#include "chimeracheckcommand.h"
+//**********************************************************************************************************************
+vector<string> ChimeraCheckCommand::getValidParameters(){
+ try {
+ string AlignArray[] = {"fasta","processors","increment","template","ksize","svg", "name","outputdir","inputdir" };
+ vector<string> myArray (AlignArray, AlignArray+(sizeof(AlignArray)/sizeof(string)));
+ return myArray;
+ }
+ catch(exception& e) {
+ m->errorOut(e, "ChimeraCheckCommand", "getValidParameters");
+ exit(1);
+ }
+}
+//**********************************************************************************************************************
+vector<string> ChimeraCheckCommand::getRequiredParameters(){
+ try {
+ string AlignArray[] = {"template","fasta"};
+ vector<string> myArray (AlignArray, AlignArray+(sizeof(AlignArray)/sizeof(string)));
+ return myArray;
+ }
+ catch(exception& e) {
+ m->errorOut(e, "ChimeraCheckCommand", "getRequiredParameters");
+ exit(1);
+ }
+}
+//**********************************************************************************************************************
+vector<string> ChimeraCheckCommand::getRequiredFiles(){
+ try {
+ vector<string> myArray;
+ return myArray;
+ }
+ catch(exception& e) {
+ m->errorOut(e, "ChimeraCheckCommand", "getRequiredFiles");
+ exit(1);
+ }
+}
+//**********************************************************************************************************************
+ChimeraCheckCommand::ChimeraCheckCommand(){
+ try {
+ vector<string> tempOutNames;
+ outputTypes["chimera"] = tempOutNames;
+ }
+ catch(exception& e) {
+ m->errorOut(e, "ChimeraCheckCommand", "ChimeraCheckCommand");
+ exit(1);
+ }
+}
//***************************************************************************************************************
-
ChimeraCheckCommand::ChimeraCheckCommand(string option) {
try {
abort = false;
OptionParser parser(option);
map<string,string> parameters = parser.getParameters();
- ValidParameters validParameter;
+ ValidParameters validParameter("chimera.check");
map<string,string>::iterator it;
//check to make sure all parameters are valid for command
if (validParameter.isValidParameter(it->first, myArray, it->second) != true) { abort = true; }
}
+ vector<string> tempOutNames;
+ outputTypes["chimera"] = tempOutNames;
+
//if the user changes the input directory command factory will send this info to us in the output parameter
string inputDir = validParameter.validFile(parameters, "inputdir", false);
if (inputDir == "not found"){ inputDir = ""; }
it = parameters.find("template");
//user has given a template file
if(it != parameters.end()){
- string path = hasPath(it->second);
+ string 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; }
}
fastafile = validParameter.validFile(parameters, "fasta", false);
if (fastafile == "not found") { fastafile = ""; m->mothurOut("fasta is a required parameter for the chimera.check command."); m->mothurOutEndLine(); abort = true; }
else {
- splitAtDash(fastafile, fastaFileNames);
+ m->splitAtDash(fastafile, 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;
- #ifdef USE_MPI
- int pid;
- MPI_Comm_size(MPI_COMM_WORLD, &processors); //set processors to the number of mpi processes running
- MPI_Comm_rank(MPI_COMM_WORLD, &pid); //find out who we are
-
- if (pid == 0) {
- #endif
-
- 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;
}
}
- in.close();
- #ifdef USE_MPI
- for (int j = 1; j < processors; j++) {
- MPI_Send(&ableToOpen, 1, MPI_INT, j, 2001, MPI_COMM_WORLD);
- }
- }else{
- MPI_Status status;
- MPI_Recv(&ableToOpen, 1, MPI_INT, 0, 2001, MPI_COMM_WORLD, &status);
+ //if you can't open it, try default location
+ 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;
}
-
- #endif
-
+ }
+
+ in.close();
+
if (ableToOpen == 1) {
m->mothurOut("Unable to open " + fastaFileNames[i] +". It will be disregarded."); m->mothurOutEndLine();
//erase from file list
namefile = validParameter.validFile(parameters, "name", false);
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;
- #ifdef USE_MPI
- int pid;
- MPI_Comm_size(MPI_COMM_WORLD, &processors); //set processors to the number of mpi processes running
- MPI_Comm_rank(MPI_COMM_WORLD, &pid); //find out who we are
-
- if (pid == 0) {
- #endif
-
- 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;
}
}
- in.close();
- #ifdef USE_MPI
- for (int j = 1; j < processors; j++) {
- MPI_Send(&ableToOpen, 1, MPI_INT, j, 2001, MPI_COMM_WORLD);
- }
- }else{
- MPI_Status status;
- MPI_Recv(&ableToOpen, 1, MPI_INT, 0, 2001, MPI_COMM_WORLD, &status);
+ //if you can't open it, try default location
+ 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;
}
-
- #endif
-
+ }
+
+ in.close();
+
if (ableToOpen == 1) {
m->mothurOut("Unable to open " + nameFileNames[i] + ". It will be disregarded."); m->mothurOutEndLine();
//erase from file list
convert(temp, ksize);
temp = validParameter.validFile(parameters, "svg", false); if (temp == "not found") { temp = "F"; }
- svg = isTrue(temp);
+ svg = m->isTrue(temp);
if (nameFileNames.size() != 0) { svg = true; }
temp = validParameter.validFile(parameters, "increment", false); if (temp == "not found") { temp = "10"; }
if (m->control_pressed) { delete chimera; return 0; }
- string outputFileName = outputDir + getRootName(getSimpleName(fastaFileNames[i])) + "chimeracheck.chimeras";
- outputNames.push_back(outputFileName);
+ if (outputDir == "") { outputDir = m->hasPath(fastaFileNames[i]); }//if user entered a file with a path then preserve it
+ string outputFileName = outputDir + m->getRootName(m->getSimpleName(fastaFileNames[i])) + "chimeracheck.chimeras";
+ outputNames.push_back(outputFileName); outputTypes["chimera"].push_back(outputFileName);
#ifdef USE_MPI
int pid, end, numSeqsPerProcessor;
int tag = 2001;
- vector<long> MPIPos;
+ vector<unsigned long int> MPIPos;
MPI_Status status;
MPI_Comm_rank(MPI_COMM_WORLD, &pid); //find out who we are
MPI_File_open(MPI_COMM_WORLD, inFileName, inMode, MPI_INFO_NULL, &inMPI); //comm, filename, mode, info, filepointer
MPI_File_open(MPI_COMM_WORLD, outFilename, outMode, MPI_INFO_NULL, &outMPI);
- if (m->control_pressed) { MPI_File_close(&inMPI); MPI_File_close(&outMPI); for (int j = 0; j < outputNames.size(); j++) { remove(outputNames[j].c_str()); } delete chimera; return 0; }
+ if (m->control_pressed) { MPI_File_close(&inMPI); MPI_File_close(&outMPI); for (int j = 0; j < outputNames.size(); j++) { remove(outputNames[j].c_str()); } outputTypes.clear(); delete chimera; return 0; }
if (pid == 0) { //you are the root process
- MPIPos = setFilePosFasta(fastaFileNames[i], numSeqs); //fills MPIPos, returns numSeqs
+ MPIPos = m->setFilePosFasta(fastaFileNames[i], numSeqs); //fills MPIPos, returns numSeqs
//send file positions to all processes
for(int j = 1; j < processors; j++) {
//align your part
driverMPI(startIndex, numSeqsPerProcessor, inMPI, outMPI, MPIPos);
- if (m->control_pressed) { MPI_File_close(&inMPI); MPI_File_close(&outMPI); for (int j = 0; j < outputNames.size(); j++) { remove(outputNames[j].c_str()); } delete chimera; return 0; }
+ if (m->control_pressed) { MPI_File_close(&inMPI); MPI_File_close(&outMPI); for (int j = 0; j < outputNames.size(); j++) { remove(outputNames[j].c_str()); } outputTypes.clear(); delete chimera; return 0; }
//wait on chidren
for(int j = 1; j < processors; j++) {
- char buf[4];
- MPI_Recv(buf, 4, MPI_CHAR, j, tag, MPI_COMM_WORLD, &status);
+ char buf[5];
+ MPI_Recv(buf, 5, MPI_CHAR, j, tag, MPI_COMM_WORLD, &status);
}
}else{ //you are a child process
MPI_Recv(&numSeqs, 1, MPI_INT, 0, tag, MPI_COMM_WORLD, &status);
//align your part
driverMPI(startIndex, numSeqsPerProcessor, inMPI, outMPI, MPIPos);
- if (m->control_pressed) { MPI_File_close(&inMPI); MPI_File_close(&outMPI); for (int j = 0; j < outputNames.size(); j++) { remove(outputNames[j].c_str()); } delete chimera; return 0; }
+ if (m->control_pressed) { MPI_File_close(&inMPI); MPI_File_close(&outMPI); for (int j = 0; j < outputNames.size(); j++) { remove(outputNames[j].c_str()); } outputTypes.clear(); delete chimera; return 0; }
//tell parent you are done.
- char buf[4];
+ char buf[5];
strcpy(buf, "done");
- MPI_Send(buf, 4, MPI_CHAR, 0, tag, MPI_COMM_WORLD);
+ MPI_Send(buf, 5, MPI_CHAR, 0, tag, MPI_COMM_WORLD);
}
//close files
MPI_Barrier(MPI_COMM_WORLD); //make everyone wait - just in case
#else
+ vector<unsigned long int> positions = m->divideFile(fastaFileNames[i], processors);
+
+ for (int s = 0; s < (positions.size()-1); s++) {
+ lines.push_back(new linePair(positions[s], positions[(s+1)]));
+ }
+
//break up file
#if defined (__APPLE__) || (__MACH__) || (linux) || (__linux)
if(processors == 1){
- ifstream inFASTA;
- openInputFile(fastaFileNames[i], inFASTA);
- getNumSeqs(inFASTA, numSeqs);
- inFASTA.close();
+ numSeqs = driver(lines[0], outputFileName, fastaFileNames[i]);
- lines.push_back(new linePair(0, numSeqs));
-
- driver(lines[0], outputFileName, fastaFileNames[i]);
-
- if (m->control_pressed) {
- for (int j = 0; j < outputNames.size(); j++) { remove(outputNames[j].c_str()); }
- for (int j = 0; j < lines.size(); j++) { delete lines[j]; } lines.clear();
- delete chimera;
- return 0;
- }
+ if (m->control_pressed) { for (int j = 0; j < outputNames.size(); j++) { remove(outputNames[j].c_str()); } for (int j = 0; j < lines.size(); j++) { delete lines[j]; } outputTypes.clear(); lines.clear(); delete chimera; return 0; }
}else{
- vector<unsigned long int> positions;
processIDS.resize(0);
- ifstream inFASTA;
- openInputFile(fastaFileNames[i], inFASTA);
-
- string input;
- while(!inFASTA.eof()){
- input = getline(inFASTA);
- if (input.length() != 0) {
- if(input[0] == '>'){ unsigned long int pos = inFASTA.tellg(); positions.push_back(pos - input.length() - 1); }
- }
- }
- inFASTA.close();
-
- numSeqs = positions.size();
-
- int numSeqsPerProcessor = numSeqs / processors;
-
- for (int j = 0; j < processors; j++) {
- unsigned long int startPos = positions[ j * numSeqsPerProcessor ];
- if(j == processors - 1){
- numSeqsPerProcessor = numSeqs - j * numSeqsPerProcessor;
- }
- lines.push_back(new linePair(startPos, numSeqsPerProcessor));
- }
-
-
- createProcesses(outputFileName, fastaFileNames[i]);
+ numSeqs = createProcesses(outputFileName, fastaFileNames[i]);
rename((outputFileName + toString(processIDS[0]) + ".temp").c_str(), outputFileName.c_str());
//append output files
for(int j=1;j<processors;j++){
- appendFiles((outputFileName + toString(processIDS[j]) + ".temp"), outputFileName);
+ m->appendFiles((outputFileName + toString(processIDS[j]) + ".temp"), outputFileName);
remove((outputFileName + toString(processIDS[j]) + ".temp").c_str());
}
if (m->control_pressed) {
- for (int j = 0; j < outputNames.size(); j++) { remove(outputNames[j].c_str()); }
+ for (int j = 0; j < outputNames.size(); j++) { remove(outputNames[j].c_str()); } outputTypes.clear();
for (int j = 0; j < lines.size(); j++) { delete lines[j]; } lines.clear();
delete chimera;
return 0;
}
#else
- ifstream inFASTA;
- openInputFile(fastaFileNames[i], inFASTA);
- getNumSeqs(inFASTA, numSeqs);
- inFASTA.close();
- lines.push_back(new linePair(0, numSeqs));
-
- driver(lines[0], outputFileName, fastaFileNames[i]);
+ numSeqs = driver(lines[0], outputFileName, fastaFileNames[i]);
- if (m->control_pressed) {
- for (int j = 0; j < lines.size(); j++) { delete lines[j]; } lines.clear();
- for (int j = 0; j < outputNames.size(); j++) { remove(outputNames[j].c_str()); }
- delete chimera;
- return 0;
- }
+ if (m->control_pressed) { for (int j = 0; j < lines.size(); j++) { delete lines[j]; } lines.clear(); for (int j = 0; j < outputNames.size(); j++) { remove(outputNames[j].c_str()); } outputTypes.clear(); delete chimera; return 0; }
#endif
#endif
delete chimera;
}
//**********************************************************************************************************************
-int ChimeraCheckCommand::driver(linePair* line, string outputFName, string filename){
+int ChimeraCheckCommand::driver(linePair* filePos, string outputFName, string filename){
try {
ofstream out;
- openOutputFile(outputFName, out);
+ m->openOutputFile(outputFName, out);
ofstream out2;
ifstream inFASTA;
- openInputFile(filename, inFASTA);
+ m->openInputFile(filename, inFASTA);
+
+ inFASTA.seekg(filePos->start);
+
+ bool done = false;
+ int count = 0;
+
+ while (!done) {
- inFASTA.seekg(line->start);
-
- for(int i=0;i<line->numSeqs;i++){
-
if (m->control_pressed) { return 1; }
- Sequence* candidateSeq = new Sequence(inFASTA); gobble(inFASTA);
+ Sequence* candidateSeq = new Sequence(inFASTA); m->gobble(inFASTA);
if (candidateSeq->getName() != "") { //incase there is a commented sequence at the end of a file
//find chimeras
}
delete candidateSeq;
+ #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((i+1) % 100 == 0){ m->mothurOut("Processing sequence: " + toString(i+1)); m->mothurOutEndLine(); }
+ if((count) % 100 == 0){ m->mothurOut("Processing sequence: " + toString(count)); m->mothurOutEndLine(); }
}
//report progress
- if((line->numSeqs) % 100 != 0){ m->mothurOut("Processing sequence: " + toString(line->numSeqs)); m->mothurOutEndLine(); }
+ if((count) % 100 != 0){ m->mothurOut("Processing sequence: " + toString(count)); m->mothurOutEndLine(); }
out.close();
inFASTA.close();
- return 0;
+ return count;
}
catch(exception& e) {
m->errorOut(e, "ChimeraCheckCommand", "driver");
}
//**********************************************************************************************************************
#ifdef USE_MPI
-int ChimeraCheckCommand::driverMPI(int start, int num, MPI_File& inMPI, MPI_File& outMPI, vector<long>& MPIPos){
+int ChimeraCheckCommand::driverMPI(int start, int num, MPI_File& inMPI, MPI_File& outMPI, vector<unsigned long int>& MPIPos){
try {
MPI_File outAccMPI;
MPI_Status status;
istringstream iss (tempBuf,istringstream::in);
delete buf4;
- Sequence* candidateSeq = new Sequence(iss); gobble(iss);
+ Sequence* candidateSeq = new Sequence(iss); m->gobble(iss);
if (candidateSeq->getName() != "") { //incase there is a commented sequence at the end of a file
//find chimeras
try {
#if defined (__APPLE__) || (__MACH__) || (linux) || (__linux)
int process = 0;
- // processIDS.resize(0);
+ int num = 0;
//loop through and create all the processes you want
while (process != processors) {
processIDS.push_back(pid); //create map from line number to pid so you can append files in correct order later
process++;
}else if (pid == 0){
- driver(lines[process], outputFileName + toString(getpid()) + ".temp", filename);
+ num = driver(lines[process], outputFileName + toString(getpid()) + ".temp", filename);
+
+ //pass numSeqs to parent
+ ofstream out;
+ string tempFile = outputFileName + toString(getpid()) + ".num.temp";
+ m->openOutputFile(tempFile, out);
+ out << num << endl;
+ out.close();
+
+ 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);
- }else { m->mothurOut("unable to spawn the necessary processes."); m->mothurOutEndLine(); exit(0); }
+ }
}
//force parent to wait until all the processes are done
wait(&temp);
}
- return 0;
+ for (int i = 0; i < processIDS.size(); i++) {
+ ifstream in;
+ string tempFile = outputFileName + 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());
+ }
+
+ return num;
#endif
}
catch(exception& e) {