basic functionality of reusing files based on xml_id.

Use xml_id_path to recreate proforma structure with tests and model_solutions containing multiple files

app/services/proforma_service/convert_exercise_to_task.rb

@@ -77,62 +77,85 @@ def uuid
     end
 
     def model_solutions
-      @exercise.files.filter {|file| file.role == 'reference_implementation' }.map do |file|
+      @exercise.files.filter {|file| file.role == 'reference_implementation' }.group_by {|file| file.xml_id_path&.split('/')&.first || "co-ms-#{file.id}" }.map do |xml_id, files|
         ProformaXML::ModelSolution.new(
-          id: "ms-#{file.id}",
-          files: model_solution_file(file)
+          id: xml_id || "ms-#{files.first.id}",
+          files: files.map {|file| model_solution_file(file) }
         )
       end
     end
 
     def model_solution_file(file)
-      [
-        task_file(file).tap do |ms_file|
-          ms_file.used_by_grader = false
-          ms_file.usage_by_lms = 'display'
-        end,
-      ]
+      task_file(file).tap do |ms_file|
+        ms_file.used_by_grader = false
+        ms_file.usage_by_lms = 'display'
+      end
     end
 
     def tests
-      @exercise.files.filter(&:teacher_defined_assessment?).map do |file|
+      @exercise.files.filter(&:teacher_defined_assessment?).group_by {|file| xml_id_from_file(file).first }.map do |xml_id, files|
         ProformaXML::Test.new(
-          id: file.id,
-          title: file.name,
-          files: test_file(file),
-          meta_data: test_meta_data(file)
+          id: xml_id || files.first.id,
+          title: files.first.name,
+          files: files.map {|file| test_file(file) },
+          meta_data: test_meta_data(files)
         )
       end
     end
 
-    def test_meta_data(file)
+    def xml_id_from_file(file)
+      type = if file.teacher_defined_assessment?
+               'test'
+             elsif file.reference_implementation?
+               'ms'
+             else
+               'file'
+             end
+      xml_id_path_parts = file.xml_id_path&.split('/')
+      xml_id_path  = []
+
+      xml_id_path << (xml_id_path_parts&.first || "co-#{type}-#{file.id}") unless type == 'file'
+      xml_id_path << (xml_id_path_parts&.last || file.id)
+
+      xml_id_path
+    end
+
+    def test_meta_data(files)
       {
         '@@order' => %w[test-meta-data],
         'test-meta-data' => {
-          '@@order' => %w[CodeOcean:feedback-message CodeOcean:weight CodeOcean:hidden-feedback],
+          '@@order' => %w[CodeOcean:test-file],
           '@xmlns' => {'CodeOcean' => 'codeocean.openhpi.de'},
-          'CodeOcean:feedback-message' => {
-            '@@order' => %w[$1],
-            '$1' => file.feedback_message,
-          },
-          'CodeOcean:weight' => {
-            '@@order' => %w[$1],
-            '$1' => file.weight,
-          },
-          'CodeOcean:hidden-feedback' => {
-            '@@order' => %w[$1],
-            '$1' => file.hidden_feedback,
-          },
+          'CodeOcean:test-file' => files.to_h do |file|
+            [
+              "CodeOcean:#{xml_id_from_file(file).last}", {
+                '@@order' => %w[CodeOcean:feedback-message CodeOcean:weight CodeOcean:hidden-feedback],
+                '@xmlns' => {'CodeOcean' => 'codeocean.openhpi.de'},
+                '@id' =>  xml_id_from_file(file).last,
+                '@name' => file.name,
+                'CodeOcean:feedback-message' => {
+                  '@@order' => %w[$1],
+                  '$1' => file.feedback_message,
+                },
+                'CodeOcean:weight' => {
+                  '@@order' => %w[$1],
+                  '$1' => file.weight,
+                },
+                'CodeOcean:hidden-feedback' => {
+                  '@@order' => %w[$1],
+                  '$1' => file.hidden_feedback,
+                },
+              }
+            ]
+          end,
         },
       }
     end
 
     def test_file(file)
-      [
-        task_file(file).tap do |t_file|
-          t_file.used_by_grader = true
-        end,
-      ]
+      task_file(file).tap do |t_file|
+        t_file.used_by_grader = true
+      end
     end
 
     def exercise_files
@@ -168,8 +191,11 @@ def task_files
     end
 
     def task_file(file)
+      file.update(xml_id_path: xml_id_from_file(file).join('/')) if file.xml_id_path.blank?
+
+      xml_id = xml_id_from_file(file).last
       task_file = ProformaXML::TaskFile.new(
-        id: file.id,
+        id: xml_id,
         filename: filename(file),
         usage_by_lms: file.read_only ? 'display' : 'edit',
         visible: file.hidden ? 'no' : 'yes'

app/services/proforma_service/convert_task_to_exercise.rb

@@ -60,44 +60,56 @@ def string_to_bool(str)
     end
 
     def files
-      model_solution_files + test_files + task_files.values.tap {|array| array.each {|file| file.role ||= 'regular_file' } }
+      model_solution_files + test_files + task_files
     end
 
     def test_files
-      @task.tests.map do |test_object|
-        task_files.delete(test_object.files.first.id).tap do |file|
-          file.weight = extract_meta_data(test_object.meta_data&.dig('test-meta-data'), 'weight').presence || 1.0
-          file.feedback_message = extract_meta_data(test_object.meta_data&.dig('test-meta-data'), 'feedback-message').presence || 'Feedback'
-          file.hidden_feedback = extract_meta_data(test_object.meta_data&.dig('test-meta-data'), 'hidden-feedback').presence || false
-          file.role ||= 'teacher_defined_test'
+      @task.tests.flat_map do |test|
+        test.files.map do |task_file|
+          codeocean_file_from_task_file(task_file, test).tap do |file|
+            file.weight = extract_meta_data(test.meta_data&.dig('test-meta-data'), 'test-file', task_file.id, 'weight').presence || 1.0
+            file.feedback_message = extract_meta_data(test.meta_data&.dig('test-meta-data'), 'test-file', task_file.id, 'feedback-message').presence || 'Feedback'
+            file.hidden_feedback = extract_meta_data(test.meta_data&.dig('test-meta-data'), 'test-file', task_file.id, 'hidden-feedback').presence || false
+            file.role = 'teacher_defined_test' unless file.teacher_defined_assessment?
+          end
         end
       end
     end
 
     def model_solution_files
-      @task.model_solutions.map do |model_solution_object|
-        task_files.delete(model_solution_object.files.first.id).tap do |file|
-          file.role ||= 'reference_implementation'
+      @task.model_solutions.flat_map do |model_solution|
+        model_solution.files.map do |task_file|
+          codeocean_file_from_task_file(task_file, model_solution).tap do |file|
+            file.role ||= 'reference_implementation'
+          end
         end
       end
     end
 
     def task_files
-      @task_files ||= @task.all_files.reject {|file| file.id == 'ms-placeholder-file' }.to_h do |task_file|
-        [task_file.id, codeocean_file_from_task_file(task_file)]
+      @task.files.reject {|file| file.id == 'ms-placeholder-file' }.map do |task_file|
+        codeocean_file_from_task_file(task_file).tap do |file|
+          file.role ||= 'regular_file'
+        end
       end
+      # @task_files ||= @task.all_files.reject {|file| file.id == 'ms-placeholder-file' }.to_h do |task_file|
+      #   [task_file.id, codeocean_file_from_task_file(task_file)]
+      # end
     end
 
-    def codeocean_file_from_task_file(file)
+    def codeocean_file_from_task_file(file, parent_object = nil)
       extension = File.extname(file.filename)
-      codeocean_file = CodeOcean::File.new(
+
+      codeocean_file = CodeOcean::File.where(context: @exercise).where('xml_id_path LIKE ?', "%#{file.id}").first_or_initialize(context: @exercise)
+      codeocean_file.assign_attributes(
         context: @exercise,
         file_type: file_type(extension),
         hidden: file.visible != 'yes', # hides 'delayed' and 'no'
         name: File.basename(file.filename, '.*'),
         read_only: file.usage_by_lms != 'edit',
         role: extract_meta_data(@task.meta_data&.dig('meta-data'), 'files', "CO-#{file.id}", 'role'),
-        path: File.dirname(file.filename).in?(['.', '']) ? nil : File.dirname(file.filename)
+        path: File.dirname(file.filename).in?(['.', '']) ? nil : File.dirname(file.filename),
+        xml_id_path: (parent_object.nil? ? '' : "#{parent_object.id}/") + file.id.to_s
       )
       if file.binary
         codeocean_file.native_file = FileIO.new(file.content.dup.force_encoding('UTF-8'), File.basename(file.filename))

app/services/proforma_service/import.rb

@@ -14,13 +14,7 @@ def execute
         import_result = importer.perform
         @task = import_result
 
-        exercise = base_exercise
-        exercise_files = exercise&.files&.to_a
-
-        exercise = ConvertTaskToExercise.call(task: @task, user: @user, exercise:)
-        exercise_files&.each(&:destroy) # feels suboptimal
-
-        exercise
+        ConvertTaskToExercise.call(task: @task, user: @user, exercise: base_exercise)
       else
         import_multi
       end

db/migrate/20240903204319_add_xml_path_to_files.rb

@@ -0,0 +1,7 @@
+# frozen_string_literal: true
+
+class AddXmlPathToFiles < ActiveRecord::Migration[7.1]
+  def change
+    add_column :files, :xml_id_path, :string, null: true, default: nil
+  end
+end

db/schema.rb

@@ -10,15 +10,15 @@
 #
 # It's strongly recommended that you check this file into your version control system.
 
-ActiveRecord::Schema[7.1].define(version: 2023_12_08_194632) do
+ActiveRecord::Schema[7.1].define(version: 2024_09_03_204319) do
   # These are extensions that must be enabled in order to support this database
   enable_extension "pg_trgm"
   enable_extension "pgcrypto"
   enable_extension "plpgsql"
 
   create_table "anomaly_notifications", id: :serial, force: :cascade do |t|
-    t.integer "contributor_id", null: false
     t.string "contributor_type", null: false
+    t.integer "contributor_id", null: false
     t.integer "exercise_id", null: false
     t.integer "exercise_collection_id", null: false
     t.jsonb "reason"
@@ -46,10 +46,10 @@
     t.string "api_key"
     t.datetime "created_at"
     t.datetime "updated_at"
+    t.string "user_type"
     t.integer "user_id"
     t.string "push_url"
     t.string "check_uuid_url"
-    t.string "user_type"
     t.index ["user_type", "user_id"], name: "index_codeharbor_links_on_user_type_and_user_id"
   end
 
@@ -137,8 +137,8 @@
   create_table "events", id: :serial, force: :cascade do |t|
     t.string "category"
     t.string "data"
-    t.integer "user_id"
     t.string "user_type"
+    t.integer "user_id"
     t.integer "exercise_id"
     t.integer "file_id"
     t.datetime "created_at", null: false
@@ -207,8 +207,8 @@
     t.datetime "created_at"
     t.datetime "updated_at"
     t.boolean "use_anomaly_detection", default: false
-    t.integer "user_id"
     t.string "user_type"
+    t.integer "user_id"
     t.index ["user_type", "user_id"], name: "index_exercise_collections_on_user_type_and_user_id"
   end
 
@@ -298,8 +298,8 @@
 
   create_table "files", id: :serial, force: :cascade do |t|
     t.text "content"
-    t.integer "context_id"
     t.string "context_type"
+    t.integer "context_id"
     t.integer "file_id"
     t.integer "file_type_id"
     t.boolean "hidden"
@@ -315,6 +315,7 @@
     t.string "path"
     t.integer "file_template_id"
     t.boolean "hidden_feedback", default: false, null: false
+    t.string "xml_id_path"
     t.index ["context_id", "context_type"], name: "index_files_on_context_id_and_context_type"
   end
 
@@ -492,8 +493,8 @@
 
   create_table "searches", id: :serial, force: :cascade do |t|
     t.integer "exercise_id", null: false
-    t.integer "user_id", null: false
     t.string "user_type", null: false
+    t.integer "user_id", null: false
     t.string "search"
     t.datetime "created_at"
     t.datetime "updated_at"
@@ -523,7 +524,7 @@
     t.bigint "user_id"
     t.integer "role", limit: 2, default: 0, null: false, comment: "Used as enum in Rails"
     t.index ["study_group_id"], name: "index_study_group_memberships_on_study_group_id"
-    t.index ["user_type", "user_id"], name: "index_study_group_memberships_on_user"
+    t.index ["user_type", "user_id"], name: "index_study_group_memberships_on_user_type_and_user_id"
   end
 
   create_table "study_groups", force: :cascade do |t|
@@ -551,8 +552,8 @@
   end
 
   create_table "subscriptions", id: :serial, force: :cascade do |t|
-    t.integer "user_id"
     t.string "user_type"
+    t.integer "user_id"
     t.integer "request_for_comment_id"
     t.string "subscription_type"
     t.datetime "created_at", null: false
@@ -619,13 +620,13 @@
     t.datetime "created_at", null: false
     t.datetime "updated_at", null: false
     t.index ["file_type_id"], name: "index_tips_on_file_type_id"
-    t.index ["user_type", "user_id"], name: "index_tips_on_user"
+    t.index ["user_type", "user_id"], name: "index_tips_on_user_type_and_user_id"
   end
 
   create_table "user_exercise_feedbacks", id: :serial, force: :cascade do |t|
     t.integer "exercise_id", null: false
-    t.integer "user_id", null: false
     t.string "user_type", null: false
+    t.integer "user_id", null: false
     t.integer "difficulty"
     t.integer "working_time_seconds"
     t.string "feedback_text"
@@ -638,8 +639,8 @@
   end
 
   create_table "user_exercise_interventions", id: :serial, force: :cascade do |t|
-    t.integer "contributor_id", null: false
     t.string "contributor_type", null: false
+    t.integer "contributor_id", null: false
     t.integer "exercise_id", null: false
     t.integer "intervention_id", null: false
     t.integer "accumulated_worktime_s"
@@ -652,16 +653,16 @@
   end
 
   create_table "user_proxy_exercise_exercises", id: :serial, force: :cascade do |t|
-    t.integer "user_id"
     t.string "user_type"
+    t.integer "user_id"
     t.integer "proxy_exercise_id"
     t.integer "exercise_id"
     t.datetime "created_at"
     t.datetime "updated_at"
     t.string "reason"
     t.index ["exercise_id"], name: "index_user_proxy_exercise_exercises_on_exercise_id"
     t.index ["proxy_exercise_id"], name: "index_user_proxy_exercise_exercises_on_proxy_exercise_id"
-    t.index ["user_type", "user_id"], name: "index_user_proxy_exercise_exercises_on_user"
+    t.index ["user_type", "user_id"], name: "index_user_proxy_exercise_exercises_on_user_type_and_user_id"
   end
 
   add_foreign_key "anomaly_notifications", "exercise_collections"