canon 0.1.19 → 0.1.20

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: fae901023e2945c8ee14c48a6de4ce793d2735d9f2b098ba9f727b9c0f10e8ad
-  data.tar.gz: 84ed342a12b39a77394275e6159eb16cc60f331c80ed161a6cd4fccc957dc06d
+  metadata.gz: 5caff8b02e8a381d4febbcbce1b0eeb4000599385428db08b8292fb35b76cfd5
+  data.tar.gz: ee62fd3b76217834e32c026bddcdf889ba9bad48772e5520b0bce7da4d3f982e
 SHA512:
-  metadata.gz: d88d544b3b961dfa5c0f9fb806f51a473e29b0a018a22dbc9ea2aebaaf459a3aa6317d1cb22c2d5dd32d69eb6162389be274a06886566ab9b63f83e613c4b276
-  data.tar.gz: cc409487c2c38791ec915584a8ebca85672ac7add362c78b5ace99bfd1a1657c6907ac387f321dbc12eda2613ffe998ae7fecdfce6895ff5e285f1f6022d250f
+  metadata.gz: 3a317a23324d26f55980d532e07c4d67b1fdfe9cf9e62afdd59ed0381be1f2ebcc6a19da92c02ab9c3e336ec10eda7885dca78d8fbe6de3717876d84f65bc05a
+  data.tar.gz: e38169560e771711fc1f79e60915e8412c0c8229244ac33f23107a87ede5e0f7fad87c47cf85a1519904256f985ed6abfb6d047d04e83c6865fc8ef23163a7d3

data/.rubocop_todo.yml

@@ -1,6 +1,6 @@
 # This configuration was generated by
 # `rubocop --auto-gen-config`
-# on 2026-03-24 10:43:04 UTC using RuboCop version 1.85.1.
+# on 2026-03-25 12:14:38 UTC using RuboCop version 1.86.0.
 # The point is for the user to remove these configuration records
 # one by one as the offenses are removed from the code base.
 # Note that changes in the inspected code, or installation of new
@@ -11,77 +11,36 @@ Gemspec/RequiredRubyVersion:
   Exclude:
     - 'canon.gemspec'
 
-# Offense count: 2
-# This cop supports safe autocorrection (--autocorrect).
-# Configuration parameters: EnforcedStyle, IndentationWidth.
-# SupportedStyles: with_first_argument, with_fixed_indentation
-Layout/ArgumentAlignment:
-  Exclude:
-    - 'lib/canon/diff_formatter/diff_detail_formatter/dimension_formatter.rb'
-
 # Offense count: 1
 # This cop supports safe autocorrection (--autocorrect).
-# Configuration parameters: EnforcedStyle, IndentationWidth.
-# SupportedStyles: with_first_element, with_fixed_indentation
-Layout/ArrayAlignment:
-  Exclude:
-    - 'lib/canon/diff/path_builder.rb'
-
-# Offense count: 6
-# This cop supports safe autocorrection (--autocorrect).
-Layout/ElseAlignment:
-  Exclude:
-    - 'lib/canon/diff_formatter/diff_detail_formatter/node_utils.rb'
-
-# Offense count: 2
-# This cop supports safe autocorrection (--autocorrect).
 # Configuration parameters: EnforcedStyleAlignWith.
-# SupportedStylesAlignWith: keyword, variable, start_of_line
-Layout/EndAlignment:
-  Exclude:
-    - 'lib/canon/diff/path_builder.rb'
-    - 'lib/canon/diff_formatter/diff_detail_formatter/node_utils.rb'
-
-# Offense count: 1
-# This cop supports safe autocorrection (--autocorrect).
-# Configuration parameters: AllowForAlignment, AllowBeforeTrailingComments, ForceEqualSignAlignment.
-Layout/ExtraSpacing:
+# SupportedStylesAlignWith: either, start_of_block, start_of_line
+Layout/BlockAlignment:
   Exclude:
-    - 'lib/canon/diff_formatter/diff_detail_formatter/text_utils.rb'
+    - 'spec/canon/diff/diff_node_mapper_comments_spec.rb'
 
 # Offense count: 1
 # This cop supports safe autocorrection (--autocorrect).
-# Configuration parameters: EnforcedStyle.
-# SupportedStyles: normal, indented_internal_methods
-Layout/IndentationConsistency:
+Layout/BlockEndNewline:
   Exclude:
-    - 'lib/canon/diff/path_builder.rb'
+    - 'spec/canon/diff/diff_node_mapper_comments_spec.rb'
 
-# Offense count: 8
+# Offense count: 2
 # This cop supports safe autocorrection (--autocorrect).
 # Configuration parameters: Width, EnforcedStyleAlignWith, AllowedPatterns.
 # SupportedStylesAlignWith: start_of_line, relative_to_receiver
 Layout/IndentationWidth:
   Exclude:
-    - 'lib/canon/diff/path_builder.rb'
-    - 'lib/canon/diff_formatter/diff_detail_formatter/node_utils.rb'
+    - 'spec/canon/diff/diff_node_mapper_comments_spec.rb'
 
-# Offense count: 841
+# Offense count: 858
 # This cop supports safe autocorrection (--autocorrect).
 # Configuration parameters: Max, AllowHeredoc, AllowURI, AllowQualifiedName, URISchemes, AllowRBSInlineAnnotation, AllowCopDirectives, AllowedPatterns, SplitStrings.
 # URISchemes: http, https
 Layout/LineLength:
   Enabled: false
 
-# Offense count: 3
-# This cop supports safe autocorrection (--autocorrect).
-# Configuration parameters: AllowInHeredoc.
-Layout/TrailingWhitespace:
-  Exclude:
-    - 'lib/canon/diff/path_builder.rb'
-    - 'lib/canon/diff_formatter/diff_detail_formatter/dimension_formatter.rb'
-
-# Offense count: 49
+# Offense count: 48
 # Configuration parameters: IgnoreLiteralBranches, IgnoreConstantBranches, IgnoreDuplicateElseBranch.
 Lint/DuplicateBranch:
   Enabled: false
@@ -130,19 +89,19 @@ Metrics/AbcSize:
 # Configuration parameters: CountComments, CountAsOne, AllowedMethods, AllowedPatterns, inherit_mode.
 # AllowedMethods: refine
 Metrics/BlockLength:
-  Max: 84
+  Max: 85
 
-# Offense count: 196
+# Offense count: 198
 # Configuration parameters: AllowedMethods, AllowedPatterns, Max.
 Metrics/CyclomaticComplexity:
   Enabled: false
 
-# Offense count: 405
+# Offense count: 413
 # Configuration parameters: CountComments, CountAsOne, AllowedMethods, AllowedPatterns.
 Metrics/MethodLength:
-  Max: 95
+  Max: 104
 
-# Offense count: 45
+# Offense count: 46
 # Configuration parameters: CountKeywordArgs, MaxOptionalParameters.
 Metrics/ParameterLists:
   Max: 9
@@ -171,14 +130,15 @@ Naming/VariableNumber:
     - 'lib/canon/comparison/markup_comparator.rb'
     - 'lib/canon/comparison/xml_comparator/diff_node_builder.rb'
 
-# Offense count: 2
+# Offense count: 3
 # Configuration parameters: MinSize.
 Performance/CollectionLiteralInLoop:
   Exclude:
     - 'lib/canon/comparison/html_comparator.rb'
     - 'lib/canon/xml/xml_base_handler.rb'
+    - 'spec/canon/diff/diff_node_mapper_comments_spec.rb'
 
-# Offense count: 68
+# Offense count: 71
 # Configuration parameters: Prefixes, AllowedPatterns.
 # Prefixes: when, with, without
 RSpec/ContextWording:
@@ -195,7 +155,7 @@ RSpec/DescribeMethod:
     - 'spec/canon/comparison/multiple_differences_spec.rb'
     - 'spec/canon/diff_formatter/character_map_customization_spec.rb'
 
-# Offense count: 699
+# Offense count: 707
 # Configuration parameters: CountAsOne.
 RSpec/ExampleLength:
   Max: 43
@@ -247,7 +207,7 @@ RSpec/MultipleDescribes:
   Exclude:
     - 'spec/canon/comparison/match_options_spec.rb'
 
-# Offense count: 537
+# Offense count: 546
 RSpec/MultipleExpectations:
   Max: 15
 
@@ -310,6 +270,17 @@ RSpec/VerifiedDoubles:
     - 'spec/canon/diff_formatter/diff_detail_formatter_spec.rb'
     - 'spec/canon/tree_diff/operation_converter_spec.rb'
 
+# Offense count: 1
+# This cop supports safe autocorrection (--autocorrect).
+# Configuration parameters: EnforcedStyle, ProceduralMethods, FunctionalMethods, AllowedMethods, AllowedPatterns, AllowBracesOnProceduralOneLiners, BracesRequiredMethods.
+# SupportedStyles: line_count_based, semantic, braces_for_chaining, always_braces
+# ProceduralMethods: benchmark, bm, bmbm, create, each_with_object, measure, new, realtime, tap, with_object
+# FunctionalMethods: let, let!, subject, watch
+# AllowedMethods: lambda, proc, it
+Style/BlockDelimiters:
+  Exclude:
+    - 'spec/canon/diff/diff_node_mapper_comments_spec.rb'
+
 # Offense count: 1
 # This cop supports safe autocorrection (--autocorrect).
 # Configuration parameters: EnforcedStyle, AllowComments.
@@ -332,37 +303,9 @@ Style/IdenticalConditionalBranches:
     - 'lib/canon/diff_formatter/by_object/base_formatter.rb'
     - 'lib/canon/diff_formatter/legend.rb'
 
-# Offense count: 2
-# This cop supports safe autocorrection (--autocorrect).
-# Configuration parameters: AllowMethodComparison, ComparisonsThreshold.
-Style/MultipleComparison:
-  Exclude:
-    - 'lib/canon/diff/path_builder.rb'
-
 # Offense count: 1
 # Configuration parameters: AllowedMethods.
 # AllowedMethods: respond_to_missing?
 Style/OptionalBooleanParameter:
   Exclude:
     - 'lib/canon/diff_formatter/debug_output.rb'
-
-# Offense count: 1
-# This cop supports safe autocorrection (--autocorrect).
-Style/RedundantParentheses:
-  Exclude:
-    - 'lib/canon/diff/path_builder.rb'
-
-# Offense count: 1
-# This cop supports safe autocorrection (--autocorrect).
-# Configuration parameters: AllowModifier.
-Style/SoleNestedConditional:
-  Exclude:
-    - 'lib/canon/diff/path_builder.rb'
-
-# Offense count: 3
-# This cop supports safe autocorrection (--autocorrect).
-# Configuration parameters: EnforcedStyleForMultiline.
-# SupportedStylesForMultiline: comma, consistent_comma, diff_comma, no_comma
-Style/TrailingCommaInArguments:
-  Exclude:
-    - 'spec/canon/diff_formatter/diff_detail_formatter_spec.rb'

data/docs/understanding/formats/xml.adoc

@@ -208,6 +208,23 @@ differently depending on the operation:
 This means documents with and without XML declarations are considered
 equivalent when using the comparison API.
 
+==== Case-insensitive declaration attributes
+
+Per the XML specification, certain XML declaration attributes have
+case-insensitive values. Canon normalizes these during formatting
+comparison so that differences in letter casing are treated as
+formatting-only rather than normative differences.
+
+**Case-insensitive attributes:**
+
+* `encoding` — e.g., `encoding="UTF-8"` equals `encoding="utf-8"`
+* `standalone` — e.g., `standalone="Yes"` equals `standalone="yes"`
+
+These attributes are normalized to lowercase during formatting-level
+comparison in `Canon::Diff::FormattingDetector`. This applies when
+determining whether a line-level difference is formatting-only (e.g.,
+in `by_line` diff mode).
+
 === Comment handling
 
 XML comments are preserved in canonical form unless `--with-comments` is explicitly set.

data/lib/canon/comparison/markup_comparator.rb

@@ -93,7 +93,9 @@ module Canon
           elsif node.is_a?(Canon::Xml::Nodes::ElementNode)
             serialize_element_node(node)
           elsif node.is_a?(Canon::Xml::Nodes::TextNode)
-            node.value
+            # Use original text (with entity references) if available,
+            # otherwise fall back to value (decoded text)
+            node.original || node.value
           elsif node.is_a?(Canon::Xml::Nodes::CommentNode)
             "<!--#{node.value}-->"
           elsif node.is_a?(Canon::Xml::Nodes::ProcessingInstructionNode)

data/lib/canon/comparison/xml_comparator/child_comparison.rb

@@ -143,8 +143,11 @@ diff_children, differences)
             children1, children2, _parent_node, comparator,
             opts, child_opts, diff_children, differences
           )
+            has_mismatch = false
+
             # Length check
             unless children1.length == children2.length
+              has_mismatch = true
               dimension = determine_dimension_for_mismatch(
                 children1, children2, comparator
               )
@@ -154,28 +157,20 @@ diff_children, differences)
                   children1, children2, comparator
                 )
 
-              # Skip creating parent-level difference for comments
-              #  when comments: :ignore
-              # The child comparison will handle the comment vs
-              # element comparison
-              # This avoids creating duplicate differences
-              match_opts = opts[:match_opts]
-              unless dimension == :comments && match_opts && match_opts[:comments] == :ignore
-                if mismatched_children.empty?
-                  comparator.send(:add_difference, parent_node, parent_node,
-                                  Comparison::MISSING_NODE, Comparison::MISSING_NODE,
-                                  dimension, opts, differences)
-                else
-                  mismatched_children.each do |child|
-                    if children1.length > children2.length # rubocop:disable Metrics/BlockNesting
-                      comparator.send(:add_difference, child, nil,
-                                      Comparison::MISSING_NODE, Comparison::MISSING_NODE,
-                                      dimension, opts, differences)
-                    else
-                      comparator.send(:add_difference, nil, child,
-                                      Comparison::MISSING_NODE, Comparison::MISSING_NODE,
-                                      dimension, opts, differences)
-                    end
+              if mismatched_children.empty?
+                comparator.send(:add_difference, parent_node, parent_node,
+                                Comparison::MISSING_NODE, Comparison::MISSING_NODE,
+                                dimension, opts, differences)
+              else
+                mismatched_children.each do |child|
+                  if children1.length > children2.length # rubocop:disable Metrics/BlockNesting
+                    comparator.send(:add_difference, child, nil,
+                                    Comparison::MISSING_NODE, Comparison::MISSING_NODE,
+                                    dimension, opts, differences)
+                  else
+                    comparator.send(:add_difference, nil, child,
+                                    Comparison::MISSING_NODE, Comparison::MISSING_NODE,
+                                    dimension, opts, differences)
                   end
                 end
               end
@@ -184,7 +179,7 @@ diff_children, differences)
             end
 
             # Compare children pairwise by position
-            result = Comparison::EQUIVALENT
+            result = has_mismatch ? Comparison::UNEQUAL_ELEMENTS : Comparison::EQUIVALENT
             children1.zip(children2).each do |child1, child2|
               # Skip if one is nil (due to different lengths)
               next if child1.nil? || child2.nil?
@@ -220,8 +215,12 @@ diff_children, differences)
               elsif !comparator.send(:same_node_type?, children1[i],
                                      children2[i])
                 # Different node types at same position
-                dimension = comparator.send(:determine_node_dimension,
-                                            children1[i])
+                # Check both nodes - if either is a comment, use :comments dimension
+                dim1 = comparator.send(:determine_node_dimension,
+                                       children1[i])
+                dim2 = comparator.send(:determine_node_dimension,
+                                       children2[i])
+                dimension = [dim1, dim2].include?(:comments) ? :comments : dim1
                 break
               end
             end

data/lib/canon/comparison.rb

@@ -224,9 +224,33 @@ module Canon
           raise Canon::CompareFormatMismatchError.new(format1, format2)
         end
 
+        # Get global config options if not defined in opts
+        # This is needed because semantic_diff doesn't go through dom_diff's config handling
+        if !(opts[:match_profile] || opts[:global_options]) && Canon::Config.instance.respond_to?(format1)
+          format_config = Canon::Config.instance.public_send(format1)
+          if format_config.match.profile
+            opts[:match_profile] =
+              format_config.match.profile
+          end
+          if format_config.match.options && !format_config.match.options.empty?
+            opts[:global_options] =
+              format_config.match.options
+          end
+        end
+
         # Resolve match options for the format
         match_opts_hash = resolve_match_options(format1, opts)
 
+        # Also read diff options from config (e.g., max_node_count for large documents)
+        # This is independent of match options and needs to be passed to TreeDiffIntegrator
+        if !match_opts_hash[:max_node_count] && Canon::Config.instance.respond_to?(format1)
+          diff_max_node = Canon::Config.instance.public_send(format1).diff.max_node_count
+          if diff_max_node > 10_000
+            match_opts_hash[:max_node_count] =
+              diff_max_node
+          end
+        end
+
         # Delegate parsing to comparators (reuses existing preprocessing logic)
         doc1, doc2 = parse_with_comparator(obj1, obj2, format1, match_opts_hash)
 
@@ -243,11 +267,21 @@ module Canon
         # Perform diff
         tree_diff_result = integrator.diff(doc1, doc2)
 
+        # Extract only match-related keys for OperationConverter and SemanticTreeMatchStrategy
+        # These components expect match options, not diff options like max_node_count
+        match_only_keys = %i[match_profile match preprocessing
+                             text_content structural_whitespace attribute_presence
+                             attribute_order attribute_values element_position
+                             comments format similarity_threshold hash_matching
+                             similarity_matching propagation sensitive_elements
+                             whitespace_sensitive_elements respect_xml_space]
+        match_options_only = match_opts_hash.slice(*match_only_keys)
+
         # Convert operations to DiffNodes for unified pipeline
         # CRITICAL: Use match_opts_hash (resolved options with profile) not opts[:match]
         converter = Canon::TreeDiff::OperationConverter.new(
           format: format1,
-          match_options: match_opts_hash,
+          match_options: match_options_only,
         )
         diff_nodes = converter.convert(tree_diff_result[:operations])
 
@@ -255,7 +289,7 @@ module Canon
         # This matches DOM diff preprocessing pattern (xml_comparator.rb:106-109)
         require_relative "comparison/strategies/semantic_tree_match_strategy"
         strategy = Comparison::Strategies::SemanticTreeMatchStrategy.new(
-          format: format1, match_options: match_opts_hash,
+          format: format1, match_options: match_options_only,
         )
         str1, str2 = strategy.preprocess_for_display(doc1, doc2)
 

data/lib/canon/diff/diff_line.rb

@@ -5,16 +5,20 @@ module Canon
     # Represents a single line in the diff output
     # Links textual representation to semantic DiffNode
     class DiffLine
-      attr_reader :line_number, :content, :type, :diff_node
+      attr_reader :line_number, :new_position, :content, :type, :diff_node
+      attr_writer :formatting
 
-      # @param line_number [Integer] The line number in the original text
+      # @param line_number [Integer] The 0-based line number in text1 (old text)
+      # @param new_position [Integer, nil] The 0-based line number in text2 (new text),
+      #   used for :changed lines where old and new positions differ
       # @param content [String] The text content of the line
       # @param type [Symbol] The type of line (:unchanged, :added, :removed, :changed)
       # @param diff_node [DiffNode, nil] The semantic diff node this line belongs to
       # @param formatting [Boolean] Whether this is a formatting-only difference
       def initialize(line_number:, content:, type:, diff_node: nil,
-formatting: false)
+                     formatting: false, new_position: nil)
         @line_number = line_number
+        @new_position = new_position
         @content = content
         @type = type
         @diff_node = diff_node
@@ -71,6 +75,7 @@ formatting: false)
       def to_h
         {
           line_number: line_number,
+          new_position: new_position,
           content: content,
           type: type,
           diff_node: diff_node&.to_h,
@@ -83,6 +88,7 @@ formatting: false)
       def ==(other)
         other.is_a?(DiffLine) &&
           line_number == other.line_number &&
+          new_position == other.new_position &&
           content == other.content &&
           type == other.type &&
           diff_node == other.diff_node &&