X-Git-Url: https://git.donarmstrong.com/?a=blobdiff_plain;f=chimerauchimecommand.cpp;h=d7df400d1c70cd8842dd48120f5d9974311ad87d;hb=a89c6295ae9a35fcaaab7fa50dcb68360dd543b0;hp=e82dc1bc8dce5ffd0887025d5816b5eec8d0eab9;hpb=037b7fccc64a5c7d5d5c23a949273a912160a400;p=mothur.git diff --git a/chimerauchimecommand.cpp b/chimerauchimecommand.cpp index e82dc1b..d7df400 100644 --- a/chimerauchimecommand.cpp +++ b/chimerauchimecommand.cpp @@ -564,14 +564,12 @@ ChimeraUchimeCommand::ChimeraUchimeCommand(string option) { temp = validParameter.validFile(parameters, "dereplicate", false); - if (temp == "not found") { - if (groupfile != "") { temp = "false"; } - else { temp = "true"; } - } + if (temp == "not found") { temp = "false"; } dups = m->isTrue(temp); if (hasName && (templatefile != "self")) { m->mothurOut("You have provided a namefile and the reference parameter is not set to self. I am not sure what reference you are trying to use, aborting."); m->mothurOutEndLine(); abort=true; } + if (hasCount && (templatefile != "self")) { m->mothurOut("You have provided a countfile and the reference parameter is not set to self. I am not sure what reference you are trying to use, aborting."); m->mothurOutEndLine(); abort=true; } if (hasGroup && (templatefile != "self")) { m->mothurOut("You have provided a group file and the reference parameter is not set to self. I am not sure what reference you are trying to use, aborting."); m->mothurOutEndLine(); abort=true; } //look for uchime exe @@ -677,7 +675,7 @@ int ChimeraUchimeCommand::execute(){ int error; if (hasCount) { CountTable ct; - ct.readTable(nameFile); + ct.readTable(nameFile, true, false); for(map::iterator it = seqs.begin(); it != seqs.end(); it++) { int num = ct.getNumSeqs(it->first); if (num == 0) { error = 1; } @@ -728,7 +726,7 @@ int ChimeraUchimeCommand::execute(){ if(processors == 1) { totalSeqs = driverGroups(outputFileName, newFasta, accnosFileName, alnsFileName, newCountFile, 0, groups.size(), groups); if (hasCount && dups) { - CountTable c; c.readTable(nameFile); + CountTable c; c.readTable(nameFile, true, false); if (!m->isBlank(newCountFile)) { ifstream in2; m->openInputFile(newCountFile, in2); @@ -758,7 +756,7 @@ int ChimeraUchimeCommand::execute(){ if (hasCount) { set doNotRemove; - CountTable c; c.readTable(newCountFile); + CountTable c; c.readTable(newCountFile, true, true); vector namesInTable = c.getNamesOfSeqs(); for (int i = 0; i < namesInTable.size(); i++) { int temp = c.getNumSeqs(namesInTable[i]); @@ -1635,17 +1633,17 @@ int ChimeraUchimeCommand::createProcesses(string outputFileName, string filename //loop through and create all the processes you want while (process != processors) { - int pid = fork(); + pid_t pid = fork(); if (pid > 0) { 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){ - num = driver(outputFileName + toString(getpid()) + ".temp", files[process], accnos + toString(getpid()) + ".temp", alns + toString(getpid()) + ".temp", numChimeras); + num = driver(outputFileName + toString(m->mothurGetpid(process)) + ".temp", files[process], accnos + toString(m->mothurGetpid(process)) + ".temp", alns + toString(m->mothurGetpid(process)) + ".temp", numChimeras); //pass numSeqs to parent ofstream out; - string tempFile = outputFileName + toString(getpid()) + ".num.temp"; + string tempFile = outputFileName + toString(m->mothurGetpid(process)) + ".num.temp"; m->openOutputFile(tempFile, out); out << num << endl; out << numChimeras << endl; @@ -1798,36 +1796,38 @@ int ChimeraUchimeCommand::createProcessesGroups(string outputFName, string filen int num = 0; CountTable newCount; - if (hasCount && dups) { newCount.readTable(nameFile); } + if (hasCount && dups) { newCount.readTable(nameFile, true, false); } //sanity check if (groups.size() < processors) { processors = groups.size(); } //divide the groups between the processors vector lines; - int numGroupsPerProcessor = groups.size() / processors; - for (int i = 0; i < processors; i++) { - int startIndex = i * numGroupsPerProcessor; - int endIndex = (i+1) * numGroupsPerProcessor; - if(i == (processors - 1)){ endIndex = groups.size(); } - lines.push_back(linePair(startIndex, endIndex)); - } - + int remainingPairs = groups.size(); + int startIndex = 0; + for (int remainingProcessors = processors; remainingProcessors > 0; remainingProcessors--) { + int numPairs = remainingPairs; //case for last processor + if (remainingProcessors != 1) { numPairs = ceil(remainingPairs / remainingProcessors); } + lines.push_back(linePair(startIndex, (startIndex+numPairs))); //startIndex, endIndex + startIndex = startIndex + numPairs; + remainingPairs = remainingPairs - numPairs; + } + #if defined (__APPLE__) || (__MACH__) || (linux) || (__linux) || (__linux__) || (__unix__) || (__unix) //loop through and create all the processes you want while (process != processors) { - int pid = fork(); + pid_t pid = fork(); if (pid > 0) { 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){ - num = driverGroups(outputFName + toString(getpid()) + ".temp", filename + toString(getpid()) + ".temp", accnos + toString(getpid()) + ".temp", alns + toString(getpid()) + ".temp", accnos + ".byCount." + toString(getpid()) + ".temp", lines[process].start, lines[process].end, groups); + num = driverGroups(outputFName + toString(m->mothurGetpid(process)) + ".temp", filename + toString(m->mothurGetpid(process)) + ".temp", accnos + toString(m->mothurGetpid(process)) + ".temp", alns + toString(m->mothurGetpid(process)) + ".temp", accnos + ".byCount." + toString(m->mothurGetpid(process)) + ".temp", lines[process].start, lines[process].end, groups); //pass numSeqs to parent ofstream out; - string tempFile = outputFName + toString(getpid()) + ".num.temp"; + string tempFile = outputFName + toString(m->mothurGetpid(process)) + ".num.temp"; m->openOutputFile(tempFile, out); out << num << endl; out.close();