canon 0.2.11 → 0.2.12

data/lib/canon/tree_diff/operations/operation_detector.rb

@@ -1,7 +1,5 @@
 # frozen_string_literal: true
 
-require_relative "../core/xml_entity_decoder"
-
 module Canon
   module TreeDiff
     module Operations
@@ -619,13 +617,12 @@ module Canon
           # Check if this node or any ancestor is whitespace-sensitive
           current = node
           while current
-            if current.respond_to?(:label)
+            if current.is_a?(Core::TreeNode)
               label = current.label.to_s.downcase
               return true if whitespace_sensitive_tags.include?(label)
             end
 
-            # Check parent
-            current = current.parent if current.respond_to?(:parent)
+            current = current.is_a?(Core::TreeNode) ? current.parent : nil
             break unless current
           end
 

data/lib/canon/tree_diff/operations.rb

@@ -0,0 +1,13 @@
+# frozen_string_literal: true
+
+module Canon
+  module TreeDiff
+    # Tree-diff operations: Operation (INSERT/DELETE/UPDATE/MOVE) and
+    # the OperationDetector that emits them.
+    module Operations
+      autoload :Operation, "canon/tree_diff/operations/operation"
+      autoload :OperationDetector,
+               "canon/tree_diff/operations/operation_detector"
+    end
+  end
+end

data/lib/canon/tree_diff.rb

@@ -1,33 +1,32 @@
 # frozen_string_literal: true
 
 module Canon
+  # Semantic tree-diff algorithm — distinct from the DOM positional
+  # diff in {Canon::Comparison}.
+  #
+  # This module computes signature-based tree matches and produces
+  # INSERT/DELETE/UPDATE/MOVE operations. Sub-namespaces:
+  #
+  # * Core — TreeNode, Matching, NodeSignature, NodeWeight,
+  #   AttributeComparator, XmlEntityDecoder
+  # * Matchers — HashMatcher, SimilarityMatcher, StructuralPropagator,
+  #   UniversalMatcher
+  # * Operations — Operation, OperationDetector
+  # * Adapters — format-specific tree adapters (XML, JSON, HTML, YAML)
+  # * OperationConverterHelpers — MetadataEnricher, ReasonBuilder,
+  #   PostProcessor, UpdateChangeHandler
+  #
+  # Top-level entry points: OperationConverter and TreeDiffIntegrator.
+  #
+  # All children are autoloaded — never `require_relative` them.
   module TreeDiff
-    # Tree diff module for semantic object tree diffing
+    autoload :Adapters, "canon/tree_diff/adapters"
+    autoload :Core, "canon/tree_diff/core"
+    autoload :Matchers, "canon/tree_diff/matchers"
+    autoload :OperationConverter, "canon/tree_diff/operation_converter"
+    autoload :OperationConverterHelpers,
+             "canon/tree_diff/operation_converter_helpers"
+    autoload :Operations, "canon/tree_diff/operations"
+    autoload :TreeDiffIntegrator, "canon/tree_diff/tree_diff_integrator"
   end
 end
-
-# Load core components
-require_relative "tree_diff/core/tree_node"
-require_relative "tree_diff/core/node_signature"
-require_relative "tree_diff/core/node_weight"
-require_relative "tree_diff/core/matching"
-
-# Load matchers
-require_relative "tree_diff/matchers/hash_matcher"
-require_relative "tree_diff/matchers/similarity_matcher"
-require_relative "tree_diff/matchers/structural_propagator"
-require_relative "tree_diff/matchers/universal_matcher"
-
-# Load operations
-require_relative "tree_diff/operations/operation"
-require_relative "tree_diff/operations/operation_detector"
-require_relative "tree_diff/operation_converter"
-
-# Load adapters
-require_relative "tree_diff/adapters/xml_adapter"
-require_relative "tree_diff/adapters/json_adapter"
-require_relative "tree_diff/adapters/html_adapter"
-require_relative "tree_diff/adapters/yaml_adapter"
-
-# Load integrator
-require_relative "tree_diff/tree_diff_integrator"

data/lib/canon/validators/base_validator.rb

@@ -1,7 +1,5 @@
 # frozen_string_literal: true
 
-require_relative "../errors"
-
 module Canon
   module Validators
     # Base class for all input validators

data/lib/canon/validators/html_validator.rb

@@ -1,7 +1,6 @@
 # frozen_string_literal: true
 
 require "nokogiri" unless RUBY_ENGINE == "opal"
-require_relative "base_validator"
 
 module Canon
   module Validators

data/lib/canon/validators/json_validator.rb

@@ -1,7 +1,6 @@
 # frozen_string_literal: true
 
 require "json"
-require_relative "base_validator"
 
 module Canon
   module Validators

data/lib/canon/validators/xml_validator.rb

@@ -1,7 +1,6 @@
 # frozen_string_literal: true
 
 require "nokogiri" unless RUBY_ENGINE == "opal"
-require_relative "base_validator"
 
 module Canon
   module Validators

data/lib/canon/validators/yaml_validator.rb

@@ -3,7 +3,6 @@
 require "yaml"
 require "date"
 require "time"
-require_relative "base_validator"
 
 module Canon
   module Validators

data/lib/canon/validators.rb

@@ -0,0 +1,12 @@
+# frozen_string_literal: true
+
+module Canon
+  # Format-specific validators that raise {Error} on invalid input.
+  module Validators
+    autoload :BaseValidator, "canon/validators/base_validator"
+    autoload :HtmlValidator, "canon/validators/html_validator"
+    autoload :JsonValidator, "canon/validators/json_validator"
+    autoload :XmlValidator, "canon/validators/xml_validator"
+    autoload :YamlValidator, "canon/validators/yaml_validator"
+  end
+end

data/lib/canon/version.rb

@@ -1,5 +1,5 @@
 # frozen_string_literal: true
 
 module Canon
-  VERSION = "0.2.11"
+  VERSION = "0.2.12"
 end

data/lib/canon/xml/c14n.rb

@@ -1,9 +1,5 @@
 # frozen_string_literal: true
 
-require_relative "data_model"
-require_relative "processor"
-require_relative "xpath_engine"
-
 module Canon
   module Xml
     # XML Canonicalization 1.1 implementation

data/lib/canon/xml/data_model.rb

@@ -2,16 +2,6 @@
 
 require "nokogiri" unless RUBY_ENGINE == "opal"
 require "set"
-require_relative "../data_model"
-require_relative "../xml_backend"
-require_relative "../xml_parsing"
-require_relative "nodes/root_node"
-require_relative "nodes/element_node"
-require_relative "nodes/namespace_node"
-require_relative "nodes/attribute_node"
-require_relative "nodes/text_node"
-require_relative "nodes/comment_node"
-require_relative "nodes/processing_instruction_node"
 
 module Canon
   module Xml

data/lib/canon/xml/line_range_mapper.rb

@@ -1,7 +1,5 @@
 # frozen_string_literal: true
 
-require_relative "../pretty_printer/xml"
-
 module Canon
   module Xml
     # Maps DOM elements to line ranges in pretty-printed XML

data/lib/canon/xml/nodes/attribute_node.rb

@@ -1,7 +1,5 @@
 # frozen_string_literal: true
 
-require_relative "../node"
-
 module Canon
   module Xml
     module Nodes

data/lib/canon/xml/nodes/comment_node.rb

@@ -1,7 +1,5 @@
 # frozen_string_literal: true
 
-require_relative "../node"
-
 module Canon
   module Xml
     module Nodes

data/lib/canon/xml/nodes/element_node.rb

@@ -1,7 +1,5 @@
 # frozen_string_literal: true
 
-require_relative "../node"
-
 module Canon
   module Xml
     module Nodes

data/lib/canon/xml/nodes/namespace_node.rb

@@ -1,7 +1,5 @@
 # frozen_string_literal: true
 
-require_relative "../node"
-
 module Canon
   module Xml
     module Nodes

data/lib/canon/xml/nodes/processing_instruction_node.rb

@@ -1,7 +1,5 @@
 # frozen_string_literal: true
 
-require_relative "../node"
-
 module Canon
   module Xml
     module Nodes

data/lib/canon/xml/nodes/root_node.rb

@@ -1,7 +1,5 @@
 # frozen_string_literal: true
 
-require_relative "../node"
-
 module Canon
   module Xml
     module Nodes

data/lib/canon/xml/nodes/text_node.rb

@@ -1,7 +1,5 @@
 # frozen_string_literal: true
 
-require_relative "../node"
-
 module Canon
   module Xml
     module Nodes

data/lib/canon/xml/nodes.rb

@@ -0,0 +1,19 @@
+# frozen_string_literal: true
+
+module Canon
+  module Xml
+    # XPath data model node types. All nodes inherit from
+    # {Canon::Xml::Node}. Children are autoloaded — never
+    # `require_relative` them.
+    module Nodes
+      autoload :AttributeNode, "canon/xml/nodes/attribute_node"
+      autoload :CommentNode, "canon/xml/nodes/comment_node"
+      autoload :ElementNode, "canon/xml/nodes/element_node"
+      autoload :NamespaceNode, "canon/xml/nodes/namespace_node"
+      autoload :ProcessingInstructionNode,
+               "canon/xml/nodes/processing_instruction_node"
+      autoload :RootNode, "canon/xml/nodes/root_node"
+      autoload :TextNode, "canon/xml/nodes/text_node"
+    end
+  end
+end

data/lib/canon/xml/processor.rb

@@ -1,10 +1,5 @@
 # frozen_string_literal: true
 
-require_relative "character_encoder"
-require_relative "namespace_handler"
-require_relative "attribute_handler"
-require_relative "xml_base_handler"
-
 module Canon
   module Xml
     # C14N 1.1 processor

data/lib/canon/xml/sax_builder.rb

@@ -1,13 +1,6 @@
 # frozen_string_literal: true
 
 require "nokogiri" unless RUBY_ENGINE == "opal"
-require_relative "nodes/root_node"
-require_relative "nodes/element_node"
-require_relative "nodes/namespace_node"
-require_relative "nodes/attribute_node"
-require_relative "nodes/text_node"
-require_relative "nodes/comment_node"
-require_relative "nodes/processing_instruction_node"
 
 module Canon
   module Xml

data/lib/canon/xml.rb

@@ -0,0 +1,33 @@
+# frozen_string_literal: true
+
+module Canon
+  # Native XML data model, parsing, and DOM utilities.
+  #
+  # This namespace holds Canon's own XML representation (independent of
+  # Nokogiri/Moxml), including:
+  # * the XPath data model nodes (Canon::Xml::Node and Canon::Xml::Nodes::*)
+  # * the SAX builder that produces them
+  # * element matching, line-range mapping, xpath, C14N, processors
+  # * canonicalization (C14n) and serialization helpers
+  #
+  # All children are autoloaded from this file. The nested Nodes namespace
+  # is itself a sibling and is autoloaded on first reference to
+  # Canon::Xml::Nodes.
+  module Xml
+    autoload :AttributeHandler, "canon/xml/attribute_handler"
+    autoload :C14n, "canon/xml/c14n"
+    autoload :CharacterEncoder, "canon/xml/character_encoder"
+    autoload :DataModel, "canon/xml/data_model"
+    autoload :ElementMatcher, "canon/xml/element_matcher"
+    autoload :LineRangeMapper, "canon/xml/line_range_mapper"
+    autoload :NamespaceHandler, "canon/xml/namespace_handler"
+    autoload :NamespaceHelper, "canon/xml/namespace_helper"
+    autoload :Node, "canon/xml/node"
+    autoload :Nodes, "canon/xml/nodes"
+    autoload :Processor, "canon/xml/processor"
+    autoload :SaxBuilder, "canon/xml/sax_builder"
+    autoload :WhitespaceNormalizer, "canon/xml/whitespace_normalizer"
+    autoload :XmlBaseHandler, "canon/xml/xml_base_handler"
+    autoload :XPathEngine, "canon/xml/xpath_engine"
+  end
+end

data/lib/canon/xml_backend.rb

@@ -1,20 +1,6 @@
 # frozen_string_literal: true
 
 module Canon
-  # Centralized XML backend detection for Canon.
-  #
-  # Canon supports two XML backends:
-  # - :nokogiri — MRI with Nokogiri installed (default, existing code path)
-  # - :moxml    — Opal runtime or MRI without Nokogiri (uses Oga via moxml)
-  #
-  # The active backend is determined once at load time and cached.
-  # All XML-related code should check `Canon::XmlBackend.moxml?` or
-  # `Canon::XmlBackend.nokogiri?` to select the appropriate code path.
-  #
-  # This module intentionally does NOT wrap Nokogiri through moxml.
-  # Each backend path is independent — the Nokogiri path is the existing
-  # battle-tested code; the moxml path is a parallel implementation for
-  # environments where Nokogiri is unavailable.
   module XmlBackend
     class << self
       def active
@@ -33,6 +19,56 @@ module Canon
         @active = nil
       end
 
+      # Whether the node is a document fragment (any variant).
+      def document_fragment?(node)
+        if nokogiri?
+          node.is_a?(Nokogiri::XML::DocumentFragment) ||
+            node.is_a?(Nokogiri::HTML4::DocumentFragment) ||
+            node.is_a?(Nokogiri::HTML5::DocumentFragment)
+        else
+          false
+        end
+      end
+
+      # Whether the node is an HTML document (any variant).
+      def html_document?(node)
+        if nokogiri?
+          node.is_a?(Nokogiri::HTML::Document) ||
+            node.is_a?(Nokogiri::HTML4::Document) ||
+            node.is_a?(Nokogiri::HTML5::Document)
+        else
+          false
+        end
+      end
+
+      # Detect HTML version from a Nokogiri node.
+      # Returns :html5 or :html4. Defaults to :html5 for non-Nokogiri nodes.
+      def html_version_from_node(node)
+        if nokogiri?
+          if node.is_a?(Nokogiri::HTML5::Document) ||
+              node.is_a?(Nokogiri::HTML5::DocumentFragment)
+            :html5
+          elsif node.is_a?(Nokogiri::HTML4::Document) ||
+              node.is_a?(Nokogiri::HTML4::DocumentFragment)
+            :html4
+          else
+            :html5
+          end
+        else
+          :html5
+        end
+      end
+
+      # Parse an HTML string into an XML fragment.
+      def xml_fragment(html_string)
+        if nokogiri?
+          Nokogiri::XML.fragment(html_string)
+        else
+          raise Canon::Error,
+                "HTML fragment parsing requires the Nokogiri backend"
+        end
+      end
+
       private
 
       def detect

data/lib/canon/xml_parsing.rb

@@ -264,8 +264,10 @@ module Canon
         node.to_xml
       end
 
-      def moxml_canonicalize(node, _options)
-        node.to_xml
+      def moxml_canonicalize(_node, _options)
+        raise Canon::Error,
+              "C14N canonicalization is not supported by the moxml backend. " \
+              "Use the Nokogiri backend or a different preprocessing option."
       end
 
       def moxml_node_type(node)

data/lib/canon.rb

@@ -1,23 +1,31 @@
 # frozen_string_literal: true
 
-require_relative "canon/version"
-require_relative "canon/errors"
-require_relative "canon/xml_backend"
-require_relative "canon/xml_parsing"
-require_relative "canon/config"
-require_relative "canon/data_model"
-require_relative "canon/html"
-require_relative "canon/formatters/xml_formatter"
-require_relative "canon/formatters/yaml_formatter"
-require_relative "canon/formatters/json_formatter"
-require_relative "canon/formatters/html_formatter"
-require_relative "canon/formatters/html4_formatter"
-require_relative "canon/formatters/html5_formatter"
-require_relative "canon/comparison"
+require "canon/version"
+require "canon/errors"
+require "nokogiri" unless RUBY_ENGINE == "opal"
+require "canon/xml_backend"
+require "canon/xml_parsing"
+require "canon/config"
+require "canon/data_model"
+require "canon/xml"
+require "canon/html"
+require "canon/formatters"
+require "canon/comparison"
+require "canon/diff"
+require "canon/tree_diff"
+require "canon/validators"
+require "canon/pretty_printer"
+require "canon/options"
+require "canon/commands"
 
-require_relative "canon/rspec_matchers" if defined?(RSpec.configure)
+require "canon/rspec_matchers" if defined?(RSpec.configure)
 
 module Canon
+  autoload :Cache, "canon/cache"
+  autoload :Cli, "canon/cli"
+  autoload :ColorDetector, "canon/color_detector"
+  autoload :DiffFormatter, "canon/diff_formatter"
+
   SUPPORTED_FORMATS = %i[xml yaml json html html4 html5 string].freeze
 
   # Format content based on the specified format type
@@ -60,6 +68,8 @@ module Canon
   # Define shorthand methods for each supported format
   # Creates parse_{format} and format_{format} methods
   SUPPORTED_FORMATS.each do |format|
+    next if format == :string # comparison-only format, no formatter
+
     define_singleton_method("parse_#{format}") do |content|
       parse(content, format)
     end

data/lib/tasks/performance.rake

@@ -1,7 +1,6 @@
 # frozen_string_literal: true
 
 require_relative "performance_comparator"
-require_relative "benchmark_runner"
 
 desc "Run performance benchmarks"
 namespace :performance do
@@ -16,66 +15,9 @@ namespace :performance do
     runner.run_benchmarks
   end
 
-  desc "Run specific benchmark category (xml_parsing, html_parsing, xml_comparison, html_comparison, formatting)"
-  task :category, [:name] do |_t, args|
-    category = args[:name]
-    unless PerformanceComparator::BENCHMARK_CATEGORIES.key?(category.to_sym)
-      puts "Unknown category: #{category}"
-      puts "Available: #{PerformanceComparator::BENCHMARK_CATEGORIES.keys.join(', ')}"
-      exit(1)
-    end
-
-    runner = BenchmarkRunner.new(run_time: 10)
-    runner.run_benchmarks
-  end
-
   desc "Quick benchmark run (faster, less accurate)"
   task :quick do
     runner = BenchmarkRunner.new(run_time: 2, warmup: 1, items: 20)
     runner.run_benchmarks
   end
-
-  desc "Run benchmarks and output as JSON"
-  task :json do
-    require "json"
-    runner = BenchmarkRunner.new(run_time: 5)
-
-    # Suppress pretty output, just get results
-    results = runner.send(:run_all_benchmarks)
-
-    output = results.each_with_object({}) do |(label, metrics), h|
-      ips = (metrics[:lower] + metrics[:upper]) / 2.0
-      deviation = ((metrics[:upper] - metrics[:lower]) / metrics[:upper] * 100).round(1)
-      h[label] = {
-        ips: ips.round(2),
-        lower: metrics[:lower].round(2),
-        upper: metrics[:upper].round(2),
-        deviation: deviation,
-      }
-    end
-
-    puts JSON.pretty_generate(output)
-  end
-
-  desc "Run benchmarks and output as YAML"
-  task :yaml do
-    require "yaml"
-    runner = BenchmarkRunner.new(run_time: 5)
-
-    # Suppress pretty output, just get results
-    results = runner.send(:run_all_benchmarks)
-
-    output = results.each_with_object({}) do |(label, metrics), h|
-      ips = (metrics[:lower] + metrics[:upper]) / 2.0
-      deviation = ((metrics[:upper] - metrics[:lower]) / metrics[:upper] * 100).round(1)
-      h[label.to_sym] = {
-        ips: ips.round(2),
-        lower: metrics[:lower].round(2),
-        upper: metrics[:upper].round(2),
-        deviation: deviation,
-      }
-    end
-
-    puts YAML.dump(output)
-  end
 end