Add already_seen flag to classification_events (#80)

- Migration to add already_seen column to classification_events
- Update fixture to a more recent looking classification payload coming from kinesis stream
(Prior fixture is from 2016, what we send to the kinesis stream from panoptes has updated plenty since then)
- Fetch already_seen from metadata on stream_events

app/models/stream_events.rb

@@ -46,4 +46,8 @@ def self.session_time(event_data)
     diff = SESSION_CAP if diff > SESSION_LIMIT
     diff
   end
+
+  def self.already_seen(event_data)
+    event_data.fetch('metadata')&.fetch('subject_selection_state', nil)&.fetch('already_seen', nil)
+  end
 end

app/models/stream_events/classification.rb

@@ -18,7 +18,8 @@ def process
         workflow_id: @links.fetch('workflow'),
         user_id: @links&.fetch('user', nil),
         user_group_ids: @data.fetch('metadata')&.fetch('user_group_ids', []),
-        session_time: StreamEvents.session_time(@data)
+        session_time: StreamEvents.session_time(@data),
+        already_seen: StreamEvents.already_seen(@data)
       }
     end
   end

db/migrate/20241127220919_add_already_seen_to_classification_events.rb

@@ -0,0 +1,5 @@
+class AddAlreadySeenToClassificationEvents < ActiveRecord::Migration[7.0]
+  def change
+    add_column :classification_events, :already_seen, :boolean
+  end
+end

db/schema.rb

@@ -10,7 +10,7 @@
 #
 # It's strongly recommended that you check this file into your version control system.
 
-ActiveRecord::Schema[7.0].define(version: 2024_09_26_233924) do
+ActiveRecord::Schema[7.0].define(version: 2024_11_27_220919) do
   # These are extensions that must be enabled in order to support this database
   enable_extension "plpgsql"
   enable_extension "timescaledb"
@@ -28,6 +28,7 @@
     t.float "session_time"
     t.datetime "created_at", null: false
     t.datetime "updated_at", null: false
+    t.boolean "already_seen"
     t.index ["event_time"], name: "classification_events_event_time_idx", order: :desc
     t.index ["user_id"], name: "index_classification_events_on_user_id"
   end

spec/models/stream_events/classification_spec.rb

@@ -23,6 +23,7 @@
       expect(prepared_payload[:user_id]).to eq(links.fetch('user'))
       expect(prepared_payload[:user_group_ids]).to eq(data.fetch('metadata').fetch('user_group_ids'))
       expect(prepared_payload[:session_time]).to eq(StreamEvents.session_time(data))
+      expect(prepared_payload[:already_seen]).to eq(StreamEvents.already_seen(data))
     end
 
     it 'sets the user_id to nil if no user in stream payload' do

spec/models/stream_events_spec.rb

@@ -111,6 +111,23 @@
         expect(StreamEvents.session_time(event_data)).to be_zero
       end
     end
+
+    context 'already_seen' do
+      it 'returns already_seen from payload if provided' do
+        expected_already_seen = event_metadata['subject_selection_state']['already_seen']
+        expect(StreamEvents.already_seen(event_data)).to eq(expected_already_seen)
+      end
+
+      it 'returns nil if subject_selection_state is not provided' do
+        event_metadata.delete('subject_selection_state')
+        expect(StreamEvents.already_seen(event_data)).to eq(nil)
+      end
+
+      it 'returns nil if subject_selection_state does not have already_seen' do
+        event_metadata['subject_selection_state'] = {}
+        expect(StreamEvents.already_seen(event_data)).to eq(nil)
+      end
+    end
   end
 
   describe 'unknown event' do