Improve how the E.M type is detected to match fair data station properties

lib/seek/fair_data_station/writer.rb

@@ -211,12 +211,14 @@ def detect_extended_metadata_type(seek_entity, datastation_entity)
 
         # collect and sort those with the most properties that match, eliminating any where no properties match
         candidates = ::ExtendedMetadataType.where(supported_type: seek_entity.class.name).includes(:extended_metadata_attributes).collect do |emt|
-          pids = collect_extended_metadata_type_attibute_pids(emt)
-          score = (property_ids - pids).length
-          emt = nil if (property_ids & pids).empty?
-          [score, emt]
+          extended_metadata_property_ids = collect_extended_metadata_type_attibute_pids(emt)
+          intersection = (property_ids & extended_metadata_property_ids)
+          difference = (property_ids | extended_metadata_property_ids) - intersection
+          emt = nil if intersection.length.zero?
+          [intersection.length, difference.length, emt]
         end.sort_by do |x|
-          x[0]
+          # order by the number of properties matched coming top, but downgraded by the number of differences
+          [-x[0], x[1]]
         end
 
         candidates.first&.last

test/factories/extended_metadata_types.rb

@@ -468,6 +468,7 @@
     supported_type { 'Assay' }
     after(:build) do |a|
       a.extended_metadata_attributes << FactoryBot.create(:study_title_extended_metadata_attribute, title: 'Facility', pid:'http://fairbydesign.nl/ontology/facility')
+      a.extended_metadata_attributes << FactoryBot.create(:study_title_extended_metadata_attribute, title: 'Protocol', pid:'http://fairbydesign.nl/ontology/protocol')
       a.extended_metadata_attributes << FactoryBot.create(:study_title_extended_metadata_attribute, title: 'Forward Primer', pid:'http://fairbydesign.nl/ontology/forwardPrimer')
       a.extended_metadata_attributes << FactoryBot.create(:study_title_extended_metadata_attribute, title: 'Instrument Model', pid:'http://fairbydesign.nl/ontology/instrument_model')
       a.extended_metadata_attributes << FactoryBot.create(:study_title_extended_metadata_attribute, title: 'Library Selection', pid:'http://fairbydesign.nl/ontology/library_selection')

test/unit/fair_data_station/fair_data_station_writer_test.rb

@@ -125,6 +125,7 @@ class FairDataStationWriterTest < ActiveSupport::TestCase
     assert_equal assay_metadata_type, assay.extended_metadata.extended_metadata_type
 
     assert_equal 'NMIMR', assay.extended_metadata.get_attribute_value('Facility')
+    assert_equal 'MiSeq Reagent Kit v3 (600-cycle) with a 20% PhiX (Illumina) spike-in', assay.extended_metadata.get_attribute_value('Protocol')
     assert_equal 'CCTACGGGNGGCWGCAG', assay.extended_metadata.get_attribute_value('Forward Primer')
     assert_equal 'Illumina MiSeq', assay.extended_metadata.get_attribute_value('Instrument Model')
     assert_equal 'PCR', assay.extended_metadata.get_attribute_value('Library Selection')
@@ -685,6 +686,38 @@ class FairDataStationWriterTest < ActiveSupport::TestCase
 
   end
 
+  test 'detect extended metadata type' do
+    virtual_demo_assay = FactoryBot.create(:fairdata_virtual_demo_assay_extended_metadata)
+    seek_test_case_assay = FactoryBot.create(:fairdata_test_case_assay_extended_metadata)
+    FactoryBot.create(:simple_assay_extended_metadata_type)
+    FactoryBot.create(:fairdata_test_case_obsv_unit_extended_metadata)
+    FactoryBot.create(:simple_observation_unit_extended_metadata_type)
+    writer = Seek::FairDataStation::Writer.new
+
+    path = "#{Rails.root}/test/fixtures/files/fair_data_station/seek-fair-data-station-test-case.ttl"
+    inv = Seek::FairDataStation::Reader.new.parse_graph(path).first
+    assay = inv.studies.first.assays.first
+    seek_assay = ::Assay.new
+    detected_type = writer.send(:detect_extended_metadata_type, seek_assay, assay)
+    assert_equal seek_test_case_assay, detected_type
+
+    path = "#{Rails.root}/test/fixtures/files/fair_data_station/demo.ttl"
+    inv = Seek::FairDataStation::Reader.new.parse_graph(path).first
+    assay = inv.studies.first.assays.first
+    detected_type = writer.send(:detect_extended_metadata_type, seek_assay, assay)
+    assert_equal virtual_demo_assay, detected_type
+
+    path = "#{Rails.root}/test/fixtures/files/fair_data_station/indpensim.ttl"
+    inv = Seek::FairDataStation::Reader.new.parse_graph(path).first
+    obs_unit = inv.studies.first.observation_units.first
+    detected_type = writer.send(:detect_extended_metadata_type, ::ObservationUnit.new, obs_unit)
+    assert_nil detected_type
+
+    inpensim_obs_unit = FactoryBot.create(:fairdata_indpensim_obsv_unit_extended_metadata)
+    detected_type = writer.send(:detect_extended_metadata_type, ::ObservationUnit.new, obs_unit)
+    assert_equal inpensim_obs_unit, detected_type
+  end
+
   private
 
   def setup_test_case_investigation