X-Git-Url: https://git.donarmstrong.com/?a=blobdiff_plain;f=removelineagecommand.cpp;h=4cec90f567c764e8770987c150b67b5ececc0d35;hb=49d2b7459c5027557564b21e9487dadafbbbdc96;hp=22453a46078a7e99b33c07351722913fbb88f854;hpb=e150b0b0664caec517485ee6d69dcdade6dcae77;p=mothur.git diff --git a/removelineagecommand.cpp b/removelineagecommand.cpp index 22453a4..4cec90f 100644 --- a/removelineagecommand.cpp +++ b/removelineagecommand.cpp @@ -57,8 +57,31 @@ string RemoveLineageCommand::getHelpString(){ exit(1); } } - - +//********************************************************************************************************************** +string RemoveLineageCommand::getOutputFileNameTag(string type, string inputName=""){ + try { + string outputFileName = ""; + map >::iterator it; + + //is this a type this command creates + it = outputTypes.find(type); + if (it == outputTypes.end()) { m->mothurOut("[ERROR]: this command doesn't create a " + type + " output file.\n"); } + else { + if (type == "fasta") { outputFileName = "pick" + m->getExtension(inputName); } + else if (type == "taxonomy") { outputFileName = "pick" + m->getExtension(inputName); } + else if (type == "name") { outputFileName = "pick" + m->getExtension(inputName); } + else if (type == "group") { outputFileName = "pick" + m->getExtension(inputName); } + else if (type == "list") { outputFileName = "pick" + m->getExtension(inputName); } + else if (type == "alignreport") { outputFileName = "pick.align.report"; } + else { m->mothurOut("[ERROR]: No definition for type " + type + " output file tag.\n"); m->control_pressed = true; } + } + return outputFileName; + } + catch(exception& e) { + m->errorOut(e, "RemoveLineageCommand", "getOutputFileNameTag"); + exit(1); + } +} //********************************************************************************************************************** RemoveLineageCommand::RemoveLineageCommand(){ try { @@ -169,16 +192,19 @@ RemoveLineageCommand::RemoveLineageCommand(string option) { //check for required parameters fastafile = validParameter.validFile(parameters, "fasta", true); - if (fastafile == "not open") { abort = true; } + if (fastafile == "not open") { fastafile = ""; abort = true; } else if (fastafile == "not found") { fastafile = ""; } + else { m->setFastaFile(fastafile); } namefile = validParameter.validFile(parameters, "name", true); - if (namefile == "not open") { abort = true; } + if (namefile == "not open") { namefile = ""; abort = true; } else if (namefile == "not found") { namefile = ""; } + else { m->setNameFile(namefile); } groupfile = validParameter.validFile(parameters, "group", true); if (groupfile == "not open") { abort = true; } else if (groupfile == "not found") { groupfile = ""; } + else { m->setGroupFile(groupfile); } alignfile = validParameter.validFile(parameters, "alignreport", true); if (alignfile == "not open") { abort = true; } @@ -187,14 +213,15 @@ RemoveLineageCommand::RemoveLineageCommand(string option) { listfile = validParameter.validFile(parameters, "list", true); if (listfile == "not open") { abort = true; } else if (listfile == "not found") { listfile = ""; } + else { m->setListFile(listfile); } taxfile = validParameter.validFile(parameters, "taxonomy", true); - if (taxfile == "not open") { abort = true; } + if (taxfile == "not open") { taxfile = ""; abort = true; } else if (taxfile == "not found") { taxfile = m->getTaxonomyFile(); if (taxfile != "") { m->mothurOut("Using " + taxfile + " as input file for the taxonomy parameter."); m->mothurOutEndLine(); } else { m->mothurOut("You have no current taxonomy file and the taxonomy parameter is required."); m->mothurOutEndLine(); abort = true; } - } + }else { m->setTaxonomyFile(taxfile); } string usedDups = "true"; string temp = validParameter.validFile(parameters, "dups", false); @@ -211,12 +238,17 @@ RemoveLineageCommand::RemoveLineageCommand(string option) { if (taxons[0] == '\'') { taxons = taxons.substr(1); } if (taxons[(taxons.length()-1)] == '\'') { taxons = taxons.substr(0, (taxons.length()-1)); } } - + m->splitAtChar(taxons, listOfTaxons, '-'); if ((fastafile == "") && (namefile == "") && (groupfile == "") && (alignfile == "") && (listfile == "") && (taxfile == "")) { m->mothurOut("You must provide one of the following: fasta, name, group, alignreport, taxonomy or listfile."); m->mothurOutEndLine(); abort = true; } if ((usedDups != "") && (namefile == "")) { m->mothurOut("You may only use dups with the name option."); m->mothurOutEndLine(); abort = true; } - + + if ((namefile == "") && ((fastafile != "") || (taxfile != ""))){ + vector files; files.push_back(fastafile); files.push_back(taxfile); + parser.getNameFile(files); + } + } } @@ -243,7 +275,7 @@ int RemoveLineageCommand::execute(){ if (listfile != "") { readList(); } - if (m->control_pressed) { for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); } return 0; } + if (m->control_pressed) { for (int i = 0; i < outputNames.size(); i++) { m->mothurRemove(outputNames[i]); } return 0; } if (outputNames.size() != 0) { m->mothurOutEndLine(); @@ -293,8 +325,7 @@ int RemoveLineageCommand::readFasta(){ try { string thisOutputDir = outputDir; if (outputDir == "") { thisOutputDir += m->hasPath(fastafile); } - string outputFileName = thisOutputDir + m->getRootName(m->getSimpleName(fastafile)) + "pick" + m->getExtension(fastafile); - + string outputFileName = thisOutputDir + m->getRootName(m->getSimpleName(fastafile)) + getOutputFileNameTag("fasta", fastafile); ofstream out; m->openOutputFile(outputFileName, out); @@ -305,7 +336,7 @@ int RemoveLineageCommand::readFasta(){ bool wroteSomething = false; while(!in.eof()){ - if (m->control_pressed) { in.close(); out.close(); remove(outputFileName.c_str()); return 0; } + if (m->control_pressed) { in.close(); out.close(); m->mothurRemove(outputFileName); return 0; } Sequence currSeq(in); name = currSeq.getName(); @@ -339,8 +370,7 @@ int RemoveLineageCommand::readList(){ try { string thisOutputDir = outputDir; if (outputDir == "") { thisOutputDir += m->hasPath(listfile); } - string outputFileName = thisOutputDir + m->getRootName(m->getSimpleName(listfile)) + "pick" + m->getExtension(listfile); - + string outputFileName = thisOutputDir + m->getRootName(m->getSimpleName(listfile)) + getOutputFileNameTag("list", listfile); ofstream out; m->openOutputFile(outputFileName, out); @@ -359,7 +389,7 @@ int RemoveLineageCommand::readList(){ //for each bin for (int i = 0; i < list.getNumBins(); i++) { - if (m->control_pressed) { in.close(); out.close(); remove(outputFileName.c_str()); return 0; } + if (m->control_pressed) { in.close(); out.close(); m->mothurRemove(outputFileName); return 0; } //parse out names that are in accnos file string binnames = list.get(i); @@ -410,8 +440,7 @@ int RemoveLineageCommand::readName(){ try { string thisOutputDir = outputDir; if (outputDir == "") { thisOutputDir += m->hasPath(namefile); } - string outputFileName = thisOutputDir + m->getRootName(m->getSimpleName(namefile)) + "pick" + m->getExtension(namefile); - + string outputFileName = thisOutputDir + m->getRootName(m->getSimpleName(namefile)) + getOutputFileNameTag("name", namefile); ofstream out; m->openOutputFile(outputFileName, out); @@ -422,7 +451,7 @@ int RemoveLineageCommand::readName(){ bool wroteSomething = false; while(!in.eof()){ - if (m->control_pressed) { in.close(); out.close(); remove(outputFileName.c_str()); return 0; } + if (m->control_pressed) { in.close(); out.close(); m->mothurRemove(outputFileName); return 0; } in >> firstCol; in >> secondCol; @@ -488,8 +517,7 @@ int RemoveLineageCommand::readGroup(){ try { string thisOutputDir = outputDir; if (outputDir == "") { thisOutputDir += m->hasPath(groupfile); } - string outputFileName = thisOutputDir + m->getRootName(m->getSimpleName(groupfile)) + "pick" + m->getExtension(groupfile); - + string outputFileName = thisOutputDir + m->getRootName(m->getSimpleName(groupfile)) + getOutputFileNameTag("group", groupfile); ofstream out; m->openOutputFile(outputFileName, out); @@ -500,7 +528,7 @@ int RemoveLineageCommand::readGroup(){ bool wroteSomething = false; while(!in.eof()){ - if (m->control_pressed) { in.close(); out.close(); remove(outputFileName.c_str()); return 0; } + if (m->control_pressed) { in.close(); out.close(); m->mothurRemove(outputFileName); return 0; } in >> name; //read from first column in >> group; //read from second column @@ -531,7 +559,7 @@ int RemoveLineageCommand::readTax(){ try { string thisOutputDir = outputDir; if (outputDir == "") { thisOutputDir += m->hasPath(taxfile); } - string outputFileName = thisOutputDir + m->getRootName(m->getSimpleName(taxfile)) + "pick" + m->getExtension(taxfile); + string outputFileName = thisOutputDir + m->getRootName(m->getSimpleName(taxfile)) + getOutputFileNameTag("taxonomy", taxfile); ofstream out; m->openOutputFile(outputFileName, out); @@ -541,120 +569,132 @@ int RemoveLineageCommand::readTax(){ bool wroteSomething = false; - bool taxonsHasConfidence = false; - vector< map > searchTaxons; - string noConfidenceTaxons = taxons; - int hasConPos = taxons.find_first_of('('); - if (hasConPos != string::npos) { - taxonsHasConfidence = true; - searchTaxons = getTaxons(taxons); - noConfidenceTaxons = removeConfidences(taxons); + vector taxonsHasConfidence; taxonsHasConfidence.resize(listOfTaxons.size(), false); + vector< vector< map > > searchTaxons; searchTaxons.resize(listOfTaxons.size()); + vector noConfidenceTaxons; noConfidenceTaxons.resize(listOfTaxons.size(), ""); + + for (int i = 0; i < listOfTaxons.size(); i++) { + noConfidenceTaxons[i] = listOfTaxons[i]; + int hasConPos = listOfTaxons[i].find_first_of('('); + if (hasConPos != string::npos) { + taxonsHasConfidence[i] = true; + searchTaxons[i] = getTaxons(listOfTaxons[i]); + noConfidenceTaxons[i] = listOfTaxons[i]; + m->removeConfidences(noConfidenceTaxons[i]); + } } while(!in.eof()){ - if (m->control_pressed) { in.close(); out.close(); remove(outputFileName.c_str()); return 0; } + if (m->control_pressed) { in.close(); out.close(); m->mothurRemove(outputFileName); return 0; } in >> name; //read from first column in >> tax; //read from second column - string newtax = tax; + bool remove = false; - //if the users file contains confidence scores we want to ignore them when searching for the taxons, unless the taxon has them - if (!taxonsHasConfidence) { + for (int j = 0; j < listOfTaxons.size(); j++) { + string newtax = tax; - int hasConfidences = tax.find_first_of('('); - if (hasConfidences != string::npos) { - newtax = removeConfidences(tax); - } - - int pos = newtax.find(taxons); - - if (pos == string::npos) { - wroteSomething = true; - out << name << '\t' << tax << endl; - }else{ //this sequence contains the taxon the user wants to remove - names.insert(name); - } - - }else{//if taxons has them and you don't them remove taxons - int hasConfidences = tax.find_first_of('('); - if (hasConfidences == string::npos) { - - int pos = newtax.find(noConfidenceTaxons); + //if the users file contains confidence scores we want to ignore them when searching for the taxons, unless the taxon has them + if (!taxonsHasConfidence[j]) { - if (pos == string::npos) { - wroteSomething = true; - out << name << '\t' << tax << endl; - }else{ //this sequence contains the taxon the user wants to remove - names.insert(name); - } - }else { //both have confidences so we want to make sure the users confidences are greater then or equal to the taxons - //first remove confidences from both and see if the taxonomy exists - - string noNewTax = tax; int hasConfidences = tax.find_first_of('('); if (hasConfidences != string::npos) { - noNewTax = removeConfidences(tax); + newtax = tax; + m->removeConfidences(newtax); } - int pos = noNewTax.find(noConfidenceTaxons); + int pos = newtax.find(noConfidenceTaxons[j]); + + if (pos == string::npos) { + //wroteSomething = true; + //out << name << '\t' << tax << endl; + }else{ //this sequence contains the taxon the user wants to remove + names.insert(name); + remove=true; break; + } - if (pos != string::npos) { //if yes, then are the confidences okay + }else{//if taxons has them and you don't them remove taxons + int hasConfidences = tax.find_first_of('('); + if (hasConfidences == string::npos) { - bool remove = false; - vector< map > usersTaxon = getTaxons(newtax); + int pos = newtax.find(noConfidenceTaxons[j]); - //the usersTaxon is most likely longer than the searchTaxons, and searchTaxon[0] may relate to userTaxon[4] - //we want to "line them up", so we will find the the index where the searchstring starts - int index = 0; - for (int i = 0; i < usersTaxon.size(); i++) { + if (pos == string::npos) { + //wroteSomething = true; + //out << name << '\t' << tax << endl; + }else{ //this sequence contains the taxon the user wants to remove + names.insert(name); + remove=true; break; + } + }else { //both have confidences so we want to make sure the users confidences are greater then or equal to the taxons + //first remove confidences from both and see if the taxonomy exists + + string noNewTax = tax; + int hasConfidences = tax.find_first_of('('); + if (hasConfidences != string::npos) { + noNewTax = tax; + m->removeConfidences(noNewTax); + } + + int pos = noNewTax.find(noConfidenceTaxons[j]); + + if (pos != string::npos) { //if yes, then are the confidences okay - if (usersTaxon[i].begin()->first == searchTaxons[0].begin()->first) { - index = i; - int spot = 0; - bool goodspot = true; - //is this really the start, or are we dealing with a taxon of the same name? - while ((spot < searchTaxons.size()) && ((i+spot) < usersTaxon.size())) { - if (usersTaxon[i+spot].begin()->first != searchTaxons[spot].begin()->first) { goodspot = false; break; } - else { spot++; } - } + vector< map > usersTaxon = getTaxons(newtax); + + //the usersTaxon is most likely longer than the searchTaxons, and searchTaxon[0] may relate to userTaxon[4] + //we want to "line them up", so we will find the the index where the searchstring starts + int index = 0; + for (int i = 0; i < usersTaxon.size(); i++) { - if (goodspot) { break; } + if (usersTaxon[i].begin()->first == searchTaxons[j][0].begin()->first) { + index = i; + int spot = 0; + bool goodspot = true; + //is this really the start, or are we dealing with a taxon of the same name? + while ((spot < searchTaxons[j].size()) && ((i+spot) < usersTaxon.size())) { + if (usersTaxon[i+spot].begin()->first != searchTaxons[j][spot].begin()->first) { goodspot = false; break; } + else { spot++; } + } + + if (goodspot) { break; } + } } - } - - for (int i = 0; i < searchTaxons.size(); i++) { - if ((i+index) < usersTaxon.size()) { //just in case, should never be false - if (usersTaxon[i+index].begin()->second < searchTaxons[i].begin()->second) { //is the users cutoff less than the search taxons + for (int i = 0; i < searchTaxons[j].size(); i++) { + + if ((i+index) < usersTaxon.size()) { //just in case, should never be false + if (usersTaxon[i+index].begin()->second < searchTaxons[j][i].begin()->second) { //is the users cutoff less than the search taxons + remove = true; + break; + } + }else { remove = true; break; } + } + + //passed the test so remove you + if (remove) { + names.insert(name); + remove=true; break; }else { - remove = true; - break; + //wroteSomething = true; + //out << name << '\t' << tax << endl; } - } - - //passed the test so remove you - if (remove) { - names.insert(name); }else { - wroteSomething = true; - out << name << '\t' << tax << endl; + //wroteSomething = true; + //out << name << '\t' << tax << endl; } - }else { - wroteSomething = true; - out << name << '\t' << tax << endl; } } + } - - - + if (!remove) { wroteSomething = true; out << name << '\t' << tax << endl; } m->gobble(in); } in.close(); @@ -713,36 +753,13 @@ vector< map > RemoveLineageCommand::getTaxons(string tax) { exit(1); } } -/**************************************************************************************************/ -string RemoveLineageCommand::removeConfidences(string tax) { - try { - - string taxon = ""; - int taxLength = tax.length(); - for(int i=0;ierrorOut(e, "RemoveLineageCommand", "removeConfidences"); - exit(1); - } -} //********************************************************************************************************************** //alignreport file has a column header line then all other lines contain 16 columns. we just want the first column since that contains the name int RemoveLineageCommand::readAlign(){ try { string thisOutputDir = outputDir; if (outputDir == "") { thisOutputDir += m->hasPath(alignfile); } - string outputFileName = thisOutputDir + m->getRootName(m->getSimpleName(alignfile)) + "pick.align.report"; + string outputFileName = thisOutputDir + m->getRootName(m->getSimpleName(alignfile)) + getOutputFileNameTag("alignreport"); ofstream out; m->openOutputFile(outputFileName, out); @@ -761,7 +778,7 @@ int RemoveLineageCommand::readAlign(){ out << endl; while(!in.eof()){ - if (m->control_pressed) { in.close(); out.close(); remove(outputFileName.c_str()); return 0; } + if (m->control_pressed) { in.close(); out.close(); m->mothurRemove(outputFileName); return 0; } in >> name; //read from first column