fix: validate response data inputs (#15)

Author: kholdrex

CHANGELOG.md

@@ -2,6 +2,13 @@
 
 All notable changes to this project are documented in this file.
 
+## [Unreleased]
+
+### Changed
+- Clarified that response data must be a `Matrix` or array of arrays containing only integer `0`, integer `1`, or `nil`; floats, strings, booleans, and other values are rejected.
+
+---
+
 ## [0.3.0] - 2025-01-14
 
 ### Changed

README.md

@@ -47,6 +47,12 @@ result = model.fit
 puts "Abilities:    #{result[:abilities]}"
 puts "Difficulties: #{result[:difficulties]}"
 ```
+
+Response data passed to model constructors must be either a `Matrix` or an
+array of arrays. Each response value must be the integer `0`, the integer `1`,
+or `nil` for missing data; floats such as `0.0`/`1.0`, strings, booleans, and
+other values are rejected.
+
 ### Using 2PL and 3PL Models
 ```ruby
 two_pl_model = IrtRuby::TwoParameterModel.new(data)

lib/irt_ruby.rb

@@ -2,6 +2,7 @@
 
 require "irt_ruby/version"
 require "matrix"
+require "irt_ruby/response_data_validator"
 require "irt_ruby/rasch_model"
 require "irt_ruby/two_parameter_model"
 require "irt_ruby/three_parameter_model"

lib/irt_ruby/rasch_model.rb

@@ -1,5 +1,7 @@
 # frozen_string_literal: true
 
+require "irt_ruby/response_data_validator"
+
 module IrtRuby
   # A class representing the Rasch model for Item Response Theory (ability - difficulty).
   # Incorporates:
@@ -20,7 +22,7 @@ def initialize(data,
       # missing_strategy: :ignore (skip), :treat_as_incorrect, :treat_as_correct
 
       @data = data
-      @data_array = data.to_a
+      @data_array = ResponseDataValidator.validate!(data)
       num_rows = @data_array.size
       num_cols = @data_array.first.size
 

lib/irt_ruby/response_data_validator.rb

@@ -0,0 +1,59 @@
+# frozen_string_literal: true
+
+module IrtRuby
+  # Validates response data accepted by IRT model constructors.
+  # @api private
+  module ResponseDataValidator
+    module_function
+
+    def validate!(data)
+      raise ArgumentError, "response data must be a Matrix or array of arrays" unless valid_data_container?(data)
+
+      data_array = data.to_a
+
+      raise ArgumentError, "response data must have at least one row" unless data_array.any?
+
+      validate_rows!(data_array)
+      validate_values!(data_array)
+
+      data_array
+    end
+
+    def validate_rows!(data_array)
+      first_row = data_array.first
+
+      raise ArgumentError, "response data must be a Matrix or array of arrays" unless first_row.is_a?(Array)
+
+      expected_columns = first_row.size
+      raise ArgumentError, "response data must have at least one column" if expected_columns.zero?
+
+      data_array.each_with_index do |row, index|
+        row_number = index + 1
+        raise ArgumentError, "response data row #{row_number} must be an Array" unless row.is_a?(Array)
+
+        next if row.size == expected_columns
+
+        raise ArgumentError, "response data must be rectangular; row #{row_number} has #{row.size} columns, expected #{expected_columns}"
+      end
+    end
+
+    def validate_values!(data_array)
+      data_array.each_with_index do |row, row_index|
+        row.each_with_index do |value, column_index|
+          next if valid_response?(value)
+
+          raise ArgumentError,
+                "response data contains invalid value #{value.inspect} at row #{row_index + 1}, column #{column_index + 1}; allowed values are 0, 1, and nil"
+        end
+      end
+    end
+
+    def valid_response?(value)
+      value.nil? || value.eql?(0) || value.eql?(1)
+    end
+
+    def valid_data_container?(data)
+      data.is_a?(Array) || (defined?(::Matrix) && data.is_a?(::Matrix))
+    end
+  end
+end

lib/irt_ruby/three_parameter_model.rb

@@ -1,5 +1,7 @@
 # frozen_string_literal: true
 
+require "irt_ruby/response_data_validator"
+
 module IrtRuby
   # A class representing the Three-Parameter model (3PL) for Item Response Theory.
   # Incorporates:
@@ -19,7 +21,7 @@ def initialize(data,
                    decay_factor: 0.5,
                    missing_strategy: :ignore)
       @data = data
-      @data_array = data.to_a
+      @data_array = ResponseDataValidator.validate!(data)
       num_rows = @data_array.size
       num_cols = @data_array.first.size
 

lib/irt_ruby/two_parameter_model.rb

@@ -1,5 +1,7 @@
 # frozen_string_literal: true
 
+require "irt_ruby/response_data_validator"
+
 module IrtRuby
   # A class representing the Two-Parameter model (2PL) for IRT.
   # Incorporates:
@@ -15,7 +17,7 @@ def initialize(data, max_iter: 1000, tolerance: 1e-6, param_tolerance: 1e-6,
                    learning_rate: 0.01, decay_factor: 0.5,
                    missing_strategy: :ignore)
       @data = data
-      @data_array = data.to_a
+      @data_array = ResponseDataValidator.validate!(data)
       num_rows = @data_array.size
       num_cols = @data_array.first.size
 

spec/irt_ruby/rasch_model_spec.rb

@@ -3,6 +3,8 @@
 require "spec_helper"
 
 RSpec.describe IrtRuby::RaschModel do
+  it_behaves_like "response data validation"
+
   let(:data_array) do
     [
       [1, 1, 0],

spec/irt_ruby/three_parameter_model_spec.rb

@@ -3,6 +3,8 @@
 require "spec_helper"
 
 RSpec.describe IrtRuby::ThreeParameterModel do
+  it_behaves_like "response data validation"
+
   let(:data_array) do
     [
       [1, 1, 0],

spec/irt_ruby/two_parameter_model_spec.rb

@@ -3,6 +3,8 @@
 require "spec_helper"
 
 RSpec.describe IrtRuby::TwoParameterModel do
+  it_behaves_like "response data validation"
+
   let(:data_array) do
     [
       [1, 1, 0],