+/**************************************************************************************************/
+//custom data structure for threads to use.
+// This is passed by void pointer so it can be any data type
+// that can be passed using a single void pointer (LPVOID).
+struct preClusterData {
+ string fastafile;
+ string namefile;
+ string groupfile, countfile;
+ string newFName, newNName, newMName;
+ MothurOut* m;
+ int start;
+ int end, count;
+ int diffs, threadID;
+ vector<string> groups;
+ vector<string> mapFileNames;
+ bool topdown;
+
+ preClusterData(){}
+ preClusterData(string f, string n, string g, string c, string nff, string nnf, string nmf, vector<string> gr, MothurOut* mout, int st, int en, int d, bool td, int tid) {
+ fastafile = f;
+ namefile = n;
+ groupfile = g;
+ newFName = nff;
+ newNName = nnf;
+ newMName = nmf;
+ m = mout;
+ start = st;
+ end = en;
+ diffs = d;
+ threadID = tid;
+ groups = gr;
+ countfile = c;
+ topdown = td;
+ count=0;
+ }
+};
+
+/**************************************************************************************************/
+#if defined (__APPLE__) || (__MACH__) || (linux) || (__linux) || (__linux__) || (__unix__) || (__unix)
+#else
+static DWORD WINAPI MyPreclusterThreadFunction(LPVOID lpParam){
+ preClusterData* pDataArray;
+ pDataArray = (preClusterData*)lpParam;
+
+ try {
+
+ //parse fasta and name file by group
+ SequenceParser* parser;
+ SequenceCountParser* cparser;
+ if (pDataArray->countfile != "") {
+ cparser = new SequenceCountParser(pDataArray->countfile, pDataArray->fastafile);
+ }else {
+ if (pDataArray->namefile != "") { parser = new SequenceParser(pDataArray->groupfile, pDataArray->fastafile, pDataArray->namefile); }
+ else { parser = new SequenceParser(pDataArray->groupfile, pDataArray->fastafile); }
+ }
+
+ int numSeqs = 0;
+ vector<seqPNode> alignSeqs;
+ //clear out old files
+ ofstream outF; pDataArray->m->openOutputFile(pDataArray->newFName, outF); outF.close();
+ ofstream outN; pDataArray->m->openOutputFile(pDataArray->newNName, outN); outN.close();
+
+ //precluster each group
+ for (int k = pDataArray->start; k < pDataArray->end; k++) {
+
+ pDataArray->count++;
+
+ int start = time(NULL);
+
+ if (pDataArray->m->control_pressed) { delete parser; return 0; }
+
+ pDataArray->m->mothurOutEndLine(); pDataArray->m->mothurOut("Processing group " + pDataArray->groups[k] + ":"); pDataArray->m->mothurOutEndLine();
+
+ map<string, string> thisNameMap;
+ vector<Sequence> thisSeqs;
+ if (pDataArray->groupfile != "") {
+ thisSeqs = parser->getSeqs(pDataArray->groups[k]);
+ }else if (pDataArray->countfile != "") {
+ thisSeqs = cparser->getSeqs(pDataArray->groups[k]);
+ }
+ if (pDataArray->namefile != "") { thisNameMap = parser->getNameMap(pDataArray->groups[k]); }
+
+ //fill alignSeqs with this groups info.
+ ////////////////////////////////////////////////////
+ //numSeqs = loadSeqs(thisNameMap, thisSeqs); same function below
+
+ int length = 0;
+ alignSeqs.clear();
+ map<string, string>::iterator it;
+ bool error = false;
+ map<string, int> thisCount;
+ if (pDataArray->countfile != "") { thisCount = cparser->getCountTable(pDataArray->groups[k]); }
+
+
+ for (int i = 0; i < thisSeqs.size(); i++) {
+
+ if (pDataArray->m->control_pressed) { delete parser; return 0; }
+
+ if (pDataArray->namefile != "") {
+ it = thisNameMap.find(thisSeqs[i].getName());
+
+ //should never be true since parser checks for this
+ if (it == thisNameMap.end()) { pDataArray->m->mothurOut(thisSeqs[i].getName() + " is not in your names file, please correct."); pDataArray->m->mothurOutEndLine(); error = true; }
+ else{
+ //get number of reps
+ int numReps = 1;
+ for(int j=0;j<(it->second).length();j++){
+ if((it->second)[j] == ','){ numReps++; }
+ }
+
+ seqPNode tempNode(numReps, thisSeqs[i], it->second);
+ alignSeqs.push_back(tempNode);
+ if (thisSeqs[i].getAligned().length() > length) { length = thisSeqs[i].getAligned().length(); }
+ }
+ }else { //no names file, you are identical to yourself
+ int numRep = 1;
+ if (pDataArray->countfile != "") {
+ map<string, int>::iterator it2 = thisCount.find(thisSeqs[i].getName());
+
+ //should never be true since parser checks for this
+ if (it2 == thisCount.end()) { pDataArray->m->mothurOut(thisSeqs[i].getName() + " is not in your count file, please correct."); pDataArray->m->mothurOutEndLine(); error = true; }
+ else { numRep = it2->second; }
+ }
+ seqPNode tempNode(numRep, thisSeqs[i], thisSeqs[i].getName());
+ alignSeqs.push_back(tempNode);
+ if (thisSeqs[i].getAligned().length() > length) { length = thisSeqs[i].getAligned().length(); }
+ }
+ }
+
+ //sanity check
+ if (error) { pDataArray->m->control_pressed = true; }
+
+ thisSeqs.clear();
+ numSeqs = alignSeqs.size();
+
+ ////////////////////////////////////////////////////
+
+ if (pDataArray->m->control_pressed) { delete parser; return 0; }
+
+ if (pDataArray->diffs > length) { pDataArray->m->mothurOut("Error: diffs is greater than your sequence length."); pDataArray->m->mothurOutEndLine(); pDataArray->m->control_pressed = true; return 0; }
+
+ ////////////////////////////////////////////////////
+ //int count = process(); - same function below
+
+ ofstream out;
+ pDataArray->m->openOutputFile(pDataArray->newMName+pDataArray->groups[k]+".map", out);
+ pDataArray->mapFileNames.push_back(pDataArray->newMName+pDataArray->groups[k]+".map");
+
+ //sort seqs by number of identical seqs
+ if (pDataArray->topdown) { sort(alignSeqs.begin(), alignSeqs.end(), comparePriorityTopDown); }
+ else { sort(alignSeqs.begin(), alignSeqs.end(), comparePriorityDownTop); }
+
+ int count = 0;
+
+ if (pDataArray->topdown) {
+ //think about running through twice...
+ for (int i = 0; i < numSeqs; i++) {
+
+ //are you active
+ // itActive = active.find(alignSeqs[i].seq.getName());
+
+ if (alignSeqs[i].active) { //this sequence has not been merged yet
+
+ string chunk = alignSeqs[i].seq.getName() + "\t" + toString(alignSeqs[i].numIdentical) + "\t" + toString(0) + "\t" + alignSeqs[i].seq.getAligned() + "\n";