#include "phylosummary.h"
#include "knn.h"
-//**********************************************************************************************************************
+//**********************************************************************************************************************
+vector<string> ClassifySeqsCommand::getValidParameters(){
+ try {
+ string AlignArray[] = {"template","fasta","name","group","search","ksize","method","processors","taxonomy","match","mismatch","gapopen","gapextend","numwanted","cutoff","probs","iters", "outputdir","inputdir"};
+ vector<string> myArray (AlignArray, AlignArray+(sizeof(AlignArray)/sizeof(string)));
+ return myArray;
+ }
+ catch(exception& e) {
+ m->errorOut(e, "ClassifySeqsCommand", "getValidParameters");
+ exit(1);
+ }
+}
+//**********************************************************************************************************************
+ClassifySeqsCommand::ClassifySeqsCommand(){
+ try {
+ abort = true;
+ //initialize outputTypes
+ vector<string> tempOutNames;
+ outputTypes["taxonomy"] = tempOutNames;
+ outputTypes["taxsummary"] = tempOutNames;
+ outputTypes["matchdist"] = tempOutNames;
+ }
+ catch(exception& e) {
+ m->errorOut(e, "ClassifySeqsCommand", "ClassifySeqsCommand");
+ exit(1);
+ }
+}
+//**********************************************************************************************************************
+vector<string> ClassifySeqsCommand::getRequiredParameters(){
+ try {
+ string Array[] = {"fasta","template","taxonomy"};
+ vector<string> myArray (Array, Array+(sizeof(Array)/sizeof(string)));
+ return myArray;
+ }
+ catch(exception& e) {
+ m->errorOut(e, "ClassifySeqsCommand", "getRequiredParameters");
+ exit(1);
+ }
+}
+//**********************************************************************************************************************
+vector<string> ClassifySeqsCommand::getRequiredFiles(){
+ try {
+ vector<string> myArray;
+ return myArray;
+ }
+ catch(exception& e) {
+ m->errorOut(e, "ClassifySeqsCommand", "getRequiredFiles");
+ exit(1);
+ }
+}
+//**********************************************************************************************************************
ClassifySeqsCommand::ClassifySeqsCommand(string option) {
try {
abort = false;
if (validParameter.isValidParameter(it->first, myArray, it->second) != true) { abort = true; }
}
+ //initialize outputTypes
+ vector<string> tempOutNames;
+ outputTypes["taxonomy"] = tempOutNames;
+ outputTypes["taxsummary"] = tempOutNames;
+ outputTypes["matchdist"] = tempOutNames;
+
//if the user changes the output directory command factory will send this info to us in the output parameter
outputDir = validParameter.validFile(parameters, "outputdir", false); if (outputDir == "not found"){ outputDir = ""; }
it = parameters.find("template");
//user has given a template file
if(it != parameters.end()){
- path = hasPath(it->second);
+ path = m->hasPath(it->second);
//if the user has not given a path then, add inputdir. else leave path alone.
if (path == "") { parameters["template"] = inputDir + it->second; }
}
it = parameters.find("taxonomy");
//user has given a template file
if(it != parameters.end()){
- path = hasPath(it->second);
+ path = m->hasPath(it->second);
//if the user has not given a path then, add inputdir. else leave path alone.
if (path == "") { parameters["taxonomy"] = inputDir + it->second; }
}
it = parameters.find("group");
//user has given a template file
if(it != parameters.end()){
- path = hasPath(it->second);
+ path = m->hasPath(it->second);
//if the user has not given a path then, add inputdir. else leave path alone.
if (path == "") { parameters["group"] = inputDir + it->second; }
}
fastaFileName = validParameter.validFile(parameters, "fasta", false);
if (fastaFileName == "not found") { m->mothurOut("fasta is a required parameter for the classify.seqs command."); m->mothurOutEndLine(); abort = true; }
else {
- splitAtDash(fastaFileName, fastaFileNames);
+ m->splitAtDash(fastaFileName, fastaFileNames);
//go through files and make sure they are good, if not, then disregard them
for (int i = 0; i < fastaFileNames.size(); i++) {
if (inputDir != "") {
- string path = hasPath(fastaFileNames[i]);
+ string path = m->hasPath(fastaFileNames[i]);
//if the user has not given a path then, add inputdir. else leave path alone.
if (path == "") { fastaFileNames[i] = inputDir + fastaFileNames[i]; }
}
int ableToOpen;
ifstream in;
- ableToOpen = openInputFile(fastaFileNames[i], in, "noerror");
+ ableToOpen = m->openInputFile(fastaFileNames[i], in, "noerror");
//if you can't open it, try default location
if (ableToOpen == 1) {
if (m->getDefaultPath() != "") { //default path is set
- string tryPath = m->getDefaultPath() + getSimpleName(fastaFileNames[i]);
+ string tryPath = m->getDefaultPath() + m->getSimpleName(fastaFileNames[i]);
m->mothurOut("Unable to open " + fastaFileNames[i] + ". Trying default " + tryPath); m->mothurOutEndLine();
- ableToOpen = openInputFile(tryPath, in, "noerror");
+ ifstream in2;
+ ableToOpen = m->openInputFile(tryPath, in2, "noerror");
+ in2.close();
+ fastaFileNames[i] = tryPath;
+ }
+ }
+
+ if (ableToOpen == 1) {
+ if (m->getOutputDir() != "") { //default path is set
+ string tryPath = m->getOutputDir() + m->getSimpleName(fastaFileNames[i]);
+ m->mothurOut("Unable to open " + fastaFileNames[i] + ". Trying output directory " + tryPath); m->mothurOutEndLine();
+ ifstream in2;
+ ableToOpen = m->openInputFile(tryPath, in2, "noerror");
+ in2.close();
fastaFileNames[i] = tryPath;
}
}
+
in.close();
if (ableToOpen == 1) {
if (namefile == "not found") { namefile = ""; }
else {
- splitAtDash(namefile, namefileNames);
+ m->splitAtDash(namefile, namefileNames);
//go through files and make sure they are good, if not, then disregard them
for (int i = 0; i < namefileNames.size(); i++) {
if (inputDir != "") {
- string path = hasPath(namefileNames[i]);
+ string path = m->hasPath(namefileNames[i]);
//if the user has not given a path then, add inputdir. else leave path alone.
if (path == "") { namefileNames[i] = inputDir + namefileNames[i]; }
}
int ableToOpen;
ifstream in;
- ableToOpen = openInputFile(namefileNames[i], in, "noerror");
+ ableToOpen = m->openInputFile(namefileNames[i], in, "noerror");
//if you can't open it, try default location
if (ableToOpen == 1) {
if (m->getDefaultPath() != "") { //default path is set
- string tryPath = m->getDefaultPath() + getSimpleName(namefileNames[i]);
+ string tryPath = m->getDefaultPath() + m->getSimpleName(namefileNames[i]);
m->mothurOut("Unable to open " + namefileNames[i] + ". Trying default " + tryPath); m->mothurOutEndLine();
- ableToOpen = openInputFile(tryPath, in, "noerror");
+ ifstream in2;
+ ableToOpen = m->openInputFile(tryPath, in2, "noerror");
+ in2.close();
+ namefileNames[i] = tryPath;
+ }
+ }
+
+ if (ableToOpen == 1) {
+ if (m->getOutputDir() != "") { //default path is set
+ string tryPath = m->getOutputDir() + m->getSimpleName(namefileNames[i]);
+ m->mothurOut("Unable to open " + namefileNames[i] + ". Trying output directory " + tryPath); m->mothurOutEndLine();
+ ifstream in2;
+ ableToOpen = m->openInputFile(tryPath, in2, "noerror");
+ in2.close();
namefileNames[i] = tryPath;
}
}
groupfile = validParameter.validFile(parameters, "group", false);
if (groupfile == "not found") { groupfile = ""; }
else {
- splitAtDash(groupfile, groupfileNames);
+ m->splitAtDash(groupfile, groupfileNames);
//go through files and make sure they are good, if not, then disregard them
for (int i = 0; i < groupfileNames.size(); i++) {
if (inputDir != "") {
- string path = hasPath(groupfileNames[i]);
+ string path = m->hasPath(groupfileNames[i]);
//if the user has not given a path then, add inputdir. else leave path alone.
if (path == "") { groupfileNames[i] = inputDir + groupfileNames[i]; }
}
int ableToOpen;
ifstream in;
- ableToOpen = openInputFile(groupfileNames[i], in, "noerror");
+ ableToOpen = m->openInputFile(groupfileNames[i], in, "noerror");
//if you can't open it, try default location
if (ableToOpen == 1) {
if (m->getDefaultPath() != "") { //default path is set
- string tryPath = m->getDefaultPath() + getSimpleName(groupfileNames[i]);
+ string tryPath = m->getDefaultPath() + m->getSimpleName(groupfileNames[i]);
m->mothurOut("Unable to open " + groupfileNames[i] + ". Trying default " + tryPath); m->mothurOutEndLine();
- ableToOpen = openInputFile(tryPath, in, "noerror");
+ ifstream in2;
+ ableToOpen = m->openInputFile(tryPath, in2, "noerror");
+ in2.close();
+ groupfileNames[i] = tryPath;
+ }
+ }
+
+ if (ableToOpen == 1) {
+ if (m->getOutputDir() != "") { //default path is set
+ string tryPath = m->getOutputDir() + m->getSimpleName(groupfileNames[i]);
+ m->mothurOut("Unable to open " + groupfileNames[i] + ". Trying output directory " + tryPath); m->mothurOutEndLine();
+ ifstream in2;
+ ableToOpen = m->openInputFile(tryPath, in2, "noerror");
+ in2.close();
groupfileNames[i] = tryPath;
}
}
+
in.close();
if (ableToOpen == 1) {
convert(temp, cutoff);
temp = validParameter.validFile(parameters, "probs", false); if (temp == "not found"){ temp = "true"; }
- probs = isTrue(temp);
+ probs = m->isTrue(temp);
temp = validParameter.validFile(parameters, "iters", false); if (temp == "not found") { temp = "100"; }
convert(temp, iters);
if (m->control_pressed) { delete classify; return 0; }
- vector<string> outputNames;
for (int s = 0; s < fastaFileNames.size(); s++) {
m->mothurOut("Classifying sequences from " + fastaFileNames[s] + " ..." ); m->mothurOutEndLine();
- string RippedTaxName = getRootName(getSimpleName(taxonomyFileName));
- RippedTaxName = getExtension(RippedTaxName.substr(0, RippedTaxName.length()-1));
+ string RippedTaxName = m->getRootName(m->getSimpleName(taxonomyFileName));
+ RippedTaxName = m->getExtension(RippedTaxName.substr(0, RippedTaxName.length()-1));
if (RippedTaxName[0] == '.') { RippedTaxName = RippedTaxName.substr(1, RippedTaxName.length()); }
RippedTaxName += ".";
- if (outputDir == "") { outputDir += hasPath(fastaFileNames[s]); }
- string newTaxonomyFile = outputDir + getRootName(getSimpleName(fastaFileNames[s])) + RippedTaxName + "taxonomy";
- string tempTaxonomyFile = outputDir + getRootName(getSimpleName(fastaFileNames[s])) + "taxonomy.temp";
- string taxSummary = outputDir + getRootName(getSimpleName(fastaFileNames[s])) + RippedTaxName + "tax.summary";
+ if (outputDir == "") { outputDir += m->hasPath(fastaFileNames[s]); }
+ string newTaxonomyFile = outputDir + m->getRootName(m->getSimpleName(fastaFileNames[s])) + RippedTaxName + "taxonomy";
+ string tempTaxonomyFile = outputDir + m->getRootName(m->getSimpleName(fastaFileNames[s])) + "taxonomy.temp";
+ string taxSummary = outputDir + m->getRootName(m->getSimpleName(fastaFileNames[s])) + RippedTaxName + "tax.summary";
if ((method == "knn") && (search == "distance")) {
- string DistName = outputDir + getRootName(getSimpleName(fastaFileNames[s])) + "match.dist";
- classify->setDistName(DistName); outputNames.push_back(DistName);
+ string DistName = outputDir + m->getRootName(m->getSimpleName(fastaFileNames[s])) + "match.dist";
+ classify->setDistName(DistName); outputNames.push_back(DistName); outputTypes["matchdist"].push_back(DistName);
}
- outputNames.push_back(newTaxonomyFile);
- outputNames.push_back(taxSummary);
+ outputNames.push_back(newTaxonomyFile); outputTypes["taxonomy"].push_back(newTaxonomyFile);
+ outputNames.push_back(taxSummary); outputTypes["taxsummary"].push_back(taxSummary);
int start = time(NULL);
int numFastaSeqs = 0;
int outMode=MPI_MODE_CREATE|MPI_MODE_WRONLY;
int inMode=MPI_MODE_RDONLY;
- //char* outNewTax = new char[newTaxonomyFile.length()];
- //memcpy(outNewTax, newTaxonomyFile.c_str(), newTaxonomyFile.length());
-
char outNewTax[1024];
strcpy(outNewTax, newTaxonomyFile.c_str());
-
- //char* outTempTax = new char[tempTaxonomyFile.length()];
- //memcpy(outTempTax, tempTaxonomyFile.c_str(), tempTaxonomyFile.length());
char outTempTax[1024];
strcpy(outTempTax, tempTaxonomyFile.c_str());
-
- //char* inFileName = new char[fastaFileNames[s].length()];
- //memcpy(inFileName, fastaFileNames[s].c_str(), fastaFileNames[s].length());
char inFileName[1024];
strcpy(inFileName, fastaFileNames[s].c_str());
MPI_File_open(MPI_COMM_WORLD, outNewTax, outMode, MPI_INFO_NULL, &outMPINewTax);
MPI_File_open(MPI_COMM_WORLD, outTempTax, outMode, MPI_INFO_NULL, &outMPITempTax);
- //delete outNewTax;
- //delete outTempTax;
- //delete inFileName;
-
- if (m->control_pressed) { MPI_File_close(&inMPI); MPI_File_close(&outMPINewTax); MPI_File_close(&outMPITempTax); delete classify; return 0; }
+ if (m->control_pressed) { outputTypes.clear(); MPI_File_close(&inMPI); MPI_File_close(&outMPINewTax); MPI_File_close(&outMPITempTax); delete classify; return 0; }
if (pid == 0) { //you are the root process
- MPIPos = setFilePosFasta(fastaFileNames[s], numFastaSeqs); //fills MPIPos, returns numSeqs
+ MPIPos = m->setFilePosFasta(fastaFileNames[s], numFastaSeqs); //fills MPIPos, returns numSeqs
//send file positions to all processes
for(int i = 1; i < processors; i++) {
//align your part
driverMPI(startIndex, numSeqsPerProcessor, inMPI, outMPINewTax, outMPITempTax, MPIPos);
- if (m->control_pressed) { MPI_File_close(&inMPI); MPI_File_close(&outMPINewTax); MPI_File_close(&outMPITempTax); for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); } delete classify; return 0; }
+ if (m->control_pressed) { outputTypes.clear(); MPI_File_close(&inMPI); MPI_File_close(&outMPINewTax); MPI_File_close(&outMPITempTax); for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); } delete classify; return 0; }
for (int i = 1; i < processors; i++) {
int done;
//align your part
driverMPI(startIndex, numSeqsPerProcessor, inMPI, outMPINewTax, outMPITempTax, MPIPos);
- if (m->control_pressed) { MPI_File_close(&inMPI); MPI_File_close(&outMPINewTax); MPI_File_close(&outMPITempTax); delete classify; return 0; }
+ if (m->control_pressed) { outputTypes.clear(); MPI_File_close(&inMPI); MPI_File_close(&outMPINewTax); MPI_File_close(&outMPITempTax); delete classify; return 0; }
int done = 0;
MPI_Send(&done, 1, MPI_INT, 0, tag, MPI_COMM_WORLD);
#else
- vector<unsigned long int> positions = divideFile(fastaFileNames[s], processors);
+ vector<unsigned long int> positions = m->divideFile(fastaFileNames[s], processors);
for (int i = 0; i < (positions.size()-1); i++) {
lines.push_back(new linePair(positions[i], positions[(i+1)]));
numFastaSeqs = createProcesses(newTaxonomyFile, tempTaxonomyFile, fastaFileNames[s]);
- rename((newTaxonomyFile + toString(processIDS[0]) + ".temp").c_str(), newTaxonomyFile.c_str());
- rename((tempTaxonomyFile + toString(processIDS[0]) + ".temp").c_str(), tempTaxonomyFile.c_str());
-
- for(int i=1;i<processors;i++){
- appendTaxFiles((newTaxonomyFile + toString(processIDS[i]) + ".temp"), newTaxonomyFile);
- appendTaxFiles((tempTaxonomyFile + toString(processIDS[i]) + ".temp"), tempTaxonomyFile);
- remove((newTaxonomyFile + toString(processIDS[i]) + ".temp").c_str());
- remove((tempTaxonomyFile + toString(processIDS[i]) + ".temp").c_str());
- }
-
}
#else
numFastaSeqs = driver(lines[0], newTaxonomyFile, tempTaxonomyFile, fastaFileNames[s]);
nameMap.clear(); //remove old names
ifstream inNames;
- openInputFile(namefileNames[s], inNames);
+ m->openInputFile(namefileNames[s], inNames);
string firstCol, secondCol;
while(!inNames.eof()) {
- inNames >> firstCol >> secondCol; gobble(inNames);
+ inNames >> firstCol >> secondCol; m->gobble(inNames);
vector<string> temp;
- splitAtComma(secondCol, temp);
+ m->splitAtComma(secondCol, temp);
nameMap[firstCol] = temp;
}
PhyloSummary taxaSum(taxonomyFileName, group);
- if (m->control_pressed) { for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); } delete classify; return 0; }
+ if (m->control_pressed) { outputTypes.clear(); for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); } delete classify; return 0; }
if (namefile == "") { taxaSum.summarize(tempTaxonomyFile); }
else {
ifstream in;
- openInputFile(tempTaxonomyFile, in);
+ m->openInputFile(tempTaxonomyFile, in);
//read in users taxonomy file and add sequences to tree
string name, taxon;
while(!in.eof()){
- in >> name >> taxon; gobble(in);
+ in >> name >> taxon; m->gobble(in);
itNames = nameMap.find(name);
}
remove(tempTaxonomyFile.c_str());
- if (m->control_pressed) { for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); } delete classify; return 0; }
+ if (m->control_pressed) { outputTypes.clear(); for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); } delete classify; return 0; }
//print summary file
ofstream outTaxTree;
- openOutputFile(taxSummary, outTaxTree);
+ m->openOutputFile(taxSummary, outTaxTree);
taxaSum.print(outTaxTree);
outTaxTree.close();
//output taxonomy with the unclassified bins added
ifstream inTax;
- openInputFile(newTaxonomyFile, inTax);
+ m->openInputFile(newTaxonomyFile, inTax);
ofstream outTax;
string unclass = newTaxonomyFile + ".unclass.temp";
- openOutputFile(unclass, outTax);
+ m->openOutputFile(unclass, outTax);
//get maxLevel from phylotree so you know how many 'unclassified's to add
int maxLevel = taxaSum.getMaxLevel();
//read taxfile - this reading and rewriting is done to preserve the confidence scores.
string name, taxon;
while (!inTax.eof()) {
- if (m->control_pressed) { for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); } remove(unclass.c_str()); delete classify; return 0; }
+ if (m->control_pressed) { outputTypes.clear(); for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); } remove(unclass.c_str()); delete classify; return 0; }
- inTax >> name >> taxon; gobble(inTax);
+ inTax >> name >> taxon; m->gobble(inTax);
string newTax = addUnclassifieds(taxon, maxLevel);
int ClassifySeqsCommand::createProcesses(string taxFileName, string tempTaxFile, string filename) {
try {
#if defined (__APPLE__) || (__MACH__) || (linux) || (__linux)
- int process = 0;
+ int process = 1;
int num = 0;
//loop through and create all the processes you want
//pass numSeqs to parent
ofstream out;
- string tempFile = toString(getpid()) + ".temp";
- openOutputFile(tempFile, out);
+ string tempFile = filename + toString(getpid()) + ".num.temp";
+ m->openOutputFile(tempFile, out);
out << num << endl;
out.close();
exit(0);
- }else { m->mothurOut("unable to spawn the necessary processes."); m->mothurOutEndLine(); exit(0); }
+ }else {
+ m->mothurOut("[ERROR]: unable to spawn the necessary processes."); m->mothurOutEndLine();
+ for (int i = 0; i < processIDS.size(); i++) { kill (processIDS[i], SIGINT); }
+ exit(0);
+ }
}
+ //parent does its part
+ num = driver(lines[0], taxFileName, tempTaxFile, filename);
+
//force parent to wait until all the processes are done
- for (int i=0;i<processors;i++) {
+ for (int i=0;i<processIDS.size();i++) {
int temp = processIDS[i];
wait(&temp);
}
for (int i = 0; i < processIDS.size(); i++) {
ifstream in;
- string tempFile = toString(processIDS[i]) + ".temp";
- openInputFile(tempFile, in);
+ string tempFile = filename + toString(processIDS[i]) + ".num.temp";
+ m->openInputFile(tempFile, in);
if (!in.eof()) { int tempNum = 0; in >> tempNum; num += tempNum; }
in.close(); remove(tempFile.c_str());
}
+ for(int i=0;i<processIDS.size();i++){
+ appendTaxFiles((taxFileName + toString(processIDS[i]) + ".temp"), taxFileName);
+ appendTaxFiles((tempTaxFile + toString(processIDS[i]) + ".temp"), tempTaxFile);
+ remove((taxFileName + toString(processIDS[i]) + ".temp").c_str());
+ remove((tempTaxFile + toString(processIDS[i]) + ".temp").c_str());
+ }
+
return num;
#endif
}
ofstream output;
ifstream input;
- openOutputFileAppend(filename, output);
- openInputFile(temp, input);
+ m->openOutputFileAppend(filename, output);
+ m->openInputFile(temp, input);
while(char c = input.get()){
if(input.eof()) { break; }
int ClassifySeqsCommand::driver(linePair* filePos, string taxFName, string tempTFName, string filename){
try {
ofstream outTax;
- openOutputFile(taxFName, outTax);
+ m->openOutputFile(taxFName, outTax);
ofstream outTaxSimple;
- openOutputFile(tempTFName, outTaxSimple);
+ m->openOutputFile(tempTFName, outTaxSimple);
ifstream inFASTA;
- openInputFile(filename, inFASTA);
+ m->openInputFile(filename, inFASTA);
string taxonomy;
bool done = false;
int count = 0;
-
+
while (!done) {
if (m->control_pressed) { return 0; }
- Sequence* candidateSeq = new Sequence(inFASTA); gobble(inFASTA);
-
+ Sequence* candidateSeq = new Sequence(inFASTA); m->gobble(inFASTA);
+
if (candidateSeq->getName() != "") {
+
taxonomy = classify->getTaxonomy(candidateSeq);
if (m->control_pressed) { delete candidateSeq; return 0; }
}
delete candidateSeq;
- unsigned long int pos = inFASTA.tellg();
- if ((pos == -1) || (pos >= filePos->end)) { break; }
+ #if defined (__APPLE__) || (__MACH__) || (linux) || (__linux)
+ unsigned long int pos = inFASTA.tellg();
+ if ((pos == -1) || (pos >= filePos->end)) { break; }
+ #else
+ if (inFASTA.eof()) { break; }
+ #endif
//report progress
if((count) % 100 == 0){ m->mothurOut("Processing sequence: " + toString(count)); m->mothurOutEndLine(); }
+
}
//report progress
if((count) % 100 != 0){ m->mothurOut("Processing sequence: " + toString(count)); m->mothurOutEndLine(); }
-
+
inFASTA.close();
outTax.close();
outTaxSimple.close();
string firstCol, secondCol;
while(!iss.eof()) {
- iss >> firstCol >> secondCol; gobble(iss);
+ iss >> firstCol >> secondCol; m->gobble(iss);
vector<string> temp;
- splitAtComma(secondCol, temp);
+ m->splitAtComma(secondCol, temp);
nameMap[firstCol] = temp;
}