canon 0.1.8 → 0.1.9

data/scripts/compare_failures.rb

@@ -1,151 +0,0 @@
-#!/usr/bin/env ruby
-# frozen_string_literal: true
-
-# Compare semantic and DOM algorithm failures to identify false positives/negatives
-# Usage: ruby scripts/compare_failures.rb /tmp/semantic_current.txt DOM_DIFF_RESULTS.md
-
-require "set"
-
-def parse_semantic_failures(file)
-  failures = []
-  File.readlines(file).each do |line|
-    # Format: "rspec ./spec/isodoc/blocks_notes_spec.rb:494 # ..."
-    if line =~ %r{rspec \./spec/isodoc/([a-z_]+_spec\.rb):(\d+)}
-      failures << { file: $1, line: $2.to_i }
-    end
-  end
-  failures
-end
-
-def parse_dom_failures(file)
-  failures = []
-  in_failures = false
-
-  File.readlines(file).each do |line|
-    if line.include?("Failed examples:")
-      in_failures = true
-      next
-    end
-
-    next unless in_failures
-
-    # Stop at Coverage report
-    break if line.include?("Coverage report")
-
-    # Format: "rspec ./spec/isodoc/blocks_notes_spec.rb:494"
-    if line =~ %r{rspec \./spec/isodoc/([a-z_]+_spec\.rb):(\d+)}
-      failures << { file: $1, line: $2.to_i }
-    end
-  end
-
-  failures
-end
-
-def categorize_failures(semantic, dom)
-  semantic_set = Set.new(semantic.map { |f| "#{f[:file]}:#{f[:line]}" })
-  dom_set = Set.new(dom.map { |f| "#{f[:file]}:#{f[:line]}" })
-
-  {
-    false_positives: semantic_set - dom_set, # Semantic fails, DOM passes
-    false_negatives: dom_set - semantic_set, # DOM fails, Semantic passes
-    common: semantic_set & dom_set, # Both fail (real failures)
-  }
-end
-
-def group_by_spec(failures)
-  failures.group_by { |f| f.split(":").first }.transform_values(&:count)
-end
-
-def main
-  semantic_file = ARGV[0] || "/tmp/semantic_current.txt"
-  dom_file = ARGV[1] || "DOM_DIFF_RESULTS.md"
-
-  puts "Parsing semantic failures from: #{semantic_file}"
-  semantic = parse_semantic_failures(semantic_file)
-
-  puts "Parsing DOM failures from: #{dom_file}"
-  dom = parse_dom_failures(dom_file)
-
-  puts "\n#{'=' * 80}"
-  puts "FAILURE COMPARISON SUMMARY"
-  puts "=" * 80
-
-  puts "\nTotal failures:"
-  puts "  Semantic: #{semantic.size}"
-  puts "  DOM:      #{dom.size}"
-
-  categories = categorize_failures(semantic, dom)
-
-  puts "\n#{'-' * 80}"
-  puts "FALSE POSITIVES (Semantic fails, DOM passes) - #{categories[:false_positives].size}"
-  puts "-" * 80
-  puts "\nBy spec file:"
-  group_by_spec(categories[:false_positives].to_a).sort_by do |_, v|
-    -v
-  end.each do |file, count|
-    puts "  #{file}: #{count}"
-  end
-
-  puts "\nDetailed list:"
-  categories[:false_positives].sort.each do |failure|
-    puts "  #{failure}"
-  end
-
-  puts "\n#{'-' * 80}"
-  puts "FALSE NEGATIVES (DOM fails, Semantic passes) - #{categories[:false_negatives].size}"
-  puts "-" * 80
-  puts "\nBy spec file:"
-  group_by_spec(categories[:false_negatives].to_a).sort_by do |_, v|
-    -v
-  end.each do |file, count|
-    puts "  #{file}: #{count}"
-  end
-
-  puts "\nDetailed list:"
-  categories[:false_negatives].sort.each do |failure|
-    puts "  #{failure}"
-  end
-
-  puts "\n#{'-' * 80}"
-  puts "COMMON FAILURES (Both algorithms fail) - #{categories[:common].size}"
-  puts "-" * 80
-  puts "\nBy spec file:"
-  group_by_spec(categories[:common].to_a).sort_by do |_, v|
-    -v
-  end.each do |file, count|
-    puts "  #{file}: #{count}"
-  end
-
-  puts "\n#{'=' * 80}"
-  puts "NEXT STEPS"
-  puts "=" * 80
-  puts "\n1. Fix false positives (#{categories[:false_positives].size} tests):"
-  puts "   - These are cases where semantic is too strict"
-  puts "   - DOM passes but semantic fails"
-  puts "   - Fix these to reduce semantic failures"
-
-  puts "\n2. Fix false negatives (#{categories[:false_negatives].size} tests):"
-  puts "   - These are cases where semantic is too lenient"
-  puts "   - Semantic passes but DOM fails"
-  puts "   - Fix these to maintain correctness"
-
-  puts "\n3. Common failures (#{categories[:common].size} tests):"
-  puts "   - These are real test failures in both algorithms"
-  puts "   - Will remain after parity is achieved"
-  puts "   - May indicate actual test/code issues"
-
-  # Save detailed results
-  output_file = "/tmp/failure_comparison.txt"
-  File.open(output_file, "w") do |f|
-    f.puts "FALSE POSITIVES (#{categories[:false_positives].size}):"
-    categories[:false_positives].sort.each { |fp| f.puts fp }
-    f.puts "\nFALSE NEGATIVES (#{categories[:false_negatives].size}):"
-    categories[:false_negatives].sort.each { |fn| f.puts fn }
-    f.puts "\nCOMMON FAILURES (#{categories[:common].size}):"
-    categories[:common].sort.each { |cf| f.puts cf }
-  end
-
-  puts "\nDetailed results saved to: #{output_file}"
-end
-
-main if __FILE__ == $PROGRAM_NAME

data/scripts/debug_attribute_extraction.rb

@@ -1,66 +0,0 @@
-#!/usr/bin/env ruby
-# frozen_string_literal: true
-
-require_relative "../lib/canon"
-require_relative "../lib/canon/diff_formatter"
-require_relative "../lib/canon/diff_formatter/diff_detail_formatter"
-
-# Test attribute values formatting
-html1 = '<table id="T1" class="MsoNormalTable" border="1"></table>'
-html2 = '<table id="T2" class="MsoNormalTable" border="2"></table>'
-
-result = Canon::Comparison.equivalent?(
-  html1,
-  html2,
-  match_algorithm: :semantic_tree,
-  ignore_attr_order: true,
-  verbose: true,
-)
-
-puts "Number of differences: #{result.differences.length}"
-puts
-
-result.differences.each_with_index do |diff, i|
-  puts "=" * 70
-  puts "Difference ##{i + 1}"
-  puts "=" * 70
-  puts "Class: #{diff.class}"
-  puts "Dimension: #{diff.dimension if diff.respond_to?(:dimension)}"
-
-  if diff.respond_to?(:node1) && diff.respond_to?(:node2)
-    node1 = diff.node1
-    node2 = diff.node2
-
-    puts "\nNode1:"
-    puts "  Class: #{node1.class}"
-    puts "  Name: #{node1.name if node1.respond_to?(:name)}"
-    if node1.respond_to?(:attributes)
-      puts "  Attributes: #{node1.attributes.inspect}"
-      puts "  Attributes class: #{node1.attributes.class}"
-      puts "  Attributes keys: #{node1.attributes.keys.inspect}"
-      node1.attributes.each do |key, val|
-        puts "    #{key.inspect} (#{key.class}) => #{val.inspect} (#{val.class})"
-        if val.respond_to?(:value)
-          puts "      val.value = #{val.value.inspect}"
-        end
-      end
-    end
-
-    puts "\nNode2:"
-    puts "  Class: #{node2.class}"
-    puts "  Name: #{node2.name if node2.respond_to?(:name)}"
-    if node2.respond_to?(:attributes)
-      puts "  Attributes: #{node2.attributes.inspect}"
-      puts "  Attributes class: #{node2.attributes.class}"
-      puts "  Attributes keys: #{node2.attributes.keys.inspect}"
-      node2.attributes.each do |key, val|
-        puts "    #{key.inspect} (#{key.class}) => #{val.inspect} (#{val.class})"
-        if val.respond_to?(:value)
-          puts "      val.value = #{val.value.inspect}"
-        end
-      end
-    end
-  end
-
-  puts
-end

data/scripts/debug_blocks_839.rb

@@ -1,115 +0,0 @@
-#!/usr/bin/env ruby
-# Debug script for blocks_spec.rb:839
-
-require "bundler/setup"
-require "nokogiri"
-
-HTML_HDR = <<~HEADER.freeze
-  <html lang="en">
-    <head/>
-    <body lang="en">
-      <div class="title-section">
-        <p>\u00a0</p>
-      </div>
-      <br/>
-      <div class="prefatory-section">
-        <p>\u00a0</p>
-      </div>
-      <br/>
-      <div class="main-section">
-         <br/>
-            <div class="TOC" id="_">
-        <h1 class="IntroTitle">Table of contents</h1>
-      </div>
-HEADER
-
-WORD_HDR = <<~HEADER.freeze
-       <html xmlns:epub="http://www.idpf.org/2007/ops" xmlns:v="urn:schemas-microsoft-com:vml" xmlns:o="urn:schemas-microsoft-com:office:office" xmlns:w="urn:schemas-microsoft-com:office:word" xmlns:m="http://schemas.microsoft.com/office/2004/12/omml" lang="en">
-           <head>
-    <style>
-      <!--
-      -->
-    </style>
-  </head>
-         <body lang="EN-US" link="blue" vlink="#954F72">
-           <div class="WordSection1">
-             <p>\u00a0</p>
-           </div>
-HEADER
-
-html5_doc = <<~HTML
-  #{HTML_HDR}
-              <br/>
-              <div id="_">
-                <h1 class="ForewordTitle">Foreword</h1>
-                <p id="_" style="text-align:left;">Test</p>
-              </div>
-            </div>
-          </body>
-      </html>
-HTML
-
-html4_doc = <<~HTML
-  #{WORD_HDR}
-  <p class="page-break">
-    <br clear="all" style="mso-special-character:line-break;page-break-before:always"/>
-  </p>
-          <div class="TOC" id="_">
-    <h1 class="IntroTitle">Table of contents</h1>
-  </div>
-  <p class="page-break">
-    <br clear="all" style="mso-special-character:line-break;page-break-before:always"/>
-  </p>
-          <div id="_">
-            <h1 class="ForewordTitle">Foreword</h1>
-            <p id="_" align="left" style="text-align:left;">Test</p>
-          </div>
-          <p>\u00a0</p>
-        </div>
-      </body>
-  </html>
-HTML
-
-puts "=" * 80
-puts "HTML5 PARSING"
-puts "=" * 80
-
-doc5 = Nokogiri::HTML5(html5_doc)
-head5 = doc5.at("//head")
-puts "HEAD element:"
-puts head5.to_html
-puts "\nCHILDREN:"
-head5.children.each_with_index do |child, i|
-  puts "  #{i}: #{child.name} - #{child.attributes.inspect}"
-end
-
-puts "\n#{'=' * 80}"
-puts "HTML4 PARSING"
-puts "=" * 80
-
-doc4 = Nokogiri::HTML4(html4_doc)
-head4 = doc4.at("//head")
-puts "HEAD element:"
-puts head4.to_html
-puts "\nCHILDREN:"
-head4.children.each_with_index do |child, i|
-  puts "  #{i}: #{child.name} - #{child.attributes.inspect}"
-end
-
-puts "\n#{'=' * 80}"
-puts "COMPARISON"
-puts "=" * 80
-puts "HTML5 head children: #{head5.children.size}"
-puts "HTML4 head children: #{head4.children.size}"
-
-meta5 = head5.xpath(".//meta")
-meta4 = head4.xpath(".//meta")
-puts "\nMETA elements:"
-puts "HTML5: #{meta5.size} meta elements"
-meta5.each_with_index do |m, i|
-  puts "  #{i}: #{m.to_html}"
-end
-puts "HTML4: #{meta4.size} meta elements"
-meta4.each_with_index do |m, i|
-  puts "  #{i}: #{m.to_html}"
-end

data/scripts/debug_meta_matching.rb

@@ -1,52 +0,0 @@
-#!/usr/bin/env ruby
-# frozen_string_literal: true
-
-require "bundler/setup"
-require "canon"
-
-# Test case: Meta element with attributes should match
-expected = <<~HTML
-  <html>
-    <head>
-      <meta http-equiv="Content-Type" content="text/html; charset=UTF-8">
-    </head>
-    <body>
-      <p>Test</p>
-    </body>
-  </html>
-HTML
-
-actual = <<~HTML
-  <html>
-    <head>
-      <meta http-equiv="Content-Type" content="text/html; charset=UTF-8">
-    </head>
-    <body>
-      <p>Test</p>
-    </body>
-  </html>
-HTML
-
-puts "=" * 80
-puts "Testing Meta Element Matching"
-puts "=" * 80
-
-result = Canon::Comparison.equivalent?(expected, actual,
-                                       format: :html4,
-                                       diff_algorithm: :semantic,
-                                       verbose: true)
-
-if result.is_a?(Canon::Comparison::ComparisonResult)
-  puts "\nResult: #{result.equivalent? ? 'PASS ✅' : 'FAIL ❌'}"
-  puts "Normative diffs: #{result.normative_differences.count}"
-  puts "Total diffs: #{result.differences.count}"
-
-  unless result.equivalent?
-    puts "\nDifferences:"
-    result.differences.each_with_index do |diff, i|
-      puts "\n  #{i + 1}. #{diff.inspect}"
-    end
-  end
-else
-  puts "Result: #{result}"
-end

data/scripts/debug_p_matching.rb

@@ -1,192 +0,0 @@
-#!/usr/bin/env ruby
-# frozen_string_literal: true
-
-require "bundler/setup"
-require "canon"
-require "nokogiri"
-
-# Read the test files
-expected_file = "/Users/mulgogi/src/mn/isodoc/spec/fixtures/html/isodoc-section-names-expected.html"
-actual_file = "/Users/mulgogi/src/mn/isodoc/spec/fixtures/html/isodoc-section-names-actual.html"
-
-expected = File.read(expected_file)
-actual = File.read(actual_file)
-
-puts "=" * 80
-puts "ANALYZING <p> ELEMENT MATCHING"
-puts "=" * 80
-
-# Parse with Nokogiri to see what we have
-doc1 = Nokogiri::HTML4(expected)
-doc2 = Nokogiri::HTML4(actual)
-
-# Find all <p> elements
-p_elements1 = doc1.css("p")
-p_elements2 = doc2.css("p")
-
-puts "\nFile 1 has #{p_elements1.size} <p> elements"
-puts "File 2 has #{p_elements2.size} <p> elements"
-
-# Group by class attribute
-p_by_class1 = p_elements1.group_by { |p| p["class"] }
-p_by_class2 = p_elements2.group_by { |p| p["class"] }
-
-puts "\nFile 1 <p> elements by class:"
-p_by_class1.each do |klass, elements|
-  puts "  #{klass.inspect}: #{elements.size} elements"
-  elements.first(3).each do |el|
-    content = el.text.strip
-    content = "#{content[0..50]}..." if content.length > 50
-    puts "    - #{content.inspect}"
-  end
-end
-
-puts "\nFile 2 <p> elements by class:"
-p_by_class2.each do |klass, elements|
-  puts "  #{klass.inspect}: #{elements.size} elements"
-  elements.first(3).each do |el|
-    content = el.text.strip
-    content = "#{content[0..50]}..." if content.length > 50
-    puts "    - #{content.inspect}"
-  end
-end
-
-# Now run Canon's tree diff to see what happens
-puts "\n#{'=' * 80}"
-puts "RUNNING CANON TREE DIFF"
-puts "=" * 80
-
-require_relative "../lib/canon/tree_diff/adapters/html_adapter"
-require_relative "../lib/canon/tree_diff/matchers/hash_matcher"
-require_relative "../lib/canon/tree_diff/matchers/similarity_matcher"
-require_relative "../lib/canon/tree_diff/operations/operation_detector"
-
-# Create trees
-adapter = Canon::TreeDiff::Adapters::HtmlAdapter.new
-tree1 = adapter.parse(expected)
-tree2 = adapter.parse(actual)
-
-puts "\nTree 1 has #{tree1.descendants.size} total nodes"
-puts "Tree 2 has #{tree2.descendants.size} total nodes"
-
-# Find <p> nodes in tree
-p_nodes1 = tree1.descendants.select { |n| n.label == "p" }
-p_nodes2 = tree2.descendants.select { |n| n.label == "p" }
-
-puts "\nTree 1 has #{p_nodes1.size} <p> nodes"
-puts "Tree 2 has #{p_nodes2.size} <p> nodes"
-
-# Group by attributes
-p_by_attrs1 = p_nodes1.group_by(&:attributes)
-p_by_attrs2 = p_nodes2.group_by(&:attributes)
-
-puts "\nTree 1 <p> nodes by attributes:"
-p_by_attrs1.each do |attrs, nodes|
-  puts "  #{attrs.inspect}: #{nodes.size} nodes"
-end
-
-puts "\nTree 2 <p> nodes by attributes:"
-p_by_attrs2.each do |attrs, nodes|
-  puts "  #{attrs.inspect}: #{nodes.size} nodes"
-end
-
-# Look at signatures
-require_relative "../lib/canon/tree_diff/core/node_signature"
-
-puts "\n#{'=' * 80}"
-puts "ANALYZING SIGNATURES"
-puts "=" * 80
-
-# Get page-break <p> nodes
-page_break_p1 = p_nodes1.select { |n| n.attributes["class"] == "page-break" }
-page_break_p2 = p_nodes2.select { |n| n.attributes["class"] == "page-break" }
-
-puts "\nFile 1 has #{page_break_p1.size} <p class=\"page-break\"> nodes"
-puts "File 2 has #{page_break_p2.size} <p class=\"page-break\"> nodes"
-
-if page_break_p1.any?
-  puts "\nFirst 3 signatures from File 1:"
-  page_break_p1.first(3).each_with_index do |node, i|
-    sig = Canon::TreeDiff::Core::NodeSignature.for(node)
-    puts "  #{i + 1}. #{sig}"
-    puts "     Children: #{node.children.size}"
-    if node.children.any?
-      node.children.each do |child|
-        child_sig = Canon::TreeDiff::Core::NodeSignature.for(child)
-        puts "       - #{child.label}: #{child_sig}"
-      end
-    end
-  end
-end
-
-if page_break_p2.any?
-  puts "\nFirst 3 signatures from File 2:"
-  page_break_p2.first(3).each_with_index do |node, i|
-    sig = Canon::TreeDiff::Core::NodeSignature.for(node)
-    puts "  #{i + 1}. #{sig}"
-    puts "     Children: #{node.children.size}"
-    if node.children.any?
-      node.children.each do |child|
-        child_sig = Canon::TreeDiff::Core::NodeSignature.for(child)
-        puts "       - #{child.label}: #{child_sig}"
-      end
-    end
-  end
-end
-
-# Run hash matcher
-puts "\n#{'=' * 80}"
-puts "RUNNING HASH MATCHER"
-puts "=" * 80
-
-options = {
-  attribute_order: :ignore,
-  text_content: :normalize,
-}
-
-matcher = Canon::TreeDiff::Matchers::HashMatcher.new(tree1, tree2, options)
-matching = matcher.match
-
-puts "\nTotal matched pairs: #{matching.size}"
-
-# Check how many <p> nodes were matched
-matched_p1 = p_nodes1.count { |n| matching.matched1?(n) }
-matched_p2 = p_nodes2.count { |n| matching.matched2?(n) }
-
-puts "Matched <p> from tree1: #{matched_p1}/#{p_nodes1.size}"
-puts "Matched <p> from tree2: #{matched_p2}/#{p_nodes2.size}"
-
-matched_page_break_p1 = page_break_p1.count { |n| matching.matched1?(n) }
-matched_page_break_p2 = page_break_p2.count { |n| matching.matched2?(n) }
-
-puts "Matched <p class=\"page-break\"> from tree1: #{matched_page_break_p1}/#{page_break_p1.size}"
-puts "Matched <p class=\"page-break\"> from tree2: #{matched_page_break_p2}/#{page_break_p2.size}"
-
-# Check unmatched page-break <p> nodes
-unmatched_p1 = page_break_p1.reject { |n| matching.matched1?(n) }
-unmatched_p2 = page_break_p2.reject { |n| matching.matched2?(n) }
-
-puts "\nUnmatched <p class=\"page-break\"> from tree1: #{unmatched_p1.size}"
-puts "Unmatched <p class=\"page-break\"> from tree2: #{unmatched_p2.size}"
-
-if unmatched_p1.any?
-  puts "\nFirst unmatched from tree1:"
-  node = unmatched_p1.first
-  puts "  Path: #{node.xpath}"
-  puts "  Signature: #{Canon::TreeDiff::Core::NodeSignature.for(node)}"
-  puts "  Children: #{node.children.size}"
-  node.children.each do |child|
-    puts "    - #{child.label}: value=#{child.value.inspect}, attrs=#{child.attributes.inspect}"
-  end
-end
-
-if unmatched_p2.any?
-  puts "\nFirst unmatched from tree2:"
-  node = unmatched_p2.first
-  puts "  Path: #{node.xpath}"
-  puts "  Signature: #{Canon::TreeDiff::Core::NodeSignature.for(node)}"
-  puts "  Children: #{node.children.size}"
-  node.children.each do |child|
-    puts "    - #{child.label}: value=#{child.value.inspect}, attrs=#{child.attributes.inspect}"
-  end
-end

data/scripts/debug_signature_matching.rb

@@ -1,118 +0,0 @@
-#!/usr/bin/env ruby
-# frozen_string_literal: true
-
-# Debug signature matching to understand why elements aren't matching
-# Usage: ruby scripts/debug_signature_matching.rb
-
-require_relative "../lib/canon"
-require "nokogiri"
-
-# Sample XML with semx elements that should match
-xml1 = <<~XML
-  <p>
-    <fmt-concept>
-      <semx element="title" source="_">word</semx>
-    </fmt-concept>
-  </p>
-XML
-
-xml2 = <<~XML
-  <p>
-    <fmt-concept>
-      <semx element="concept" source="_">word</semx>
-    </fmt-concept>
-  </p>
-XML
-
-puts "=" * 80
-puts "SIGNATURE MATCHING DEBUG"
-puts "=" * 80
-
-# Parse both
-doc1 = Nokogiri::XML(xml1)
-doc2 = Nokogiri::XML(xml2)
-
-# Create adapter
-adapter = Canon::TreeDiff::Adapters::XMLAdapter.new
-
-# Convert to tree
-tree1 = adapter.to_tree(doc1.root)
-tree2 = adapter.to_tree(doc2.root)
-
-puts "\nTree 1 structure:"
-def print_tree(node, indent = 0)
-  prefix = "  " * indent
-  if node.text?
-    puts "#{prefix}#text: #{node.value.inspect}"
-  else
-    attrs = node.attributes.empty? ? "" : " {#{node.attributes.inspect}}"
-    puts "#{prefix}<#{node.label}>#{attrs}"
-    node.children.each { |c| print_tree(c, indent + 1) }
-  end
-end
-
-print_tree(tree1)
-
-puts "\nTree 2 structure:"
-print_tree(tree2)
-
-# Get semx nodes
-semx1 = tree1.descendants.find { |n| n.label == "semx" }
-semx2 = tree2.descendants.find { |n| n.label == "semx" }
-
-puts "\n#{'-' * 80}"
-puts "SEMX NODE COMPARISON"
-puts "-" * 80
-
-puts "\nSemx 1:"
-puts "  Label: #{semx1.label}"
-puts "  Value: #{semx1.value.inspect}"
-puts "  Attributes: #{semx1.attributes.inspect}"
-
-puts "\nSemx 2:"
-puts "  Label: #{semx2.label}"
-puts "  Value: #{semx2.value.inspect}"
-puts "  Attributes: #{semx2.attributes.inspect}"
-
-# Compute signatures
-sig1_strict = Canon::TreeDiff::Core::NodeSignature.for(semx1,
-                                                       include_attributes: true)
-sig2_strict = Canon::TreeDiff::Core::NodeSignature.for(semx2,
-                                                       include_attributes: true)
-
-sig1_loose = Canon::TreeDiff::Core::NodeSignature.for(semx1,
-                                                      include_attributes: false)
-sig2_loose = Canon::TreeDiff::Core::NodeSignature.for(semx2,
-                                                      include_attributes: false)
-
-puts "\n#{'-' * 80}"
-puts "SIGNATURE COMPARISON"
-puts "-" * 80
-
-puts "\nStrict signatures (with attributes):"
-puts "  Semx 1: #{sig1_strict.signature_string}"
-puts "  Semx 2: #{sig2_strict.signature_string}"
-puts "  Match? #{sig1_strict == sig2_strict}"
-
-puts "\nLoose signatures (without attributes):"
-puts "  Semx 1: #{sig1_loose.signature_string}"
-puts "  Semx 2: #{sig2_loose.signature_string}"
-puts "  Match? #{sig1_loose == sig2_loose}"
-
-puts "\n#{'-' * 80}"
-puts "ANALYSIS"
-puts "-" * 80
-
-if sig1_strict != sig2_strict
-  puts "\n⚠️  ISSUE FOUND:"
-  puts "Strict signatures don't match due to attribute differences!"
-  puts "This prevents HashMatcher from considering these nodes as candidates."
-  puts "\nDifference:"
-  puts "  File 1: element='title'"
-  puts "  File 2: element='concept'"
-  puts "\nSOLUTION:"
-  puts "HashMatcher should use LOOSE signatures (no attributes) to find candidates,"
-  puts "then check attributes separately during matching."
-end
-
-puts "\n#{'=' * 80}"