Add function to read back taxonomy grouping json

lib/genesis/taxonomy/functions/kmer.cpp

@@ -682,6 +682,38 @@ void write_taxonomy_grouping_to_json(
     utils::JsonWriter().write( doc, target );
 }
 
+// --------------------------------------------------------------------------
+//     read_taxonomy_grouping_from_json
+// --------------------------------------------------------------------------
+
+std::vector<TaxonomyGroupData> read_taxonomy_grouping_from_json(
+    std::shared_ptr<utils::BaseInputSource> source
+) {
+    std::vector<TaxonomyGroupData> result;
+    auto doc = utils::JsonReader().read( source );
+    auto& arr = doc.get_array();
+    result.reserve( arr.size() );
+    for( auto& child : arr ) {
+        TaxonomyGroupData elem;
+        elem.group_index = child.at( "group_index" ).get_number_unsigned();
+        if( elem.group_index != result.size() ) {
+            throw std::runtime_error(
+                "Taxonomy grouping json file contains " + std::to_string( result.size() ) +
+                "entries, but with non-consecutive group indices. "
+                "Found group_index" + std::to_string( elem.group_index )
+            );
+        }
+        elem.num_sequences = child.at( "num_sequences" ).get_number_unsigned();
+        elem.sum_seq_lengths = child.at( "sum_seq_lengths" ).get_number_unsigned();
+        for( auto const& tax : child.at( "taxa" ).get_array() ) {
+            elem.taxa.push_back( tax.get_string() );
+        }
+        result.push_back( std::move( elem ));
+        child.clear();
+    }
+    return result;
+}
+
 // --------------------------------------------------------------------------
 //     write_kmer_taxonomy_to_json
 // --------------------------------------------------------------------------

lib/genesis/taxonomy/functions/kmer.hpp

@@ -222,6 +222,29 @@ std::string grouped_taxonomy_report( Taxonomy const& tax );
 //     Json Format
 // =================================================================================================
 
+/**
+ * @brief Data of a single group, as a POD.
+ *
+ * The functions for grouping of a Taxonomy, group_by_taxon_sizes() and
+ * group_with_target_number_of_groups(), produce annotations of the Taxonomy in form of KmerTaxonData
+ * for each Taxon. We can store the data in a json file with write_taxonomy_grouping_to_json(),
+ * which is meant for inspection of the results of grouping.
+ *
+ * When reading back this json data, we however might not always want to reconstruct all data on
+ * the Taxonomy where it came from; instead, for some use cases, we might just be interested in the
+ * data per group, without the underlying Taxonomy. To this end, read_taxonomy_grouping_from_json()
+ * reads such a file again, and produces a vector of elements containing the group data.
+ *
+ * This data structure captures the data there. See write_taxonomy_grouping_to_json() for details.
+ */
+struct TaxonomyGroupData
+{
+    size_t group_index;
+    size_t num_sequences;
+    size_t sum_seq_lengths;
+    std::vector<std::string> taxa;
+};
+
 /**
  * @brief Write the resulting list of groups of taxonomic grouping to json.
  *
@@ -234,6 +257,15 @@ void write_taxonomy_grouping_to_json(
     std::shared_ptr<utils::BaseOutputTarget> target
 );
 
+/**
+ * @brief Read data back in that was written by write_taxonomy_grouping_to_json().
+ *
+ * See TaxonomyGroupData for a description of the resuling POD struct per taxonomic group.
+ */
+std::vector<TaxonomyGroupData> read_taxonomy_grouping_from_json(
+    std::shared_ptr<utils::BaseInputSource> source
+);
+
 /**
  * @brief Write a Taxonomy with KmerTaxonData, in our internal Json format.
  *