fix(codaveri): fix crash on import question with empty data files

app/services/course/assessment/question/programming_codaveri/c_sharp/c_sharp_package_service.rb

@@ -97,27 +97,6 @@ def extract_supporting_solution_files
     end
   end
 
-  # Extracts filename and content of a data file and append it to the
-  # [:additionalFiles] array for the problem management API request body.
-  #
-  # @param [Pathname] pathname The pathname of the file.
-  # @param [String] content The content of the file.
-  def extract_supporting_file(filename, content)
-    supporting_file_object = default_codaveri_data_file_template
-
-    supporting_file_object[:type] = 'internal' # 'external' s3 upload not yet implemented by codaveri
-    supporting_file_object[:path] = filename.to_s
-    if content.force_encoding('UTF-8').valid_encoding?
-      supporting_file_object[:content] = content
-      supporting_file_object[:encoding] = 'utf8'
-    else
-      supporting_file_object[:content] = Base64.strict_encode64(content)
-      supporting_file_object[:encoding] = 'base64'
-    end
-
-    @data_files.append(supporting_file_object)
-  end
-
   # Extracts test cases from the built dummy reports and append all the test cases to the
   # [:IOTestcases] array for the problem management API request body.
   def extract_test_cases # rubocop:disable Metrics/AbcSize

app/services/course/assessment/question/programming_codaveri/go/go_package_service.rb

@@ -97,27 +97,6 @@ def extract_supporting_solution_files
     end
   end
 
-  # Extracts filename and content of a data file and append it to the
-  # [:additionalFiles] array for the problem management API request body.
-  #
-  # @param [Pathname] pathname The pathname of the file.
-  # @param [String] content The content of the file.
-  def extract_supporting_file(filename, content)
-    supporting_file_object = default_codaveri_data_file_template
-
-    supporting_file_object[:type] = 'internal' # 'external' s3 upload not yet implemented by codaveri
-    supporting_file_object[:path] = filename.to_s
-    if content.force_encoding('UTF-8').valid_encoding?
-      supporting_file_object[:content] = content
-      supporting_file_object[:encoding] = 'utf8'
-    else
-      supporting_file_object[:content] = Base64.strict_encode64(content)
-      supporting_file_object[:encoding] = 'base64'
-    end
-
-    @data_files.append(supporting_file_object)
-  end
-
   # Extracts test cases from the built dummy reports and append all the test cases to the
   # [:IOTestcases] array for the problem management API request body.
   def extract_test_cases # rubocop:disable Metrics/AbcSize

app/services/course/assessment/question/programming_codaveri/java/java_package_service.rb

@@ -103,23 +103,11 @@ def extract_supporting_tests_files
     end
   end
 
-  def extract_supporting_file(filename, content)
-    supporting_file_object = default_codaveri_data_file_template
-
-    supporting_file_object[:type] = 'internal' # 'external' s3 upload not yet implemented by codaveri
-    supporting_file_object[:path] = filename.to_s
-    # TODO: remove filename.to_s.downcase.end_with?('.java') check
-    # For now, only plaintext files that require compiling (e.g. *.java) will use 'utf8' ecoding
-    # Pending Codaveri 'utf8' encoding support for all plaintext files in compiled languages
-    if content.force_encoding('UTF-8').valid_encoding? && filename.to_s.downcase.end_with?('.java')
-      supporting_file_object[:content] = content
-      supporting_file_object[:encoding] = 'utf8'
-    else
-      supporting_file_object[:content] = Base64.strict_encode64(content)
-      supporting_file_object[:encoding] = 'base64'
-    end
-
-    @data_files.append(supporting_file_object)
+  # TODO: remove filename.to_s.downcase.end_with?('.java') check
+  # For now, only plaintext files that require compiling (e.g. *.java) will use 'utf8' ecoding
+  # Pending Codaveri 'utf8' encoding support for all plaintext files in compiled languages
+  def utf8_encodable?(filename, utf8_content)
+    super && filename.to_s.downcase.end_with?('.java')
   end
 
   def extract_template

app/services/course/assessment/question/programming_codaveri/java_script/java_script_package_service.rb

@@ -97,27 +97,6 @@ def extract_supporting_solution_files
     end
   end
 
-  # Extracts filename and content of a data file and append it to the
-  # [:additionalFiles] array for the problem management API request body.
-  #
-  # @param [Pathname] pathname The pathname of the file.
-  # @param [String] content The content of the file.
-  def extract_supporting_file(filename, content)
-    supporting_file_object = default_codaveri_data_file_template
-
-    supporting_file_object[:type] = 'internal' # 'external' s3 upload not yet implemented by codaveri
-    supporting_file_object[:path] = filename.to_s
-    if content.force_encoding('UTF-8').valid_encoding?
-      supporting_file_object[:content] = content
-      supporting_file_object[:encoding] = 'utf8'
-    else
-      supporting_file_object[:content] = Base64.strict_encode64(content)
-      supporting_file_object[:encoding] = 'base64'
-    end
-
-    @data_files.append(supporting_file_object)
-  end
-
   # Extracts test cases from the built dummy reports and append all the test cases to the
   # [:IOTestcases] array for the problem management API request body.
   def extract_test_cases # rubocop:disable Metrics/AbcSize

app/services/course/assessment/question/programming_codaveri/language_package_service.rb

@@ -61,6 +61,40 @@ def process_evaluator
 
   private
 
+  # Extracts filename and content of a data file and append it to the
+  # [:additionalFiles] array for the problem management API request body.
+  #
+  # @param [Pathname] filename The pathname of the file.
+  # @param [String] content The content of the file.
+  def extract_supporting_file(filename, content)
+    supporting_file_object = default_codaveri_data_file_template
+
+    supporting_file_object[:type] = 'internal' # 'external' s3 upload not yet implemented by codaveri
+    supporting_file_object[:path] = filename.to_s
+    # `content` is read straight out of the zip and may be frozen (rubyzip returns a frozen empty
+    # string literal for zero-byte entries), so tag the encoding on a copy rather than in place.
+    utf8_content = content.dup.force_encoding('UTF-8')
+    if utf8_encodable?(filename, utf8_content)
+      supporting_file_object[:content] = utf8_content
+      supporting_file_object[:encoding] = 'utf8'
+    else
+      supporting_file_object[:content] = Base64.strict_encode64(content)
+      supporting_file_object[:encoding] = 'base64'
+    end
+
+    @data_files.append(supporting_file_object)
+  end
+
+  # Whether a supporting file may be sent to Codaveri as plaintext 'utf8' rather than 'base64'.
+  # Concrete services may narrow this further; see the Java package service.
+  #
+  # @param [Pathname] filename The pathname of the file.
+  # @param [String] utf8_content The content of the file, tagged as UTF-8.
+  # @return [Boolean]
+  def utf8_encodable?(_filename, utf8_content)
+    utf8_content.valid_encoding?
+  end
+
   # Defines the default solution template as indicated in the Codevari API problem management spec.
   #
   # @return [Hash]

app/services/course/assessment/question/programming_codaveri/python/python_package_service.rb

@@ -97,27 +97,6 @@ def extract_supporting_solution_files
     end
   end
 
-  # Extracts filename and content of a data file and append it to the
-  # [:additionalFiles] array for the problem management API request body.
-  #
-  # @param [Pathname] pathname The pathname of the file.
-  # @param [String] content The content of the file.
-  def extract_supporting_file(filename, content)
-    supporting_file_object = default_codaveri_data_file_template
-
-    supporting_file_object[:type] = 'internal' # 'external' s3 upload not yet implemented by codaveri
-    supporting_file_object[:path] = filename.to_s
-    if content.force_encoding('UTF-8').valid_encoding?
-      supporting_file_object[:content] = content
-      supporting_file_object[:encoding] = 'utf8'
-    else
-      supporting_file_object[:content] = Base64.strict_encode64(content)
-      supporting_file_object[:encoding] = 'base64'
-    end
-
-    @data_files.append(supporting_file_object)
-  end
-
   # Extracts test cases from 'autograde.py' and append all the test cases to the
   # [:resources][0][:exprTestcases] array for the problem management API request body.
   def extract_test_cases

app/services/course/assessment/question/programming_codaveri/r/r_package_service.rb

@@ -97,27 +97,6 @@ def extract_supporting_solution_files
     end
   end
 
-  # Extracts filename and content of a data file and append it to the
-  # [:additionalFiles] array for the problem management API request body.
-  #
-  # @param [Pathname] pathname The pathname of the file.
-  # @param [String] content The content of the file.
-  def extract_supporting_file(filename, content)
-    supporting_file_object = default_codaveri_data_file_template
-
-    supporting_file_object[:type] = 'internal' # 'external' s3 upload not yet implemented by codaveri
-    supporting_file_object[:path] = filename.to_s
-    if content.force_encoding('UTF-8').valid_encoding?
-      supporting_file_object[:content] = content
-      supporting_file_object[:encoding] = 'utf8'
-    else
-      supporting_file_object[:content] = Base64.strict_encode64(content)
-      supporting_file_object[:encoding] = 'base64'
-    end
-
-    @data_files.append(supporting_file_object)
-  end
-
   # Extracts test cases from the built dummy reports and append all the test cases to the
   # [:IOTestcases] array for the problem management API request body.
   def extract_test_cases # rubocop:disable Metrics/AbcSize

app/services/course/assessment/question/programming_codaveri/rust/rust_package_service.rb

@@ -97,27 +97,6 @@ def extract_supporting_solution_files
     end
   end
 
-  # Extracts filename and content of a data file and append it to the
-  # [:additionalFiles] array for the problem management API request body.
-  #
-  # @param [Pathname] pathname The pathname of the file.
-  # @param [String] content The content of the file.
-  def extract_supporting_file(filename, content)
-    supporting_file_object = default_codaveri_data_file_template
-
-    supporting_file_object[:type] = 'internal' # 'external' s3 upload not yet implemented by codaveri
-    supporting_file_object[:path] = filename.to_s
-    if content.force_encoding('UTF-8').valid_encoding?
-      supporting_file_object[:content] = content
-      supporting_file_object[:encoding] = 'utf8'
-    else
-      supporting_file_object[:content] = Base64.strict_encode64(content)
-      supporting_file_object[:encoding] = 'base64'
-    end
-
-    @data_files.append(supporting_file_object)
-  end
-
   # Extracts test cases from the built dummy reports and append all the test cases to the
   # [:IOTestcases] array for the problem management API request body.
   def extract_test_cases # rubocop:disable Metrics/AbcSize

app/services/course/assessment/question/programming_codaveri/type_script/type_script_package_service.rb

@@ -97,27 +97,6 @@ def extract_supporting_solution_files
     end
   end
 
-  # Extracts filename and content of a data file and append it to the
-  # [:additionalFiles] array for the problem management API request body.
-  #
-  # @param [Pathname] pathname The pathname of the file.
-  # @param [String] content The content of the file.
-  def extract_supporting_file(filename, content)
-    supporting_file_object = default_codaveri_data_file_template
-
-    supporting_file_object[:type] = 'internal' # 'external' s3 upload not yet implemented by codaveri
-    supporting_file_object[:path] = filename.to_s
-    if content.force_encoding('UTF-8').valid_encoding?
-      supporting_file_object[:content] = content
-      supporting_file_object[:encoding] = 'utf8'
-    else
-      supporting_file_object[:content] = Base64.strict_encode64(content)
-      supporting_file_object[:encoding] = 'base64'
-    end
-
-    @data_files.append(supporting_file_object)
-  end
-
   # Extracts test cases from the built dummy reports and append all the test cases to the
   # [:IOTestcases] array for the problem management API request body.
   def extract_test_cases # rubocop:disable Metrics/AbcSize

spec/services/course/assessment/question/programming_codaveri/language_package_service_spec.rb

@@ -0,0 +1,115 @@
+# frozen_string_literal: true
+require 'rails_helper'
+
+RSpec.describe Course::Assessment::Question::ProgrammingCodaveri::LanguagePackageService do
+  let(:package_path) do
+    File.join(Rails.root, 'spec/fixtures/course/programming_question_template_codaveri_empty_data_file.zip')
+  end
+  let(:package) { Course::Assessment::ProgrammingPackage.new(package_path) }
+  let(:main_files) { package.main_files }
+  let(:test_files) { package.test_files }
+  subject { described_class.new(nil, package) }
+
+  def extract(filename, content)
+    subject.send(:extract_supporting_file, filename, content)
+    subject.data_files.last
+  end
+
+  describe '.extract_supporting_file' do
+    context 'when the file is valid UTF-8 plaintext' do
+      it 'extracts the content as utf8' do
+        filename = Pathname.new('data.csv')
+
+        expect(extract(filename, main_files[filename])).to eq(
+          type: 'internal',
+          path: 'data.csv',
+          content: "codon,amino_acid\nAAA,Lys\nAAC,Asn\n",
+          encoding: 'utf8'
+        )
+      end
+    end
+
+    context 'when the file is not valid UTF-8' do
+      it 'extracts the content as base64' do
+        filename = Pathname.new('binary.dat')
+
+        expect(extract(filename, main_files[filename])).to eq(
+          type: 'internal',
+          path: 'binary.dat',
+          content: Base64.strict_encode64("\xFF\xFE\x00\x01\x02".b),
+          encoding: 'base64'
+        )
+      end
+    end
+
+    # Zero-byte zip entries are read back as a frozen empty string, which used to be mutated
+    # in place by `force_encoding` and raise FrozenError.
+    context 'when the file is empty' do
+      it 'extracts the file as empty utf8 content' do
+        filename = Pathname.new('empty.csv')
+        content = main_files[filename]
+        expect(content).to be_frozen
+
+        expect(extract(filename, content)).to eq(
+          type: 'internal',
+          path: 'empty.csv',
+          content: '',
+          encoding: 'utf8'
+        )
+      end
+
+      it 'extracts a zero-byte file in the tests folder' do
+        filename = Pathname.new('empty.csv')
+
+        expect { extract(filename, test_files[filename]) }.not_to raise_error
+      end
+    end
+
+    it 'does not re-tag the encoding of the content read from the package' do
+      filename = Pathname.new('data.csv')
+      content = main_files[filename]
+
+      extract(filename, content)
+
+      expect(content.encoding).to eq(Encoding::ASCII_8BIT)
+    end
+
+    it 'appends every extracted file to the data files' do
+      subject.send(:extract_supporting_file, Pathname.new('data.csv'), main_files[Pathname.new('data.csv')])
+      subject.send(:extract_supporting_file, Pathname.new('empty.csv'), test_files[Pathname.new('empty.csv')])
+
+      expect(subject.data_files.map { |file| file[:path] }).to eq(['data.csv', 'empty.csv'])
+    end
+  end
+
+  describe '.utf8_encodable?' do
+    it 'accepts any valid UTF-8 content' do
+      expect(subject.send(:utf8_encodable?, Pathname.new('data.csv'), 'plaintext')).to eq(true)
+      expect(subject.send(:utf8_encodable?, Pathname.new('empty.csv'), '')).to eq(true)
+      expect(subject.send(:utf8_encodable?, Pathname.new('binary.dat'), "\xFF\xFE".b.force_encoding('UTF-8'))).
+        to eq(false)
+    end
+
+    # The Java package service narrows the base implementation, pending Codaveri 'utf8' encoding
+    # support for all plaintext files in compiled languages.
+    context 'when the concrete service restricts utf8 encoding' do
+      subject { Course::Assessment::Question::ProgrammingCodaveri::Java::JavaPackageService.new(nil, package) }
+
+      it 'only accepts plaintext files that require compiling' do
+        expect(subject.send(:utf8_encodable?, Pathname.new('Helper.java'), 'class Helper {}')).to eq(true)
+        expect(subject.send(:utf8_encodable?, Pathname.new('data.csv'), 'plaintext')).to eq(false)
+      end
+
+      it 'extracts an empty supporting file as base64' do
+        filename = Pathname.new('empty.csv')
+
+        expect(extract(filename, main_files[filename])).to eq(
+          type: 'internal',
+          path: 'empty.csv',
+          content: '',
+          encoding: 'base64'
+        )
+      end
+    end
+  end
+end