrix::read(NameAssignment* nameMap){ exit(1); } } - /***********************************************************************/ -ReadColumnMatrix::~ReadColumnMatrix(){ - //delete D; - //delete list; +int ReadColumnMatrix::read(CountTable* countTable){ + try { + + string firstName, secondName; + float distance; + int nseqs = countTable->size(); + + DMatrix->resize(nseqs); + list = new ListVector(countTable->getListVector()); + + 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... + + while(fileHandle && lt == 1){ //let's assume it's a triangular matrix... + + + fileHandle >> firstName; m->gobble(fileHandle); + fileHandle >> secondName; m->gobble(fileHandle); + fileHandle >> distance; // get the row and column names and distance + + if (m->control_pressed) { fileHandle.close(); delete reading; return 0; } + + int itA = countTable->get(firstName); + int itB = countTable->get(secondName); + + if (m->control_pressed) { 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 > itB){ + PDistCell value(itA, distance); + + + if(refRow == refCol){ // in other words, if we haven't loaded refRow and refCol... + refRow = itA; + refCol = itB; + DMatrix->addCell(itB, value); + } + else if(refRow == itA && refCol == itB){ + lt = 0; + } + else{ + DMatrix->addCell(itB, value); + } + } + else if(itA < itB){ + PDistCell value(itB, distance); + + if(refRow == refCol){ // in other words, if we haven't loaded refRow and refCol... + refRow = itA; + refCol = itB; + DMatrix->addCell(itA, value); + } + else if(refRow == itB && refCol == itA){ + lt = 0; + } + else{ + DMatrix->addCell(itA, value); + } + } + reading->update(itA * nseqs); + } + m->gobble(fileHandle); + } + + if(lt == 0){ // oops, it was square + + fileHandle.close(); //let's start over + DMatrix->clear(); //let's start over + + m->openInputFile(distFile, fileHandle); //let's start over + + while(fileHandle){ + fileHandle >> firstName; m->gobble(fileHandle); + fileHandle >> secondName; m->gobble(fileHandle); + fileHandle >> distance; // get the row and column names and distance + + if (m->control_pressed) { fileHandle.close(); delete reading; return 0; } + + int itA = countTable->get(firstName); + int itB = countTable->get(secondName); + + + if (m->control_pressed) { 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){ + PDistCell value(itA, distance); + DMatrix->addCell(itB, value); + reading->update(itA * 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) { + m->errorOut(e, "ReadColumnMatrix", "read"); + exit(1); + } } +/***********************************************************************/ +ReadColumnMatrix::~ReadColumnMatrix(){} +/***********************************************************************/