Fix gzstream compilation, update on HLL

src/Makefile

@@ -1,9 +1,10 @@
 CXX = g++
 FOPENMP?=-fopenmp
-CXXFLAGS = -Wall -std=c++0x $(FOPENMP) -O2 -Wfatal-errors ${CPPFLAGS}
+CXXFLAGS = -Wall -std=c++0x $(FOPENMP) -I./gzstream -O2 -Wfatal-errors ${CPPFLAGS}
 #CXXFLAGS = -Wall -std=c++11 $(FOPENMP) -O3 -Wfatal-errors
 PROGS = classify db_sort set_lcas make_seqid_to_taxid_map db_shrink build_taxdb grade_classification dump_taxdb read_uid_mapping
-LIBFLAGS = -L. -I./gzstream -L./gzstream -lz -lgzstream ${LDFLAGS}
+#LIBFLAGS = -L. -lz -lgzstream ${LDFLAGS}
+LIBFLAGS = -L. -lz ${LDFLAGS}
 
 .PHONY: all install clean
 
@@ -25,7 +26,7 @@ grade_classification: taxdb.h report-cols.h
 
 read_uid_mapping: quickfile.o
 
-classify: classify.cpp krakendb.o quickfile.o krakenutil.o seqreader.o uid_mapping.o hyperloglogplus.h taxdb.h report-cols.h
+classify: classify.cpp krakendb.o quickfile.o krakenutil.o seqreader.o uid_mapping.o gzstream.o hyperloglogplus.h taxdb.h report-cols.h
 	$(CXX) $(CXXFLAGS) -o classify $^ $(LIBFLAGS)
 
 build_taxdb: taxdb.h report-cols.h quickfile.o
@@ -43,6 +44,9 @@ krakendb.o: krakendb.cpp krakendb.hpp quickfile.hpp
 seqreader.o: seqreader.cpp seqreader.hpp quickfile.hpp
 	$(CXX) $(CXXFLAGS) -c seqreader.cpp
 
+gzstream.o: gzstream/gzstream.C gzstream/gzstream.h
+	$(CXX) $(CXXFLAGS) -c -O gzstream/gzstream.C
+
 quickfile.o: quickfile.cpp quickfile.hpp
 	$(CXX) $(CXXFLAGS) -c quickfile.cpp
 

src/classify.cpp

@@ -114,19 +114,19 @@ ostream* cout_or_file(string file) {
 }
 
 void loadKrakenDB(KrakenDB& database, string DB_filename, string Index_filename) {
-	QuickFile db_file;
-	db_file.open_file(DB_filename);
-	if (Populate_memory) {
-		db_file.load_file();
-	}
-	database = KrakenDB(db_file.ptr());
-	QuickFile idx_file;
-	idx_file.open_file(Index_filename);
-	if (Populate_memory)
-		idx_file.load_file();
-
-	KrakenDBIndex db_index(idx_file.ptr());
-	database.set_index(&db_index);
+  QuickFile db_file;
+  db_file.open_file(DB_filename);
+  if (Populate_memory) {
+    db_file.load_file();
+  }
+  database = KrakenDB(db_file.ptr());
+  QuickFile idx_file;
+  idx_file.open_file(Index_filename);
+  if (Populate_memory)
+    idx_file.load_file();
+
+  KrakenDBIndex db_index(idx_file.ptr());
+  database.set_index(&db_index);
 }
 
 int main(int argc, char **argv) {
@@ -242,18 +242,18 @@ int main(int argc, char **argv) {
       taxdb.readGenomeSizes(fname);
     }
 
-	TaxReport<uint32_t,ReadCounts> rep = TaxReport<uint32_t, ReadCounts>(*Report_output, taxdb, taxon_counts, false);
-	rep.setReportCols(vector<string> { 
-		"%",
-		"reads", 
+  TaxReport<uint32_t,ReadCounts> rep = TaxReport<uint32_t, ReadCounts>(*Report_output, taxdb, taxon_counts, false);
+  rep.setReportCols(vector<string> { 
+    "%",
+    "reads", 
     "taxReads",
     "kmers",
     "dup",
     "cov", 
-		"taxID", 
-		"rank", 
-		"taxName"});
-	rep.printReport("kraken","blu");
+    "taxID", 
+    "rank", 
+    "taxName"});
+  rep.printReport("kraken","blu");
   }
 
   for (size_t i = 0; i < Open_fstreams.size(); ++i) {
@@ -367,11 +367,11 @@ void process_file(char *filename) {
 
 inline
 uint32_t get_taxon_for_kmer(KrakenDB& database, uint64_t* kmer_ptr, uint64_t& current_bin_key,
-		int64_t& current_min_pos, int64_t& current_max_pos) {
-	uint32_t* val_ptr = database.kmer_query(
-			database.canonical_representation(*kmer_ptr), &current_bin_key,
-			&current_min_pos, &current_max_pos);
-	return val_ptr ? *val_ptr : 0;
+    int64_t& current_min_pos, int64_t& current_max_pos) {
+  uint32_t* val_ptr = database.kmer_query(
+      database.canonical_representation(*kmer_ptr), &current_bin_key,
+      &current_min_pos, &current_max_pos);
+  return val_ptr ? *val_ptr : 0;
 }
 
 
@@ -512,10 +512,10 @@ bool classify_sequence(DNASequence &dna, ostringstream &koss,
         taxon = get_taxon_for_kmer(*KrakenDatabases[i], kmer_ptr,
         db_statuses[i].current_bin_key, db_statuses[i].current_min_pos, db_statuses[i].current_max_pos);
 
-	  //uint32_t* val_ptr = KrakenDatabases[i]->kmer_query(
-		//	KrakenDatabases[i]->canonical_representation(*kmer_ptr), &db_statuses[i].current_bin_key,
-		//	&db_statuses[i].current_min_pos, &db_statuses[i].current_max_pos);
-	  //taxon = val_ptr ? *val_ptr : 0;
+        //uint32_t* val_ptr = KrakenDatabases[i]->kmer_query(
+        //  KrakenDatabases[i]->canonical_representation(*kmer_ptr), &db_statuses[i].current_bin_key,
+        //  &db_statuses[i].current_min_pos, &db_statuses[i].current_max_pos);
+        //taxon = val_ptr ? *val_ptr : 0;
           if (taxon) break;
         }
 

src/gzstream/Makefile

@@ -33,7 +33,7 @@
 # ----------------------------------------------------------------------------
 
 # CXX      = CC -n32 -LANG:std   # for SGI Irix 6.5, MIPSpro CC version 7.30
-CXX      = g++-7   # for Linux RedHat 6.1, g++ version 2.95.2
+CXX      = g++   # for Linux RedHat 6.1, g++ version 2.95.2
 
 CPPFLAGS = -I. -O -fPIC
 LDFLAGS  = -L. -lgzstream -lz

src/hyperloglogbias.h

@@ -8,7 +8,9 @@
 #ifndef HYPERLOGLOGBIAS_H_
 #define HYPERLOGLOGBIAS_H_
 
-const double rawEstimateData_precision4[] = {
+// Bias data from https://docs.google.com/document/d/1gyjfMHy43U9OWBXxfaeG-3MjGzejW1dlpyMwEYAAWEI/view?fullscreen
+
+static const double rawEstimateData_precision4[] = {
     11, 11.717, 12.207, 12.7896, 13.2882, 13.8204, 14.3772, 14.9342, 15.5202, 16.161, 16.7722, 17.4636, 18.0396, 18.6766, 19.3566, 20.0454, 20.7936, 21.4856, 22.2666, 22.9946, 23.766, 24.4692, 25.3638, 26.0764, 26.7864, 27.7602, 28.4814, 29.433, 30.2926, 31.0664, 31.9996, 32.7956, 33.5366, 34.5894, 35.5738, 36.2698, 37.3682, 38.0544, 39.2342, 40.0108, 40.7966, 41.9298, 42.8704, 43.6358, 44.5194, 45.773, 46.6772, 47.6174, 48.4888, 49.3304, 50.2506, 51.4996, 52.3824, 53.3078, 54.3984, 55.5838, 56.6618, 57.2174, 58.3514, 59.0802, 60.1482, 61.0376, 62.3598, 62.8078, 63.9744, 64.914, 65.781, 67.1806, 68.0594, 68.8446, 69.7928, 70.8248, 71.8324, 72.8598, 73.6246, 74.7014, 75.393, 76.6708, 77.2394
 };