[MISC] Update layout- and add build-subnamespace

include/hibf/build/bin_size_in_bits.hpp

@@ -17,7 +17,7 @@
 
 #include <hibf/platform.hpp>
 
-namespace seqan::hibf
+namespace seqan::hibf::build
 {
 
 struct bin_size_parameters
@@ -35,4 +35,4 @@ inline size_t bin_size_in_bits(bin_size_parameters const & params)
     return result;
 }
 
-} // namespace seqan::hibf
+} // namespace seqan::hibf::build

include/hibf/build/build_data.hpp

@@ -15,7 +15,7 @@
 #include <hibf/detail/timer.hpp> // for concurrent, timer
 #include <hibf/layout/graph.hpp> // for node_data
 
-namespace seqan::hibf
+namespace seqan::hibf::build
 {
 
 struct build_data
@@ -40,4 +40,4 @@ struct build_data
     }
 };
 
-} // namespace seqan::hibf
+} // namespace seqan::hibf::build

include/hibf/build/chopper_pack_record.hpp

@@ -13,7 +13,7 @@
 
 #include <hibf/platform.hpp>
 
-namespace seqan::hibf
+namespace seqan::hibf::build
 {
 
 struct chopper_pack_record
@@ -45,4 +45,4 @@ struct chopper_pack_record
     }
 };
 
-} // namespace seqan::hibf
+} // namespace seqan::hibf::build

include/hibf/build/compute_kmers.hpp

@@ -18,11 +18,11 @@
 #include <hibf/contrib/robin_hood.hpp> // for unordered_flat_set
 #include <hibf/layout/layout.hpp>      // for layout
 
-namespace seqan::hibf
+namespace seqan::hibf::build
 {
 
 void compute_kmers(robin_hood::unordered_flat_set<uint64_t> & kmers,
                    build_data const & data,
                    layout::layout::user_bin const & record);
 
-} // namespace seqan::hibf
+} // namespace seqan::hibf::build

include/hibf/build/construct_ibf.hpp

@@ -15,7 +15,7 @@
 #include <hibf/interleaved_bloom_filter.hpp> // for interleaved_bloom_filter
 #include <hibf/layout/graph.hpp>
 
-namespace seqan::hibf
+namespace seqan::hibf::build
 {
 
 seqan::hibf::interleaved_bloom_filter construct_ibf(robin_hood::unordered_flat_set<uint64_t> & parent_kmers,
@@ -25,4 +25,4 @@ seqan::hibf::interleaved_bloom_filter construct_ibf(robin_hood::unordered_flat_s
                                                     build_data & data,
                                                     bool is_root);
 
-} // namespace seqan::hibf
+} // namespace seqan::hibf::build

include/hibf/build/insert_into_ibf.hpp

@@ -16,7 +16,7 @@
 #include <hibf/interleaved_bloom_filter.hpp> // for interleaved_bloom_filter
 #include <hibf/layout/layout.hpp>            // for layout
 
-namespace seqan::hibf
+namespace seqan::hibf::build
 {
 
 // automatically does naive splitting if number_of_bins > 1
@@ -30,4 +30,4 @@ void insert_into_ibf(build_data const & data,
                      layout::layout::user_bin const & record,
                      seqan::hibf::interleaved_bloom_filter & ibf);
 
-} // namespace seqan::hibf
+} // namespace seqan::hibf::build

include/hibf/build/update_parent_kmers.hpp

@@ -17,7 +17,7 @@
 #include <hibf/detail/timer.hpp>       // for concurrent, timer
 #include <hibf/platform.hpp>
 
-namespace seqan::hibf
+namespace seqan::hibf::build
 {
 
 inline void update_parent_kmers(robin_hood::unordered_flat_set<uint64_t> & parent_kmers,
@@ -31,4 +31,4 @@ inline void update_parent_kmers(robin_hood::unordered_flat_set<uint64_t> & paren
     merge_kmers_timer += local_merge_kmers_timer;
 }
 
-} // namespace seqan::hibf
+} // namespace seqan::hibf::build

include/hibf/build/update_user_bins.hpp

@@ -13,12 +13,12 @@
 
 #include <hibf/layout/layout.hpp> // for layout
 
-namespace seqan::hibf
+namespace seqan::hibf::build
 {
 
 inline void update_user_bins(std::vector<int64_t> & filename_indices, layout::layout::user_bin const & record)
 {
     std::fill_n(filename_indices.begin() + record.storage_TB_id, record.number_of_technical_bins, record.idx);
 }
 
-} // namespace seqan::hibf
+} // namespace seqan::hibf::build

include/hibf/layout/data_store.hpp

@@ -11,7 +11,7 @@
 #include <hibf/layout/layout.hpp>      // for layout
 #include <hibf/sketch/hyperloglog.hpp> // for hyperloglog
 
-namespace seqan::hibf
+namespace seqan::hibf::layout
 {
 
 struct data_store
@@ -41,7 +41,7 @@ struct data_store
     double false_positive_rate{};
 
     //!\brief The layout that is built by layout::hierarchical_binning.
-    layout::layout * hibf_layout; // Will be modified by {simple,hierarchical}_binning.
+    layout * hibf_layout; // Will be modified by {simple,hierarchical}_binning.
 
     //!\brief The kmer counts associated with the above files used to layout user bin into technical bins.
     std::vector<size_t> const * kmer_counts{}; // Pointed to data should not be modified.
@@ -79,4 +79,4 @@ struct data_store
     //!\}
 };
 
-} // namespace seqan::hibf
+} // namespace seqan::hibf::layout

include/hibf/layout/execute.hpp

@@ -5,9 +5,9 @@
 #include <hibf/config.hpp>            // for config
 #include <hibf/layout/data_store.hpp> // for data_store
 
-namespace seqan::hibf
+namespace seqan::hibf::layout
 {
 
 size_t execute(config const &, data_store &);
 
-} // namespace seqan::hibf
+} // namespace seqan::hibf::layout

include/hibf/layout/print_matrix.hpp

@@ -4,7 +4,7 @@
 
 #include <hibf/platform.hpp>
 
-namespace seqan::hibf
+namespace seqan::hibf::layout
 {
 
 // helper function to print a matrix when debugging
@@ -28,4 +28,4 @@ void print_matrix(matrix_type const & matrix,
     std::cerr << '\n';
 }
 
-} // namespace seqan::hibf
+} // namespace seqan::hibf::layout

src/build/compute_kmers.cpp

@@ -21,7 +21,7 @@
 #include <hibf/detail/timer.hpp>        // for concurrent, timer
 #include <hibf/layout/layout.hpp>       // for layout
 
-namespace seqan::hibf
+namespace seqan::hibf::build
 {
 
 void compute_kmers(robin_hood::unordered_flat_set<uint64_t> & kmers,
@@ -35,4 +35,4 @@ void compute_kmers(robin_hood::unordered_flat_set<uint64_t> & kmers,
     data.user_bin_io_timer += local_user_bin_io_timer;
 }
 
-} // namespace seqan::hibf
+} // namespace seqan::hibf::build

src/build/construct_ibf.cpp

@@ -21,7 +21,7 @@
 #include <hibf/interleaved_bloom_filter.hpp>  // for interleaved_bloom_filter, bin_count, bin_size, hash_fun...
 #include <hibf/layout/graph.hpp>
 
-namespace seqan::hibf
+namespace seqan::hibf::build
 {
 
 seqan::hibf::interleaved_bloom_filter construct_ibf(robin_hood::unordered_flat_set<uint64_t> & parent_kmers,
@@ -54,4 +54,4 @@ seqan::hibf::interleaved_bloom_filter construct_ibf(robin_hood::unordered_flat_s
     return ibf;
 }
 
-} // namespace seqan::hibf
+} // namespace seqan::hibf::build

src/build/insert_into_ibf.cpp

@@ -22,7 +22,7 @@
 #include <hibf/interleaved_bloom_filter.hpp>        // for interleaved_bloom_filter, bin_index
 #include <hibf/layout/layout.hpp>                   // for layout
 
-namespace seqan::hibf
+namespace seqan::hibf::build
 {
 
 // automatically does naive splitting if number_of_bins > 1
@@ -70,4 +70,4 @@ void insert_into_ibf(build_data const & data,
     data.fill_ibf_timer += local_fill_ibf_timer;
 }
 
-} // namespace seqan::hibf
+} // namespace seqan::hibf::build

src/hierarchical_interleaved_bloom_filter.cpp

@@ -36,7 +36,7 @@
 size_t hierarchical_build(hierarchical_interleaved_bloom_filter & hibf,
                           robin_hood::unordered_flat_set<uint64_t> & parent_kmers,
                           layout::graph::node const & current_node,
-                          build_data & data,
+                          build::build_data & data,
                           bool is_root)
 {
     size_t const ibf_pos{data.request_ibf_idx()};
@@ -62,8 +62,8 @@
         {
             // we assume that the max record is at the beginning of the list of remaining records.
             auto const & record = current_node.remaining_records[0];
-            compute_kmers(kmers, data, record);
-            update_user_bins(filename_indices, record);
+            build::compute_kmers(kmers, data, record);
+            build::update_user_bins(filename_indices, record);
 
             return record.number_of_technical_bins;
         }
@@ -117,9 +117,9 @@
                 size_t const mutex_id{parent_bin_index / 64};
                 std::lock_guard<std::mutex> guard{local_ibf_mutex[mutex_id]};
                 ibf_positions[parent_bin_index] = ibf_pos;
-                insert_into_ibf(kmers, 1, parent_bin_index, ibf, data.fill_ibf_timer);
+                build::insert_into_ibf(kmers, 1, parent_bin_index, ibf, data.fill_ibf_timer);
                 if (!is_root)
-                    update_parent_kmers(parent_kmers, kmers, data.merge_kmers_timer);
+                    build::update_parent_kmers(parent_kmers, kmers, data.merge_kmers_timer);
             }
         }
     };
@@ -134,17 +134,21 @@
 
         if (is_root && record.number_of_technical_bins == 1) // no splitting needed
         {
-            insert_into_ibf(data, record, ibf);
+            build::insert_into_ibf(data, record, ibf);
         }
         else
         {
             compute_kmers(kmers, data, record);
-            insert_into_ibf(kmers, record.number_of_technical_bins, record.storage_TB_id, ibf, data.fill_ibf_timer);
+            build::insert_into_ibf(kmers,
+                                   record.number_of_technical_bins,
+                                   record.storage_TB_id,
+                                   ibf,
+                                   data.fill_ibf_timer);
             if (!is_root)
-                update_parent_kmers(parent_kmers, kmers, data.merge_kmers_timer);
+                build::update_parent_kmers(parent_kmers, kmers, data.merge_kmers_timer);
         }
 
-        update_user_bins(filename_indices, record);
+        build::update_user_bins(filename_indices, record);
         kmers.clear();
     }
 
@@ -157,7 +161,7 @@
 
 size_t hierarchical_build(hierarchical_interleaved_bloom_filter & hibf,
                           layout::graph::node const & root_node,
-                          build_data & data)
+                          build::build_data & data)
 {
     robin_hood::unordered_flat_set<uint64_t> root_kmers{};
     return hierarchical_build(hibf, root_kmers, root_node, data, true);
@@ -174,7 +178,7 @@
     hibf.user_bins.set_user_bin_count(hibf_layout.user_bins.size());
     hibf.next_ibf_id.resize(number_of_ibfs);
 
-    build_data data{.config = config, .ibf_graph = {hibf_layout}};
+    build::build_data data{.config = config, .ibf_graph = {hibf_layout}};
 
     layout::graph::node const & root_node = data.ibf_graph.root;
 

src/layout/compute_layout.cpp

@@ -55,7 +55,7 @@ compute_layout(config const & config, std::vector<size_t> & kmer_counts, std::ve
                      .kmer_counts = std::addressof(kmer_counts),
                      .sketches = std::addressof(sketches)};
 
-    size_t const max_hibf_id = seqan::hibf::execute(config, store);
+    size_t const max_hibf_id = seqan::hibf::layout::execute(config, store);
     store.hibf_layout->top_level_max_bin_id = max_hibf_id;
 
     // sort records ascending by the number of bin indices (corresponds to the IBF levels)

src/layout/execute.cpp

@@ -12,10 +12,10 @@
 #include <hibf/layout/hierarchical_binning.hpp>   // for hierarchical_binning
 #include <hibf/next_multiple_of_64.hpp>           // for next_multiple_of_64
 
-namespace seqan::hibf
+namespace seqan::hibf::layout
 {
 
-size_t execute(seqan::hibf::config const & config, seqan::hibf::data_store & data)
+size_t execute(seqan::hibf::config const & config, seqan::hibf::layout::data_store & data)
 {
     seqan::hibf::config config_copy{config};
 
@@ -41,11 +41,11 @@ size_t execute(seqan::hibf::config const & config, seqan::hibf::data_store & dat
     }
 
     data.fpr_correction =
-        layout::compute_fpr_correction({.fpr = config_copy.maximum_false_positive_rate, // prevent clang-format
-                                        .hash_count = config_copy.number_of_hash_functions,
-                                        .t_max = config_copy.tmax});
+        compute_fpr_correction({.fpr = config_copy.maximum_false_positive_rate, // prevent clang-format
+                                .hash_count = config_copy.number_of_hash_functions,
+                                .t_max = config_copy.tmax});
 
     return seqan::hibf::layout::hierarchical_binning{data, config_copy}.execute();
 }
 
-} // namespace seqan::hibf
+} // namespace seqan::hibf::layout

test/unit/hibf/layout/hierarchical_binning_test.cpp

@@ -19,7 +19,7 @@ TEST(hierarchical_binning_test, small_example)
     seqan::hibf::layout::layout hibf_layout{};
     std::vector<size_t> kmer_counts{500, 1000, 500, 500, 500, 500, 500, 500};
 
-    seqan::hibf::data_store data{.hibf_layout = &hibf_layout, .kmer_counts = &kmer_counts};
+    seqan::hibf::layout::data_store data{.hibf_layout = &hibf_layout, .kmer_counts = &kmer_counts};
 
     data.fpr_correction =
         seqan::hibf::layout::compute_fpr_correction({.fpr = 0.05, .hash_count = 2, .t_max = config.tmax});
@@ -49,7 +49,7 @@ TEST(hierarchical_binning_test, another_example)
 
     seqan::hibf::layout::layout hibf_layout{};
     std::vector<size_t> kmer_counts{50, 1000, 1000, 50, 5, 10, 10, 5};
-    seqan::hibf::data_store data{.hibf_layout = &hibf_layout, .kmer_counts = &kmer_counts};
+    seqan::hibf::layout::data_store data{.hibf_layout = &hibf_layout, .kmer_counts = &kmer_counts};
 
     data.fpr_correction =
         seqan::hibf::layout::compute_fpr_correction({.fpr = 0.05, .hash_count = 2, .t_max = config.tmax});
@@ -80,7 +80,7 @@ TEST(hierarchical_binning_test, high_level_max_bin_id_is_0)
 
     seqan::hibf::layout::layout hibf_layout{};
     std::vector<size_t> kmer_counts{500, 500, 500, 500};
-    seqan::hibf::data_store data{.hibf_layout = &hibf_layout, .kmer_counts = &kmer_counts};
+    seqan::hibf::layout::data_store data{.hibf_layout = &hibf_layout, .kmer_counts = &kmer_counts};
 
     data.fpr_correction =
         seqan::hibf::layout::compute_fpr_correction({.fpr = 0.05, .hash_count = 2, .t_max = config.tmax});
@@ -105,7 +105,7 @@ TEST(hierarchical_binning_test, knuts_example)
 
     seqan::hibf::layout::layout hibf_layout{};
     std::vector<size_t> kmer_counts{60, 600, 1000, 800, 800};
-    seqan::hibf::data_store data{.hibf_layout = &hibf_layout, .kmer_counts = &kmer_counts};
+    seqan::hibf::layout::data_store data{.hibf_layout = &hibf_layout, .kmer_counts = &kmer_counts};
 
     data.fpr_correction =
         seqan::hibf::layout::compute_fpr_correction({.fpr = 0.05, .hash_count = 2, .t_max = config.tmax});
@@ -133,7 +133,7 @@ TEST(hierarchical_binning_test, four_level_hibf)
 
     seqan::hibf::layout::layout hibf_layout{};
     std::vector<size_t> kmer_counts{11090, 5080, 3040, 1020, 510, 500};
-    seqan::hibf::data_store data{.hibf_layout = &hibf_layout, .kmer_counts = &kmer_counts};
+    seqan::hibf::layout::data_store data{.hibf_layout = &hibf_layout, .kmer_counts = &kmer_counts};
 
     data.fpr_correction =
         seqan::hibf::layout::compute_fpr_correction({.fpr = 0.05, .hash_count = 2, .t_max = config.tmax});
@@ -166,7 +166,7 @@ TEST(hierarchical_binning_test, tb0_is_a_merged_bin)
 
     seqan::hibf::layout::layout hibf_layout{};
     std::vector<size_t> kmer_counts{500, 500, 500, 500};
-    seqan::hibf::data_store data{.hibf_layout = &hibf_layout, .kmer_counts = &kmer_counts};
+    seqan::hibf::layout::data_store data{.hibf_layout = &hibf_layout, .kmer_counts = &kmer_counts};
 
     data.fpr_correction =
         seqan::hibf::layout::compute_fpr_correction({.fpr = 0.05, .hash_count = 2, .t_max = config.tmax});
@@ -194,7 +194,7 @@ TEST(hierarchical_binning_test, tb0_is_a_merged_bin_and_leads_to_recursive_call)
 
     seqan::hibf::layout::layout hibf_layout{};
     std::vector<size_t> kmer_counts{500, 500, 500, 500, 500, 500, 500, 500};
-    seqan::hibf::data_store data{.hibf_layout = &hibf_layout, .kmer_counts = &kmer_counts};
+    seqan::hibf::layout::data_store data{.hibf_layout = &hibf_layout, .kmer_counts = &kmer_counts};
 
     data.fpr_correction =
         seqan::hibf::layout::compute_fpr_correction({.fpr = 0.05, .hash_count = 2, .t_max = config.tmax});