changed reading of name file to use buffered reads. note the splitAtWhiteSpace functi...

[mothur.git] / trimseqscommand.h

diff --git a/trimseqscommand.h b/trimseqscommand.h
index 9ba64c4b62b7c7e87f213472b730a91bd0b37305..e280c8a1d4b50336f125dda5788d91d2af76c09a 100644 (file)
--- a/trimseqscommand.h
+++ b/trimseqscommand.h
@@ -63,6 +63,7 @@ private:
        vector<string> revPrimer, outputNames;
        set<string> filesToRemove;
        map<string, int> barcodes;
+    map<string, int> rbarcodes;
        vector<string> groupVector;
        map<string, int> primers;
     vector<string>  linker;
@@ -101,6 +102,7 @@ struct trimData {
        double qRollAverage, qThreshold, qWindowAverage, qAverage;
     vector<string> revPrimer;
        map<string, int> barcodes;
+    map<string, int> rbarcodes;
        map<string, int> primers;
     vector<string>  linker;
     vector<string>  spacer;
@@ -112,7 +114,7 @@ struct trimData {
     
        trimData(){}
        trimData(string fn, string qn, string nf, string tn, string sn, string tqn, string sqn, string tnn, string snn, string gn, vector<vector<string> > ffn, vector<vector<string> > qfn, vector<vector<string> > nfn, unsigned long long lstart, unsigned long long lend, unsigned long long qstart, unsigned long long qend,  MothurOut* mout,
-                      int pd, int bd, int ld, int sd, int td, map<string, int> pri, map<string, int> bar, vector<string> revP, vector<string> li, vector<string> spa, 
+                      int pd, int bd, int ld, int sd, int td, map<string, int> pri, map<string, int> bar, map<string, int> rbar, vector<string> revP, vector<string> li, vector<string> spa, 
                       vector<string> priNameVector, vector<string> barNameVector, bool cGroup, bool aFiles, bool keepF, int keepfi, int removeL,
                       int WindowStep, int WindowSize, int WindowAverage, bool trim, double Threshold, double Average, double RollAverage,
                       int minL, int maxA, int maxH, int maxL, bool fli, map<string, string> nm) {
@@ -141,6 +143,7 @@ struct trimData {
         sdiffs = sd;
         tdiffs = td;
         barcodes = bar;
+        rbarcodes = rbar;
         primers = pri;      numFPrimers = primers.size();
         revPrimer = revP;   numRPrimers = revPrimer.size();
         linker = li;        numLinkers = linker.size();
@@ -237,7 +240,7 @@ static DWORD WINAPI MyTrimThreadFunction(LPVOID lpParam){
                }
                
                
-               TrimOligos trimOligos(pDataArray->pdiffs, pDataArray->bdiffs, pDataArray->ldiffs, pDataArray->sdiffs, pDataArray->primers, pDataArray->barcodes, pDataArray->revPrimer, pDataArray->linker, pDataArray->spacer);
+               TrimOligos trimOligos(pDataArray->pdiffs, pDataArray->bdiffs, pDataArray->ldiffs, pDataArray->sdiffs, pDataArray->primers, pDataArray->barcodes, pDataArray->rbarcodes, pDataArray->revPrimer, pDataArray->linker, pDataArray->spacer);
         
                pDataArray->count = pDataArray->lineEnd;
                for(int i = 0; i < pDataArray->lineEnd; i++){ //end is the number of sequences to process
@@ -277,7 +280,13 @@ static DWORD WINAPI MyTrimThreadFunction(LPVOID lpParam){
                                        if(success > pDataArray->bdiffs)                {       trashCode += 'b';       }
                                        else{ currentSeqsDiffs += success;  }
                                }
-                               
+                
+                               if(pDataArray->rbarcodes.size() != 0){
+                                       success = trimOligos.stripRBarcode(currSeq, currQual, barcodeIndex);
+                                       if(success > pDataArray->bdiffs)                {       trashCode += 'b';       }
+                                       else{ currentSeqsDiffs += success;  }
+                               }
+                
                 if(pDataArray->numSpacers != 0){
                                        success = trimOligos.stripSpacer(currSeq, currQual);
                                        if(success > pDataArray->sdiffs)                {       trashCode += 's';       }
@@ -404,11 +413,13 @@ static DWORD WINAPI MyTrimThreadFunction(LPVOID lpParam){
                                                        
                                                        outGroupsFile << currSeq.getName() << '\t' << thisGroup << endl;
                                                        
+                            int numRedundants = 0;
                                                        if (pDataArray->nameFile != "") {
                                                                map<string, string>::iterator itName = pDataArray->nameMap.find(currSeq.getName());
                                                                if (itName != pDataArray->nameMap.end()) { 
                                                                        vector<string> thisSeqsNames; 
                                                                        pDataArray->m->splitAtChar(itName->second, thisSeqsNames, ',');
+                                    numRedundants = thisSeqsNames.size()-1; //we already include ourselves below
                                                                        for (int k = 1; k < thisSeqsNames.size(); k++) { //start at 1 to skip self
                                                                                outGroupsFile << thisSeqsNames[k] << '\t' << thisGroup << endl;
                                                                        }
@@ -416,8 +427,8 @@ static DWORD WINAPI MyTrimThreadFunction(LPVOID lpParam){
                                                        }
                                                        
                                                        map<string, int>::iterator it = pDataArray->groupCounts.find(thisGroup);
-                                                       if (it == pDataArray->groupCounts.end()) {      pDataArray->groupCounts[thisGroup] = 1; }
-                                                       else { pDataArray->groupCounts[it->first]++; }
+                                                       if (it == pDataArray->groupCounts.end()) {      pDataArray->groupCounts[thisGroup] = 1 + numRedundants; }
+                                                       else { pDataArray->groupCounts[it->first] += (1 + numRedundants); }
                             
                                                }
                                        }