Be able to export a full list of media clusters.

lib/check_search.rb

@@ -335,7 +335,7 @@ def medias_get_search_result(query)
 
   def self.get_exported_data(query, team_id)
     team = Team.find(team_id)
-    search = CheckSearch.new(query, nil, team_id)
+    Team.current = team
 
     # Prepare the export
     data = []
@@ -344,31 +344,39 @@ def self.get_exported_data(query, team_id)
     fields.each { |tt| header << tt.label }
     data << header
 
-    # No pagination for the export
-    search.set_option('esoffset', 0)
-    search.set_option('eslimit', CheckConfig.get(:export_csv_maximum_number_of_results, 10000, :integer))
-
-    # Iterate through each result and generate an output row for the CSV
-    search.medias.find_each do |pm|
-      row = [
-        pm.claim_description&.description,
-        pm.full_url,
-        pm.status_i18n,
-        pm.author_name.to_s.gsub(/ \[.*\]$/, ''),
-        pm.created_at.strftime("%Y-%m-%d %H:%M:%S"),
-        pm.published_at&.strftime("%Y-%m-%d %H:%M:%S"),
-        pm.linked_items_count,
-        pm.tags_as_sentence
-      ]
-      annotations = pm.get_annotations('task').map(&:load)
-      fields.each do |field|
-        annotation = annotations.find { |a| a.team_task_id == field.id }
-        answer = (annotation ? (begin annotation.first_response_obj.file_data[:file_urls].join("\n") rescue annotation.first_response.to_s end) : '')
-        answer = begin JSON.parse(answer).collect{ |x| x['url'] }.join(', ') rescue answer end
-        row << answer
+    # Paginate
+    search = CheckSearch.new(query, nil, team_id)
+    search_after = [0]
+    while true
+      result = $repository.search(_source: 'annotated_id', query: search.medias_query, sort: [{ annotated_id: { order: :asc } }], size: 10000, search_after: search_after).results
+      ids = result.collect{ |i| i['annotated_id'] }.uniq.map(&:to_i)
+      break if ids.empty?
+
+      # Iterate through each result and generate an output row for the CSV
+      ProjectMedia.where(id: ids, team_id: search.team_condition(team_id)).find_each do |pm|
+        row = [
+          pm.claim_description&.description,
+          pm.full_url,
+          pm.status_i18n,
+          pm.author_name.to_s.gsub(/ \[.*\]$/, ''),
+          pm.created_at.strftime("%Y-%m-%d %H:%M:%S"),
+          pm.published_at&.strftime("%Y-%m-%d %H:%M:%S"),
+          pm.linked_items_count,
+          pm.tags_as_sentence
+        ]
+        annotations = pm.get_annotations('task').map(&:load)
+        fields.each do |field|
+          annotation = annotations.find { |a| a.team_task_id == field.id }
+          answer = (annotation ? (begin annotation.first_response_obj.file_data[:file_urls].join("\n") rescue annotation.first_response.to_s end) : '')
+          answer = begin JSON.parse(answer).collect{ |x| x['url'] }.join(', ') rescue answer end
+          row << answer
+        end
+        data << row
       end
-      data << row
+
+      search_after = [ids.max]
     end
+
     data
   end
 

test/lib/list_export_test.rb

@@ -26,18 +26,23 @@ def teardown
     end
   end
 
-  test "should export media CSV" do
+  test "should export media (including child media) CSV" do
+    setup_elasticsearch
     t = create_team
     create_team_task team_id: t.id, fieldset: 'tasks'
-    2.times { create_project_media team: t }
+    parent = create_project_media team: t, disable_es_callbacks: false
+    child = create_project_media team: t, disable_es_callbacks: false
+    create_relationship source_id: parent.id, target_id: child.id, relationship_type: Relationship.confirmed_type
 
-    export = ListExport.new(:media, '{}', t.id)
+    sleep 2 # Wait for indexing
+
+    export = ListExport.new(:media, { show_similar: true }.to_json, t.id)
     csv_url = export.generate_csv_and_send_email(create_user)
     response = Net::HTTP.get_response(URI(csv_url))
     assert_equal 200, response.code.to_i
     csv_content = CSV.parse(response.body, headers: true)
-    assert_equal 2, csv_content.size
     assert_equal 2, export.number_of_rows
+    assert_equal 2, csv_content.size
   end
 
   test "should export feed CSV" do