Add option -debug to control -vmer debugging output.

arangrhie · Apr 19, 2021 · 09c026d · 09c026d
1 parent 34e9418
commit 09c026d
Show file tree

Hide file tree

Showing 3 changed files with 41 additions and 45 deletions.
diff --git a/src/merfin/merfin-globals.H b/src/merfin/merfin-globals.H
@@ -229,9 +229,7 @@ public:
 
   uint32                  reportType    = OP_NONE;
 
-
-
-
+  bool                    debug         = false;
 
 public:
   double                  peak          = 0;

diff --git a/src/merfin/merfin-variants.C b/src/merfin/merfin-variants.C
@@ -142,7 +142,7 @@ processVariants(void *G, void *T, void *S) {
 
   //  Initialize the per-thread data if needed.
 
-  if (t->oDebug == nullptr) {
+  if ((g->debug == true) && (t->oDebug == nullptr)) {
     char  name[FILENAME_MAX+1];
 
     snprintf(name, FILENAME_MAX, "%s.%02d.debug.gz", g->outName, t->threadID);
@@ -237,42 +237,44 @@ processVariants(void *G, void *T, void *S) {
 
     //  Save debug info.
 
-    for (uint64 idx = 0; idx < seqMer->seqs.size(); idx++) {
-      fprintf(t->oDebug->file(), "%lu\t%s:%u-%u\t%s\t%u\t%.5f\t%.5f\t%.5f\t%.5f\t%.5f\t",
-              t->varMerId++,
-              s->seq.ident(),
-              rStart,
-              rEnd,
-              seqMer->seqs[idx].c_str(),  //  seq
-              seqMer->numMs[idx],         //  missing
-              seqMer->getMinAbsK(idx),
-              seqMer->getMaxAbsK(idx),
-              seqMer->getMedAbsK(idx),
-              seqMer->getAvgAbsK(idx),
-              //seqMer->getAvgAbsdK(idx, RefAvgK),
-              seqMer->getTotdK(idx));
-
-      //  new vcf records
-      //  fprintf(stderr, "%s:%u-%u seqMer->gtPaths[idx].size() %d\n", s->seq.ident(), rStart, rEnd, seqMer->gtPaths[idx].size());
-
-      if  ( seqMer->gtPaths[idx].size() > 0 ) {
-        for (uint64 i = 0; i < seqMer->gtPaths[idx].size(); i++) {
-          // Ignore the ref-allele (0/0) GTs
-          // print only the non-ref allele variants for fixing
-          int altIdx = seqMer->gtPaths[idx][i];
-          if (altIdx > 0) {
-            fprintf(t->oDebug->file(), "%s %u . %s %s . PASS . GT 1/1  ",
-                    s->seq.ident(),
-                    gts[i]->_pos+1,
-                    gts[i]->_alleles[0],
-                    gts[i]->_alleles[altIdx]
-                    );
+    if (t->oDebug) {
+      for (uint64 idx = 0; idx < seqMer->seqs.size(); idx++) {
+        fprintf(t->oDebug->file(), "%lu\t%s:%u-%u\t%s\t%u\t%.5f\t%.5f\t%.5f\t%.5f\t%.5f\t",
+                t->varMerId++,
+                s->seq.ident(),
+                rStart,
+                rEnd,
+                seqMer->seqs[idx].c_str(),  //  seq
+                seqMer->numMs[idx],         //  missing
+                seqMer->getMinAbsK(idx),
+                seqMer->getMaxAbsK(idx),
+                seqMer->getMedAbsK(idx),
+                seqMer->getAvgAbsK(idx),
+                //seqMer->getAvgAbsdK(idx, RefAvgK),
+                seqMer->getTotdK(idx));
+
+        //  new vcf records
+        //  fprintf(stderr, "%s:%u-%u seqMer->gtPaths[idx].size() %d\n", s->seq.ident(), rStart, rEnd, seqMer->gtPaths[idx].size());
+
+        if  ( seqMer->gtPaths[idx].size() > 0 ) {
+          for (uint64 i = 0; i < seqMer->gtPaths[idx].size(); i++) {
+            // Ignore the ref-allele (0/0) GTs
+            // print only the non-ref allele variants for fixing
+            int altIdx = seqMer->gtPaths[idx][i];
+            if (altIdx > 0) {
+              fprintf(t->oDebug->file(), "%s %u . %s %s . PASS . GT 1/1  ",
+                      s->seq.ident(),
+                      gts[i]->_pos+1,
+                      gts[i]->_alleles[0],
+                      gts[i]->_alleles[altIdx]
+                      );
+            }
           }
         }
-      }
 
-      fprintf(t->oDebug->file(), "\n");
-    }
+        fprintf(t->oDebug->file(), "\n");
+      }
+    }  //  End of debug dump.
 
     //  Generate output VCFs.
 

diff --git a/src/merfin/merfin.C b/src/merfin/merfin.C
@@ -108,14 +108,6 @@ main(int32 argc, char **argv) {
     } else if (strcmp(argv[arg], "-memory") == 0) {
       G->maxMemory = strtodouble(argv[++arg]);
 
-#if 0
-    } else if (strcmp(argv[arg], "-memory1") == 0) {
-      G->maxMemory += strtodouble(argv[++arg]);
-
-    } else if (strcmp(argv[arg], "-memory2") == 0) {
-      G->maxMemory += strtodouble(argv[++arg]);
-#endif
-
     } else if (strcmp(argv[arg], "-nosplit") == 0) {
       G->nosplit = true;
 
@@ -140,6 +132,9 @@ main(int32 argc, char **argv) {
     } else if (strcmp(argv[arg], "-comb") == 0) {
       G->comb = strtouint32(argv[++arg]);
 
+    } else if (strcmp(argv[arg], "-debug") == 0) {
+      G->debug = true;
+
     } else {
       char *s = new char [1024];
       snprintf(s, 1024, "Unknown option '%s'.\n", argv[arg]);
@@ -233,6 +228,7 @@ main(int32 argc, char **argv) {
     //fprintf(stderr, "                        if chosen, use bcftools to compress and index, and consensus -H 1 -f <seq.fata> to polish.\n");
     //fprintf(stderr, "                        first ALT in heterozygous alleles are better supported by avg. |k*|.\n");
     fprintf(stderr, "             -lookup <probabilities> use probabilities to adjust multiplicity to copy number\n");
+    fprintf(stderr, "             -debug     output a debug log, into <output>.THREAD_ID.debug.gz\n");
     fprintf(stderr, "\n");
     fprintf(stderr, "  -completeness\n");
     fprintf(stderr, "   Compute kmer completeness using expected copy numbers for all kmers.\n");
-Original file line number
+Diff line change
@@ Expand Up / @@ -229,9 +229,7 @@ public: @@
       uint32                  reportType    = OP_NONE;
+      bool                    debug         = false;
     public:
       double                  peak          = 0;
@@ Expand Down @@