canon 0.1.22 → 0.1.23

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: ba5fabd7a7d0057f365952f8aecc01381ef88f2f1dd28679e9e8947945b54104
-  data.tar.gz: 3e4119f7fc69a7c2534957c7654b6a567a868208179262fc921ea1266b5a1a6c
+  metadata.gz: '08086e588946bb0f5f39a8235984058b5a5e8aea956b9652e0e366ff84500232'
+  data.tar.gz: bb078d67c1a68da1ca88b3d29fde336e291b0ff95186ade125ee1ccfb1a82de4
 SHA512:
-  metadata.gz: 44a74c7bb2010a2aa22855258affa6889003585edba717538bb579c63f6694757b21b74bc929b07bd0f87660f0bb3bd02f158b0a91be18017e938100f1fee491
-  data.tar.gz: be61d26774ae22a5547b2ba0ec1d578b11d37be0899add26d7f14ac63c197637568b74a9d0cad76535e5ab6a82d297af7b917f225f972361f09549964917576b
+  metadata.gz: 35f7888d948f86fa4b504b060ee6d9976dc38b32232e8c86b9b8986ee08448cdad265fe4d091e810e5f63f0b0702e125317e07d15761b5cd6404905190546234
+  data.tar.gz: 4f57564289fdff8c73020f3ed9f9184b363688062cf1b89baea79f0d25c57905bcbc225b5e946afeb30305b91402b557ae65ce61b99a0d63b8bb1c569ec2d57f

data/.rubocop_todo.yml

@@ -1,6 +1,6 @@
 # This configuration was generated by
 # `rubocop --auto-gen-config`
-# on 2026-03-31 12:44:31 UTC using RuboCop version 1.86.0.
+# on 2026-04-01 00:03:16 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
@@ -11,20 +11,32 @@ Gemspec/RequiredRubyVersion:
   Exclude:
     - 'canon.gemspec'
 
-# Offense count: 1
+# Offense count: 2
 # This cop supports safe autocorrection (--autocorrect).
-# Configuration parameters: EmptyLineBetweenMethodDefs, EmptyLineBetweenClassDefs, EmptyLineBetweenModuleDefs, DefLikeMacros, AllowAdjacentOneLineDefs, NumberOfEmptyLines.
-Layout/EmptyLineBetweenDefs:
+# Configuration parameters: EnforcedStyleAlignWith.
+# SupportedStylesAlignWith: either, start_of_block, start_of_line
+Layout/BlockAlignment:
   Exclude:
-    - 'lib/canon/diff_formatter/by_line/xml_formatter.rb'
+    - 'lib/canon/tree_diff/matchers/hash_matcher.rb'
+    - 'spec/canon/hash_matcher_regression_spec.rb'
 
-# Offense count: 1
+# Offense count: 2
 # This cop supports safe autocorrection (--autocorrect).
-Layout/EmptyLines:
+Layout/BlockEndNewline:
   Exclude:
-    - 'lib/canon/diff_formatter/by_line/xml_formatter.rb'
+    - 'lib/canon/tree_diff/matchers/hash_matcher.rb'
+    - 'spec/canon/hash_matcher_regression_spec.rb'
+
+# Offense count: 4
+# This cop supports safe autocorrection (--autocorrect).
+# Configuration parameters: Width, EnforcedStyleAlignWith, AllowedPatterns.
+# SupportedStylesAlignWith: start_of_line, relative_to_receiver
+Layout/IndentationWidth:
+  Exclude:
+    - 'lib/canon/tree_diff/matchers/hash_matcher.rb'
+    - 'spec/canon/hash_matcher_regression_spec.rb'
 
-# Offense count: 1094
+# Offense count: 1103
 # This cop supports safe autocorrection (--autocorrect).
 # Configuration parameters: Max, AllowHeredoc, AllowURI, AllowQualifiedName, URISchemes, AllowRBSInlineAnnotation, AllowCopDirectives, AllowedPatterns, SplitStrings.
 # URISchemes: http, https
@@ -76,7 +88,7 @@ Lint/UselessConstantScoping:
   Exclude:
     - 'lib/canon/diff_formatter/theme.rb'
 
-# Offense count: 297
+# Offense count: 298
 # Configuration parameters: AllowedMethods, AllowedPatterns, CountRepeatedAttributes, Max.
 Metrics/AbcSize:
   Enabled: false
@@ -92,7 +104,7 @@ Metrics/BlockLength:
 Metrics/BlockNesting:
   Max: 4
 
-# Offense count: 261
+# Offense count: 262
 # Configuration parameters: AllowedMethods, AllowedPatterns, Max.
 Metrics/CyclomaticComplexity:
   Enabled: false
@@ -107,7 +119,7 @@ Metrics/MethodLength:
 Metrics/ParameterLists:
   Max: 10
 
-# Offense count: 212
+# Offense count: 213
 # Configuration parameters: AllowedMethods, AllowedPatterns, Max.
 Metrics/PerceivedComplexity:
   Enabled: false
@@ -146,7 +158,7 @@ Performance/CollectionLiteralInLoop:
 RSpec/ContextWording:
   Enabled: false
 
-# Offense count: 33
+# Offense count: 34
 # Configuration parameters: IgnoredMetadata.
 RSpec/DescribeClass:
   Enabled: false
@@ -157,7 +169,7 @@ RSpec/DescribeMethod:
     - 'spec/canon/comparison/multiple_differences_spec.rb'
     - 'spec/canon/diff_formatter/character_map_customization_spec.rb'
 
-# Offense count: 736
+# Offense count: 741
 # Configuration parameters: CountAsOne.
 RSpec/ExampleLength:
   Max: 44
@@ -209,7 +221,7 @@ RSpec/MultipleDescribes:
   Exclude:
     - 'spec/canon/comparison/match_options_spec.rb'
 
-# Offense count: 590
+# Offense count: 595
 RSpec/MultipleExpectations:
   Max: 15
 
@@ -273,6 +285,18 @@ RSpec/VerifiedDoubles:
     - 'spec/canon/diff_formatter/diff_detail_formatter_spec.rb'
     - 'spec/canon/tree_diff/operation_converter_spec.rb'
 
+# Offense count: 2
+# 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:
+    - 'lib/canon/tree_diff/matchers/hash_matcher.rb'
+    - 'spec/canon/hash_matcher_regression_spec.rb'
+
 # Offense count: 1
 # This cop supports safe autocorrection (--autocorrect).
 # Configuration parameters: EnforcedStyle, AllowComments.

data/lib/canon/diff_formatter/theme.rb

@@ -497,22 +497,9 @@ module Canon
 
         private
 
-        # Deep dup a hash, handling nested hashes and arrays
+        # Delegate to module-level deep_dup
         def deep_dup(obj)
-          case obj
-          when Hash
-            obj.transform_values { |v| deep_dup(v) }
-          when Array
-            obj.map { |v| deep_dup(v) }
-          when String, Symbol, Numeric, TrueClass, FalseClass, NilClass
-            obj
-          else
-            begin
-              obj.dup
-            rescue StandardError
-              obj
-            end
-          end
+          Theme.deep_dup(obj)
         end
 
         def deep_merge!(target, source)
@@ -606,9 +593,29 @@ module Canon
       # @param name [Symbol] Theme name
       # @return [Hash] Theme hash
       # @raise [ArgumentError] if theme not found
+      # Deep copy a value, handling nested hashes and arrays
+      def self.deep_dup(obj)
+        case obj
+        when Hash
+          obj.transform_values { |v| deep_dup(v) }
+        when Array
+          obj.map { |v| deep_dup(v) }
+        when String, Symbol, Numeric, TrueClass, FalseClass, NilClass
+          obj
+        else
+          begin
+            obj.dup
+          rescue StandardError
+            obj
+          end
+        end
+      end
+
       def self.[](name)
-        THEMES[name] || raise(ArgumentError,
-                              "Unknown theme: #{name}. Valid: #{THEMES.keys}")
+        theme = THEMES[name] || raise(ArgumentError,
+                                      "Unknown theme: #{name}. Valid: #{THEMES.keys}")
+        # Return a deep copy to prevent mutation of theme constants
+        deep_dup(theme)
       end
 
       # List available theme names

data/lib/canon/tree_diff/matchers/hash_matcher.rb

@@ -93,19 +93,34 @@ module Canon
           end
           return if candidates.empty?
 
-          best_match = find_best_match(node2, candidates)
-          return unless best_match
-
-          if @matching.add(best_match, node2)
-            @matched_tree1 << best_match
-            @matched_tree2 << node2
-            propagate_to_ancestors(best_match, node2)
+          # When multiple candidates have identical signatures (common with
+          # duplicate subtrees like MathML formulas), sort by sibling position
+          # proximity to prefer matching nodes at the same position within
+          # their parent. This reduces cross-matching that causes cascading
+          # prefix closure failures.
+          if candidates.size > 1
+            pos2 = node2.position || 0
+            candidates = candidates.sort_by do |c|
+              pos1 = c.position || 0
+              (pos1 - pos2).abs
+            end
           end
-        end
 
-        # @return [TreeNode, nil]
-        def find_best_match(node2, candidates)
-          candidates.find { |node1| subtrees_match?(node1, node2) }
+          # Try each candidate until one passes both subtree matching
+          # AND the prefix closure constraint in matching.add.
+          # When multiple candidates have identical subtrees (e.g., labels
+          # with the same text child), the first may fail prefix closure
+          # due to ancestor cross-matching, but a later candidate succeeds.
+          candidates.each do |candidate|
+            next unless subtrees_match?(candidate, node2)
+
+            if @matching.add(candidate, node2)
+              @matched_tree1 << candidate
+              @matched_tree2 << node2
+              propagate_to_ancestors(candidate, node2)
+              return
+            end
+          end
         end
 
         def subtrees_match?(node1, node2)

data/lib/canon/version.rb

@@ -1,5 +1,5 @@
 # frozen_string_literal: true
 
 module Canon
-  VERSION = "0.1.22"
+  VERSION = "0.1.23"
 end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: canon
 version: !ruby/object:Gem::Version
-  version: 0.1.22
+  version: 0.1.23
 platform: ruby
 authors:
 - Ribose Inc.
 autorequire:
 bindir: exe
 cert_chain: []
-date: 2026-03-31 00:00:00.000000000 Z
+date: 2026-04-01 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: diff-lcs