coradoc 2.0.4 → 2.0.5

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 69f496efd91c2d85e4e057be629f8ee00492c40497661c0af940be65f148e092
-  data.tar.gz: a2668ca45e7d23bf613d966c36b4c5da70311be0dcdbc0fba4d988844f49b65f
+  metadata.gz: ba4d5d05249449db87eabe01151f28977069d451c5d8b3069d8c7c59238e05df
+  data.tar.gz: 225bfd2be3033634b97443c5949ca0a4c9851c4a2cdfdbfe7059edbdd6d44d1d
 SHA512:
-  metadata.gz: '04584b5faee36ac872400d4d17458221931de4f733aae75839dc40f282d4f863948c9105eb80fefb84f6945639c76c55ccf62fd93d8fd442268f36698950625d'
-  data.tar.gz: 00c91cfd40f731c1fffab6ae507d326140e5c7830a8d10bf6a5bb75857ba0b69a552fc4a27d8bd3abb95d4a16159a516df62232bc538fd79fe55a7023e95af0b
+  metadata.gz: e5fce433e6b057839f9cb93efc4ac9722dcf812046667082121586a975f273da925d5a87c2cd3c46cd0456dd06f104c36df51a3f9012a043f8ec04674c7fa7ea
+  data.tar.gz: df5ee8e79dad51920ee9a5bc89ddd90f159396d2cd76c8956323b1439129ea6f9f56c2a6b2f5b2cad518c5c93fd0c04a0668006126deb1cda33d7651b6d0137a

data/.rubocop_todo.yml

@@ -1,6 +1,6 @@
 # This configuration was generated by
 # `rubocop --auto-gen-config`
-# on 2026-05-08 04:52:25 UTC using RuboCop version 1.75.8.
+# on 2026-05-08 15:00:33 UTC using RuboCop version 1.75.8.
 # 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
@@ -29,6 +29,12 @@ Layout/ClosingParenthesisIndentation:
   Exclude:
     - 'lib/coradoc/document_manipulator.rb'
 
+# Offense count: 1
+# This cop supports safe autocorrection (--autocorrect).
+Layout/EmptyLineAfterGuardClause:
+  Exclude:
+    - 'coradoc-adoc/spec/coradoc/asciidoc/integration_pipeline_spec.rb'
+
 # Offense count: 1
 # This cop supports safe autocorrection (--autocorrect).
 # Configuration parameters: EnforcedStyle, IndentationWidth.
@@ -133,7 +139,7 @@ Lint/Void:
   Exclude:
     - 'coradoc-html/lib/coradoc/html.rb'
 
-# Offense count: 236
+# Offense count: 237
 # Configuration parameters: AllowedMethods, AllowedPatterns, CountRepeatedAttributes.
 Metrics/AbcSize:
   Max: 178
@@ -142,7 +148,7 @@ Metrics/AbcSize:
 # Configuration parameters: CountComments, CountAsOne, AllowedMethods, AllowedPatterns, inherit_mode.
 # AllowedMethods: refine
 Metrics/BlockLength:
-  Max: 148
+  Max: 151
 
 # Offense count: 4
 # Configuration parameters: CountBlocks, CountModifierForms.
@@ -154,12 +160,12 @@ Metrics/BlockNesting:
 Metrics/ClassLength:
   Max: 556
 
-# Offense count: 179
+# Offense count: 180
 # Configuration parameters: AllowedMethods, AllowedPatterns.
 Metrics/CyclomaticComplexity:
   Max: 32
 
-# Offense count: 384
+# Offense count: 385
 # Configuration parameters: CountComments, CountAsOne, AllowedMethods, AllowedPatterns.
 Metrics/MethodLength:
   Max: 246
@@ -175,7 +181,7 @@ Metrics/ParameterLists:
   Max: 7
   MaxOptionalParameters: 4
 
-# Offense count: 122
+# Offense count: 123
 # Configuration parameters: AllowedMethods, AllowedPatterns.
 Metrics/PerceivedComplexity:
   Max: 27
@@ -201,11 +207,12 @@ Naming/MethodName:
   Exclude:
     - 'coradoc-markdown/lib/coradoc/markdown.rb'
 
-# Offense count: 17
+# Offense count: 20
 # Configuration parameters: MinNameLength, AllowNamesEndingInNumbers, AllowedNames, ForbiddenNames.
 # AllowedNames: as, at, by, cc, db, id, if, in, io, ip, of, on, os, pp, to
 Naming/MethodParameterName:
   Exclude:
+    - 'coradoc-adoc/spec/coradoc/asciidoc/integration_pipeline_spec.rb'
     - 'coradoc-docx/lib/coradoc/docx/transform/from_core_model.rb'
     - 'coradoc-docx/lib/coradoc/docx/transform/rules/run_rule.rb'
     - 'coradoc-html/lib/coradoc/html/converters/base.rb'
@@ -238,12 +245,6 @@ Naming/VariableName:
   Exclude:
     - 'coradoc-markdown/lib/coradoc/markdown.rb'
 
-# Offense count: 12
-# This cop supports unsafe autocorrection (--autocorrect-all).
-RSpec/BeEq:
-  Exclude:
-    - 'coradoc-adoc/spec/coradoc/asciidoc/model/element_classification_spec.rb'
-
 # Offense count: 1
 RSpec/BeforeAfterAll:
   Exclude:
@@ -252,20 +253,12 @@ RSpec/BeforeAfterAll:
     - '**/spec/support/**/*.rb'
     - 'spec/benchmark/performance_spec.rb'
 
-# Offense count: 20
+# Offense count: 21
 # Configuration parameters: IgnoredMetadata.
 RSpec/DescribeClass:
   Enabled: false
 
-# Offense count: 1
-# This cop supports unsafe autocorrection (--autocorrect-all).
-# Configuration parameters: SkipBlocks, EnforcedStyle, OnlyStaticConstants.
-# SupportedStyles: described_class, explicit
-RSpec/DescribedClass:
-  Exclude:
-    - 'coradoc-adoc/spec/coradoc/asciidoc/model/base_spec.rb'
-
-# Offense count: 485
+# Offense count: 500
 # Configuration parameters: CountAsOne.
 RSpec/ExampleLength:
   Max: 36
@@ -316,7 +309,7 @@ RSpec/MessageSpies:
 RSpec/MultipleDescribes:
   Enabled: false
 
-# Offense count: 619
+# Offense count: 628
 RSpec/MultipleExpectations:
   Max: 12
 
@@ -431,6 +424,12 @@ Style/HashLikeCase:
     - 'coradoc-html/lib/coradoc/html/input/converters/table.rb'
     - 'coradoc-html/lib/coradoc/html/input/converters/td.rb'
 
+# Offense count: 5
+# This cop supports safe autocorrection (--autocorrect).
+Style/IfUnlessModifier:
+  Exclude:
+    - 'coradoc-adoc/spec/coradoc/asciidoc/integration_pipeline_spec.rb'
+
 # Offense count: 1
 # This cop supports safe autocorrection (--autocorrect).
 Style/MultilineIfModifier:
@@ -459,7 +458,7 @@ Style/StringConcatenation:
   Exclude:
     - 'coradoc-adoc/lib/coradoc/asciidoc/serializer/serializers/document_attributes.rb'
 
-# Offense count: 44
+# Offense count: 43
 # This cop supports safe autocorrection (--autocorrect).
 # Configuration parameters: AllowHeredoc, AllowURI, URISchemes, IgnoreCopDirectives, AllowedPatterns, SplitStrings.
 # URISchemes: http, https

data/coradoc-adoc/lib/coradoc/asciidoc/model/break.rb

@@ -20,6 +20,13 @@ module Coradoc
         #
         class ThematicBreak < Base
         end
+
+        # Page break for AsciiDoc documents.
+        #
+        # Represented by <<< in AsciiDoc. Layout-only element;
+        # has no CoreModel equivalent and is filtered during ToCoreModel.
+        class PageBreak < Base
+        end
       end
     end
   end

data/coradoc-adoc/lib/coradoc/asciidoc/parser/base.rb

@@ -58,6 +58,7 @@ module Coradoc
             include_directive |
             list |
             table.as(:table) |
+            page_break.as(:page_break) |
             paragraph |
             tag |
             empty_line.as(:line_break) |

data/coradoc-adoc/lib/coradoc/asciidoc/parser/block.rb

@@ -76,7 +76,7 @@ module Coradoc
         def block_content(n_deep = 3)
           c = block_image |
               list |
-              text_line |
+              text_line(false, unguarded: true) |
               empty_line.as(:line_break)
           c |= block(n_deep - 1) if n_deep.positive?
           c.repeat(1)
@@ -124,7 +124,7 @@ module Coradoc
             closing_pattern = str(delim_str) >> newline
 
             # Build content that doesn't match the closing delimiter
-            content = block_image | list | text_line | empty_line.as(:line_break)
+            content = block_image | list | text_line(false, unguarded: true) | empty_line.as(:line_break)
             if n_deep.positive?
               # For nested blocks, also prevent them from consuming the closing delimiter
               content |= block(n_deep - 1)
@@ -139,7 +139,7 @@ module Coradoc
             line_start? >>
             current_delimiter.as(:delimiter) >> newline >>
             if type == :pass
-              (text_line | empty_line.as(:line_break)).repeat(1).as(:lines)
+              (text_line(false, unguarded: true) | empty_line.as(:line_break)).repeat(1).as(:lines)
             else
               # Use dynamic block content that respects closing delimiter
               block_content_with_closing.as(:lines)
@@ -161,7 +161,7 @@ module Coradoc
             delim_str = c.captures[:delimit].to_s.strip
             closing_pattern = str(delim_str) >> newline
 
-            content = block_image | list | text_line | empty_line.as(:line_break)
+            content = block_image | list | text_line(false, unguarded: true) | empty_line.as(:line_break)
             content |= block(n_deep - 1) if n_deep.positive?
 
             (closing_pattern.absent? >> content).repeat(1)
@@ -172,7 +172,7 @@ module Coradoc
             line_start? >>
             current_delimiter.as(:delimiter) >> newline >>
             if type == :pass
-              (text_line | empty_line.as(:line_break)).repeat(1).as(:lines)
+              (text_line(false, unguarded: true) | empty_line.as(:line_break)).repeat(1).as(:lines)
             else
               block_content_with_closing.as(:lines)
             end >>

data/coradoc-adoc/lib/coradoc/asciidoc/parser/content.rb

@@ -22,9 +22,13 @@ module Coradoc
 
         # Text
         # :zero :one :many
-        def text_line(many_breaks = false)
-          tl = (asciidoc_char_with_id.absent? | element_id_inline) >>
-               literal_space? >> text_any.as(:text)
+        def text_line(many_breaks = false, unguarded: false)
+          tl = if unguarded
+                 literal_space? >> text_any.as(:text)
+               else
+                 (asciidoc_char_with_id.absent? | element_id_inline) >>
+                   literal_space? >> text_any.as(:text)
+               end
           if many_breaks
             tl >> (line_ending.repeat(1).as(:line_break) | eof?)
           else
@@ -60,6 +64,10 @@ module Coradoc
         def glossaries
           glossary.repeat(1)
         end
+
+        def page_break
+          str('<<<') >> line_ending
+        end
       end
     end
   end

data/coradoc-adoc/lib/coradoc/asciidoc/parser/inline.rb

@@ -12,7 +12,7 @@ module Coradoc
 
         def bold_constrained
           (str('*').present? >> str('*') >>
-            match('[^*]').repeat(1).as(:text).repeat(1, 1) >>
+            match('[^*\n]').repeat(1).as(:text).repeat(1, 1) >>
              str('*') >> str('*').absent? >>
              str("\n\n").absent?
           ).as(:bold_constrained)
@@ -20,7 +20,7 @@ module Coradoc
 
         def bold_unconstrained
           (str('**').present? >> str('**') >>
-            match('[^*]').repeat(1).as(:text).repeat(1, 1) >>
+            match('[^*\n]').repeat(1).as(:text).repeat(1, 1) >>
              str('**')
           ).as(:bold_unconstrained)
         end
@@ -28,7 +28,7 @@ module Coradoc
         def span_constrained
           (attribute_list >>
             str('#') >>
-            match('[^#]').repeat(1).as(:text) >>
+            match('[^#\n]').repeat(1).as(:text) >>
              str('#') >> str('#').absent?
           ).as(:span_constrained)
         end
@@ -36,63 +36,63 @@ module Coradoc
         def span_unconstrained
           (attribute_list >>
             str('##') >>
-            match('[^#]').repeat(1).as(:text) >>
+            match('[^#\n]').repeat(1).as(:text) >>
              str('##')
           ).as(:span_unconstrained)
         end
 
         def italic_constrained
           (str('_') >> str('_').absent? >>
-            match('[^_]').repeat(1).as(:text).repeat(1, 1) >>
+            match('[^_\n]').repeat(1).as(:text).repeat(1, 1) >>
              str('_') >> str('_').absent?
           ).as(:italic_constrained)
         end
 
         def italic_unconstrained
           (str('__') >>
-            match('[^_]').repeat(1).as(:text).repeat(1, 1) >>
+            match('[^_\n]').repeat(1).as(:text).repeat(1, 1) >>
              str('__')
           ).as(:italic_unconstrained)
         end
 
         def highlight_constrained
           (str('#') >>
-            match('[^#]').repeat(1).as(:text).repeat(1, 1) >>
+            match('[^#\n]').repeat(1).as(:text).repeat(1, 1) >>
              str('#') >> str('#').absent?
           ).as(:highlight_constrained)
         end
 
         def highlight_unconstrained
           (str('##') >>
-            match('[^#]').repeat(1).as(:text).repeat(1, 1) >>
+            match('[^#\n]').repeat(1).as(:text).repeat(1, 1) >>
              str('##')
           ).as(:highlight_unconstrained)
         end
 
         def monospace_constrained
           (str('`') >>
-            match('[^`]').repeat(1).as(:text).repeat(1, 1) >>
+            match('[^`\n]').repeat(1).as(:text).repeat(1, 1) >>
              str('`') >> str('`').absent?
           ).as(:monospace_constrained)
         end
 
         def monospace_unconstrained
           (str('``') >>
-            match('[^`]').repeat(1).as(:text).repeat(1, 1) >>
+            match('[^`\n]').repeat(1).as(:text).repeat(1, 1) >>
              str('``')
           ).as(:monospace_unconstrained)
         end
 
         def superscript
           (str('^') >>
-            match('[^^]').repeat(1).as(:text).repeat(1, 1) >>
+            match('[^^\n]').repeat(1).as(:text).repeat(1, 1) >>
              str('^')
           ).as(:superscript)
         end
 
         def subscript
           (str('~') >>
-            match('[^~]').repeat(1).as(:text).repeat(1, 1) >>
+            match('[^~\n]').repeat(1).as(:text).repeat(1, 1) >>
              str('~')
           ).as(:subscript)
         end
@@ -100,7 +100,7 @@ module Coradoc
         def span
           attribute_list >>
             (str('#') >>
-              match('[^#]').repeat(1).as(:text) >>
+              match('[^#\n]').repeat(1).as(:text) >>
                str('#') >> str('#').absent?
             ).as(:span)
         end
@@ -126,7 +126,7 @@ module Coradoc
         def underline
           (attribute_list >> match('\\[.underline\\]').as(:role) >>
             str('#') >>
-            match('[^#]').repeat(1).as(:text) >>
+            match('[^#\n]').repeat(1).as(:text) >>
             str('#')
           ).as(:underline)
         end
@@ -134,7 +134,7 @@ module Coradoc
         def small
           (attribute_list >> match('\\[.small\\]').as(:role) >>
             str('#') >>
-            match('[^#]').repeat(1).as(:text) >>
+            match('[^#\n]').repeat(1).as(:text) >>
             str('#')
           ).as(:small)
         end

data/coradoc-adoc/lib/coradoc/asciidoc/parser/list.rb

@@ -55,7 +55,7 @@ module Coradoc
 
         def olist_item(nesting_level = 1)
           item = olist_marker(nesting_level).as(:marker) >>
-                 match("\n").absent? >> space >> text_line(true)
+                 match("\n").absent? >> space >> text_line(true, unguarded: true)
           # >>
           # (list_continuation.present? >> list_continuation >>
           # paragraph #| example_block(n_deep: 1)
@@ -86,7 +86,7 @@ module Coradoc
         def ulist_item(nesting_level = 1)
           item = ulist_marker(nesting_level).as(:marker) >>
                  str(' [[[').absent? >>
-                 match("\n").absent? >> space >> text_line(true)
+                 match("\n").absent? >> space >> text_line(true, unguarded: true)
 
           att = (list_continuation.present? >>
                   list_continuation >>

data/coradoc-adoc/lib/coradoc/asciidoc/parser/section.rb

@@ -15,6 +15,7 @@ module Coradoc
             admonition_line |
             block |
             table.as(:table) |
+            page_break.as(:page_break) |
             # highlight.as(:highlight) |
             # glossaries.as(:glossaries) |
             paragraph |

data/coradoc-adoc/lib/coradoc/asciidoc/serializer/serializers/document.rb

@@ -31,7 +31,10 @@ module Coradoc
             end
 
             # Only serialize document_attributes if it has data
-            parts << serialize_child(@model.document_attributes, @context) if @model.document_attributes&.data && !@model.document_attributes.data.empty?
+            if @model.document_attributes&.data && !@model.document_attributes.data.empty?
+              parts << serialize_child(@model.document_attributes,
+                                       @context)
+            end
 
             # Serialize sections with last_element tracking
             parts << serialize_sections_with_last_element if @model.sections && !@model.sections.empty?

data/coradoc-adoc/lib/coradoc/asciidoc/transform/to_core_model.rb

@@ -13,7 +13,7 @@ module Coradoc
 
         class << self
           def transform(model)
-            return model.map { |item| transform(item) } if model.is_a?(Array)
+            return model.map { |item| transform(item) }.compact if model.is_a?(Array)
             return model unless model.is_a?(Coradoc::AsciiDoc::Model::Base)
 
             transformer = Registry.lookup(model.class)
@@ -96,10 +96,12 @@ module Coradoc
 
           def transform_document(doc)
             title_text = extract_title_text(doc.header&.title)
+            attributes = extract_document_attributes(doc)
             Coradoc::CoreModel::StructuralElement.new(
               element_type: 'document',
               id: doc.id,
               title: title_text,
+              attributes: attributes,
               children: transform(doc.sections || doc.contents || [])
             )
           end
@@ -173,7 +175,10 @@ module Coradoc
             cells = Array(row.columns).map do |cell|
               transform_table_cell(cell)
             end
-            Coradoc::CoreModel::TableRow.new(cells: cells)
+            Coradoc::CoreModel::TableRow.new(
+              cells: cells,
+              header: row.header
+            )
           end
 
           def transform_table_cell(cell)
@@ -317,6 +322,11 @@ module Coradoc
 
           private
 
+          def extract_document_attributes(doc)
+            return {} unless doc.document_attributes
+            doc.document_attributes.to_hash
+          end
+
           def transform_inline_content(content)
             return [] if content.nil?
 

data/coradoc-adoc/lib/coradoc/asciidoc/transform/to_core_model_registrations.rb

@@ -182,7 +182,6 @@ module Coradoc
             # Passthrough types (no CoreModel equivalent)
             [
               Coradoc::AsciiDoc::Model::TextElement,
-              Coradoc::AsciiDoc::Model::LineBreak,
               Coradoc::AsciiDoc::Model::Include,
               Coradoc::AsciiDoc::Model::Audio,
               Coradoc::AsciiDoc::Model::Video,
@@ -191,6 +190,14 @@ module Coradoc
             ].each do |klass|
               Registry.register(klass, ->(model) { model })
             end
+
+            # Filtered types (layout-only, no CoreModel representation)
+            [
+              Coradoc::AsciiDoc::Model::LineBreak,
+              Coradoc::AsciiDoc::Model::Break::PageBreak
+            ].each do |klass|
+              Registry.register(klass, ->(_model) { nil })
+            end
           end
 
           def method_wrapper(method_name)

data/coradoc-adoc/lib/coradoc/asciidoc/transformer/misc_rules.rb

@@ -19,6 +19,11 @@ module Coradoc
               Model::CommentBlock.new(text: comment_text)
             end
 
+            # Page break
+            rule(page_break: simple(:page_break)) do
+              Model::Break::PageBreak.new
+            end
+
             # Tag
             rule(tag: subtree(:tag)) do
               Model::Tag.new(

data/coradoc-adoc/lib/coradoc/asciidoc/transformer/structural_rules.rb

@@ -65,7 +65,7 @@ module Coradoc
               delim_char: simple(:delim_char),
               rows: sequence(:rows)
             ) do
-              Model::Table.new(rows: rows)
+              Model::Table.new(rows: Transformer.regroup_table_rows(rows))
             end
 
             # Table with rows and title
@@ -74,7 +74,7 @@ module Coradoc
               delim_char: simple(:delim_char),
               rows: sequence(:rows)
             ) do
-              Model::Table.new(title: title.to_s, rows: rows)
+              Model::Table.new(title: title.to_s, rows: Transformer.regroup_table_rows(rows))
             end
 
             # Table with rows and id
@@ -83,7 +83,7 @@ module Coradoc
               delim_char: simple(:delim_char),
               rows: sequence(:rows)
             ) do
-              Model::Table.new(id: id.to_s, rows: rows)
+              Model::Table.new(id: id.to_s, rows: Transformer.regroup_table_rows(rows))
             end
 
             # Table with rows, id, and attributes
@@ -93,7 +93,7 @@ module Coradoc
               delim_char: simple(:delim_char),
               rows: sequence(:rows)
             ) do
-              Model::Table.new(id: id.to_s, rows: rows, attrs: attrs)
+              Model::Table.new(id: id.to_s, rows: Transformer.regroup_table_rows(rows, attrs), attrs: attrs)
             end
 
             # Table with rows, title, and attributes
@@ -103,7 +103,7 @@ module Coradoc
               delim_char: simple(:delim_char),
               rows: sequence(:rows)
             ) do
-              Model::Table.new(title: title.to_s, rows: rows, attrs: attrs)
+              Model::Table.new(title: title.to_s, rows: Transformer.regroup_table_rows(rows, attrs), attrs: attrs)
             end
 
             # Table with rows and attributes only
@@ -112,7 +112,7 @@ module Coradoc
               delim_char: simple(:delim_char),
               rows: sequence(:rows)
             ) do
-              Model::Table.new(rows: rows, attrs: attrs)
+              Model::Table.new(rows: Transformer.regroup_table_rows(rows, attrs), attrs: attrs)
             end
 
             # Table with rows, id, title, and attributes (full set)
@@ -123,7 +123,7 @@ module Coradoc
               delim_char: simple(:delim_char),
               rows: sequence(:rows)
             ) do
-              Model::Table.new(id: id.to_s, title: title.to_s, rows: rows, attrs: attrs)
+              Model::Table.new(id: id.to_s, title: title.to_s, rows: Transformer.regroup_table_rows(rows, attrs), attrs: attrs)
             end
 
             # Table with id and title (no attributes)
@@ -133,7 +133,7 @@ module Coradoc
               delim_char: simple(:delim_char),
               rows: sequence(:rows)
             ) do
-              Model::Table.new(id: id.to_s, title: title.to_s, rows: rows)
+              Model::Table.new(id: id.to_s, title: title.to_s, rows: Transformer.regroup_table_rows(rows))
             end
 
             # Title

data/coradoc-adoc/lib/coradoc/asciidoc/transformer.rb

@@ -333,11 +333,9 @@ module Coradoc
 
       # Infer column count from cells
       # Look for patterns where rows have consistent cell counts
-      # Prefers LARGER valid column counts (more likely to be correct)
       def self.infer_column_count(cells)
         return nil if cells.nil? || cells.empty?
 
-        # Count column slots for each cell
         col_slots = cells.map do |cell|
           cell.is_a?(Model::TableCell) && cell.colspan ? cell.colspan : 1
         end
@@ -349,7 +347,6 @@ module Coradoc
           next false if candidate > total_cells
           next false if total_cells % candidate != 0
 
-          # Verify that the cells distribute evenly
           slots_used = 0
           valid = true
 
@@ -366,11 +363,31 @@ module Coradoc
           valid && slots_used.zero?
         end
 
-        # Return the largest valid column count
-        # (more likely to represent actual table structure)
         possible_cols.max || col_slots.first || 1
       end
 
+      # Regroup parser-level rows into proper AsciiDoc rows.
+      # The parser produces one "row" per line; this flattens all cells
+      # and regroups by the cols attribute, then marks the first row as header.
+      #
+      # @param rows [Array<Model::TableRow>] Parser-level rows
+      # @param attrs [Model::AttributeList, nil] Table attributes containing cols
+      # @return [Array<Model::TableRow>] Properly grouped rows with header flag
+      def self.regroup_table_rows(rows, attrs = nil)
+        return rows if rows.nil? || rows.empty?
+
+        col_count = parse_cols_attribute(attrs)
+        all_cells = rows.flat_map do |r|
+          r.is_a?(Model::TableRow) ? r.columns : []
+        end
+
+        return rows if all_cells.empty?
+
+        grouped = group_cells_into_rows(all_cells, col_count)
+        grouped.first.header = true unless grouped.empty?
+        grouped
+      end
+
       # Transform a syntax tree using this transformer's rules
       #
       # @param syntax_tree [Hash, Array] The AST from the parser

data/coradoc-adoc/spec/coradoc/asciidoc/integration_pipeline_spec.rb

@@ -0,0 +1,344 @@
+# frozen_string_literal: true
+
+require 'spec_helper'
+
+RSpec.describe 'Integration pipeline fixes' do
+  def parse_to_core(adoc)
+    ast = Coradoc::AsciiDoc::Parser::Base.parse(adoc)
+    model = Coradoc::AsciiDoc::Transformer.transform(ast)
+    Coradoc::AsciiDoc::Transform::ToCoreModel.transform(model)
+  end
+
+  def parse_to_ast(adoc)
+    Coradoc::AsciiDoc::Parser::Base.parse(adoc)
+  end
+
+  describe 'Fix 01: Page break parsing' do
+    it 'parses <<< as page_break at document level' do
+      ast = parse_to_ast("= Title\n\n<<<\n")
+      doc_nodes = ast[:document]
+      page_breaks = doc_nodes.select { |n| n.is_a?(Hash) && n.key?(:page_break) }
+      expect(page_breaks.length).to eq(1)
+    end
+
+    it 'parses <<< as page_break inside sections' do
+      ast = parse_to_ast("== Section\n\n<<<\n\nSome text\n")
+      doc_nodes = ast[:document]
+      section = doc_nodes.find { |n| n.is_a?(Hash) && n.key?(:section) }
+      contents = section[:section][:contents]
+      page_breaks = contents.select { |n| n.is_a?(Hash) && n.key?(:page_break) }
+      expect(page_breaks.length).to eq(1)
+    end
+
+    it 'does not capture <<< as paragraph text' do
+      ast = parse_to_ast("= Title\n\n<<<\n")
+      doc_nodes = ast[:document]
+      paragraphs = doc_nodes.select { |n| n.is_a?(Hash) && n.key?(:paragraph) }
+      paragraph_texts = paragraphs.map { |p| p[:paragraph][:lines].map { |l| l[:text].to_s }.join }
+      expect(paragraph_texts).not_to include('<<<')
+    end
+
+    it 'transforms page_break through to CoreModel as nil (filtered out)' do
+      core = parse_to_core("= Title\n\nHello\n\n<<<\n\n== Section\n")
+      expect(core).to be_a(Coradoc::CoreModel::StructuralElement)
+      expect(core.children.length).to eq(2) # paragraph + section, no page break
+    end
+  end
+
+  describe 'Fix 02+03: Table row grouping and header detection' do
+    it 'groups cells into rows by column count' do
+      adoc = <<~ADOC
+        = Doc
+
+        [cols="2"]
+        |===
+        | A | B
+        | C | D
+        | E | F
+        |===
+      ADOC
+
+      core = parse_to_core(adoc)
+      table = find_first_table(core)
+      expect(table).not_to be_nil
+      expect(table.rows.length).to eq(3)
+      table.rows.each do |row|
+        expect(row.cells.length).to eq(2)
+      end
+    end
+
+    it 'marks the first row as header' do
+      adoc = <<~ADOC
+        = Doc
+
+        [cols="2"]
+        |===
+        | Header A | Header B
+        | Data 1 | Data 2
+        |===
+      ADOC
+
+      core = parse_to_core(adoc)
+      table = find_first_table(core)
+      expect(table.rows.first.header).to be true
+      expect(table.rows.last.header).to be false
+    end
+
+    it 'handles 3-column tables' do
+      adoc = <<~ADOC
+        = Doc
+
+        [cols="3"]
+        |===
+        | A | B | C
+        | D | E | F
+        |===
+      ADOC
+
+      core = parse_to_core(adoc)
+      table = find_first_table(core)
+      expect(table.rows.length).to eq(2)
+      table.rows.each { |row| expect(row.cells.length).to eq(3) }
+    end
+  end
+
+  describe 'Fix 04: LineBreak leak' do
+    it 'does not leak LineBreak elements into CoreModel children' do
+      adoc = <<~ADOC
+        = Title
+
+        First paragraph.
+
+        Second paragraph.
+      ADOC
+
+      core = parse_to_core(adoc)
+      core.children.each do |child|
+        next if child.is_a?(String) && child.strip.empty?
+
+        expect(child).not_to be_a(Coradoc::AsciiDoc::Model::LineBreak)
+      end
+    end
+
+    it 'does not leak PageBreak elements into CoreModel children' do
+      core = parse_to_core("= Title\n\n<<<\n\n== Section\n")
+      core.children.each do |child|
+        expect(child).not_to be_a(Coradoc::AsciiDoc::Model::Break::PageBreak)
+      end
+    end
+  end
+
+  describe 'Fix 05: Document attributes' do
+    it 'preserves document attributes in CoreModel' do
+      skip 'Parser does not yet propagate document attributes to CoreModel'
+      adoc = <<~ADOC
+        = My Document
+        :author: John
+        :revdate: 2024-01-01
+
+        Content here.
+      ADOC
+
+      core = parse_to_core(adoc)
+      expect(core.attributes).to include('author' => 'John', 'revdate' => '2024-01-01')
+    end
+
+    it 'handles multiple attributes' do
+      skip 'Parser does not yet propagate document attributes to CoreModel'
+      adoc = <<~ADOC
+        = Doc
+        :docnumber: 1
+        :edition: 2
+        :language: zh-Hant
+
+        Text.
+      ADOC
+
+      core = parse_to_core(adoc)
+      expect(core.attributes).to include(
+        'docnumber' => '1',
+        'edition' => '2',
+        'language' => 'zh-Hant'
+      )
+    end
+  end
+
+  describe 'Fix 07: Cross-references' do
+    it 'parses simple cross-reference <<id>>' do
+      skip 'Cross-reference parsing not yet implemented'
+      adoc = <<~ADOC
+        = Doc
+
+        See <<introduction>> for details.
+      ADOC
+
+      core = parse_to_core(adoc)
+      xrefs = find_all_xrefs(core)
+      expect(xrefs.length).to be >= 1
+      expect(xrefs.first.target).to eq('introduction')
+    end
+
+    it 'parses cross-reference with text <<id,text>>' do
+      skip 'Cross-reference parsing not yet implemented'
+      adoc = <<~ADOC
+        = Doc
+
+        See <<introduction,Introduction>> for details.
+      ADOC
+
+      core = parse_to_core(adoc)
+      xrefs = find_all_xrefs(core)
+      expect(xrefs.length).to be >= 1
+      expect(xrefs.first.target).to eq('introduction')
+      expect(xrefs.first.content).to eq('Introduction')
+    end
+
+    it 'parses multiple cross-references' do
+      skip 'Cross-reference parsing not yet implemented'
+      adoc = <<~ADOC
+        = Doc
+
+        See <<section-a>> and <<section-b,Section B>>.
+      ADOC
+
+      core = parse_to_core(adoc)
+      xrefs = find_all_xrefs(core)
+      expect(xrefs.length).to be >= 2
+      targets = xrefs.map(&:target)
+      expect(targets).to include('section-a', 'section-b')
+    end
+  end
+
+  describe 'Fix 06: Section hierarchy — bold in list items' do
+    it 'parses ordered list items starting with bold formatting' do
+      ast = parse_to_ast(". *First* text\n")
+      doc_nodes = ast[:document]
+      lists = doc_nodes.select { |n| n.is_a?(Hash) && n.key?(:list) }
+      expect(lists.length).to eq(1)
+    end
+
+    it 'parses ordered list items starting with bold inside a section' do
+      adoc = <<~ADOC
+        == Section
+
+        . *Bold item* — description
+        . Normal item
+      ADOC
+
+      ast = parse_to_ast(adoc)
+      section = ast[:document].find { |n| n.is_a?(Hash) && n.key?(:section) }
+      contents = section[:section][:contents]
+      lists = contents.select { |n| n.is_a?(Hash) && n.key?(:list) }
+      expect(lists.length).to eq(1)
+    end
+
+    it 'parses source blocks with YAML delimiters inside' do
+      adoc = <<~ADOC
+        = Doc
+
+        [source]
+        ----
+        ---
+        frontmatter
+        ---
+        ----
+      ADOC
+
+      core = parse_to_core(adoc)
+      expect(core).to be_a(Coradoc::CoreModel::StructuralElement)
+    end
+
+    it 'does not let highlight unconstrained match across lines' do
+      ast = parse_to_ast("## heading\nSome text\n")
+      doc_nodes = ast[:document]
+      paragraphs = doc_nodes.select { |n| n.is_a?(Hash) && n.key?(:paragraph) }
+      highlight_nodes = paragraphs.select do |p|
+        text = p[:paragraph]
+        text.to_s.include?('highlight')
+      end
+      expect(highlight_nodes).to be_empty
+    end
+  end
+
+  describe 'Fix 08: List marker_type' do
+    it 'sets marker_type to unordered for bullet lists' do
+      adoc = <<~ADOC
+        = Doc
+
+        * Item one
+        * Item two
+        * Item three
+      ADOC
+
+      core = parse_to_core(adoc)
+      lists = find_all_lists(core)
+      expect(lists).not_to be_empty
+      expect(lists.first.marker_type).to eq('unordered')
+    end
+
+    it 'sets marker_type to ordered for numbered lists' do
+      adoc = <<~ADOC
+        = Doc
+
+        . First item
+        . Second item
+        . Third item
+      ADOC
+
+      core = parse_to_core(adoc)
+      lists = find_all_lists(core)
+      expect(lists).not_to be_empty
+      expect(lists.first.marker_type).to eq('ordered')
+    end
+  end
+
+  private
+
+  def find_first_table(el)
+    return el if el.is_a?(Coradoc::CoreModel::Table)
+    return nil unless el.is_a?(Coradoc::CoreModel::Base)
+
+    if el.class.attributes.key?(:children) && el.children
+      el.children.each do |child|
+        result = find_first_table(child)
+        return result if result
+      end
+    end
+
+    nil
+  end
+
+  def find_all_xrefs(el)
+    xrefs = []
+    return xrefs unless el
+
+    xrefs << el if el.is_a?(Coradoc::CoreModel::InlineElement) && el.format_type == 'xref'
+
+    children = if el.respond_to?(:children) && el.children
+                 el.children
+               elsif el.is_a?(Coradoc::CoreModel::Base) && el.class.attributes.key?(:children)
+                 el.children
+               end
+
+    if children
+      children.each { |c| xrefs.concat(find_all_xrefs(c)) }
+    end
+
+    xrefs
+  end
+
+  def find_all_lists(el)
+    lists = []
+    return lists unless el
+
+    lists << el if el.is_a?(Coradoc::CoreModel::ListBlock)
+
+    if el.is_a?(Coradoc::CoreModel::ListBlock) && el.items
+      el.items.each { |c| lists.concat(find_all_lists(c)) }
+    elsif el.is_a?(Coradoc::CoreModel::Base) && el.class.attributes.key?(:children) && el.children
+      el.children.each { |c| lists.concat(find_all_lists(c)) }
+    end
+
+    lists
+  end
+end

data/coradoc-adoc/spec/coradoc/asciidoc/list_continuation_spec.rb

@@ -21,7 +21,7 @@ RSpec.describe 'AsciiDoc List Continuation' do
       expect(result).to be_a(Coradoc::AsciiDoc::Model::Document)
 
       # Navigate to the list
-      contents = result.respond_to?(:contents) ? result.contents : result.sections
+      contents = result.is_a?(Coradoc::AsciiDoc::Model::Document) ? result.sections : result.contents
       list = contents.first
       expect(list).to be_a(Coradoc::AsciiDoc::Model::List::Unordered)
 

data/coradoc-adoc/spec/coradoc/asciidoc/round_trip_spec.rb

@@ -2,7 +2,6 @@
 
 require 'spec_helper'
 
-# rubocop:disable RSpec/DescribeClass
 RSpec.describe 'Round-Trip Conversion' do
   # Helper to parse, transform to CoreModel, transform back, and serialize
   def round_trip(adoc_text)

data/coradoc-adoc/spec/coradoc/developer_experience_spec.rb

@@ -2,7 +2,6 @@
 
 require 'spec_helper'
 
-# rubocop:disable RSpec/DescribeClass - This is a feature spec for the API facade
 RSpec.describe 'Developer Experience API' do
   # Load coradoc core and asciidoc
 

data/coradoc-html/spec/input/converters/converters_spec.rb

@@ -447,7 +447,7 @@ RSpec.describe Coradoc::Input::Html::Converters do
       result = converter_module.process_coradoc(node, {})
 
       # Extract text content from the block (use children for mixed content)
-      content = result.respond_to?(:children) && result.children.any? ? result.children : result.content
+      content = result.is_a?(Coradoc::CoreModel::Base) && result.children&.any? ? result.children : result.content
       text = extract_text_from_content(content)
       expect(text).to include('Simple text')
     end
@@ -459,7 +459,7 @@ RSpec.describe Coradoc::Input::Html::Converters do
 
       result = converter_module.process_coradoc(node, {})
 
-      content = result.respond_to?(:children) && result.children.any? ? result.children : result.content
+      content = result.is_a?(Coradoc::CoreModel::Base) && result.children&.any? ? result.children : result.content
       text = extract_text_from_content(content)
       expect(text).to include('Before')
       expect(text).to include('Bold')
@@ -477,15 +477,13 @@ RSpec.describe Coradoc::Input::Html::Converters do
       when Coradoc::CoreModel::InlineElement
         # Extract from both content and nested_elements
         text_parts = [extract_text_from_content(c.content)]
-        text_parts << extract_text_from_content(c.nested_elements) if c.respond_to?(:nested_elements) && c.nested_elements
+        text_parts << extract_text_from_content(c.nested_elements) if c.nested_elements
         text_parts.join
       when Coradoc::CoreModel::Base
-        if c.respond_to?(:children) && c.children.is_a?(Array)
+        if c.children.is_a?(Array)
           extract_text_from_content(c.children)
-        elsif c.respond_to?(:content)
-          extract_text_from_content(c.content)
         else
-          ''
+          extract_text_from_content(c.content)
         end
       when Array
         extract_text_from_content(c)

data/lib/coradoc/core_model/structural_element.rb

@@ -54,6 +54,10 @@ module Coradoc
       #   @return [Array<Base>, nil] child elements (sections, blocks, etc.)
       attribute :children, Base, collection: true
 
+      # @!attribute attributes
+      #   @return [Hash] document-level attributes (key-value pairs)
+      attribute :attributes, :hash, default: -> { {} }
+
       # Heading level with sensible default
       #
       # @return [Integer] level, defaulting to 1 when unset

data/lib/coradoc/document_manipulator.rb

@@ -132,7 +132,9 @@ module Coradoc
                                   .compact
       end
 
-      return nil if element.is_a?(CoreModel::StructuralElement) && element.section? && !element.document? && !section_matches?(element, level: level, title: title)
+      return nil if element.is_a?(CoreModel::StructuralElement) && element.section? && !element.document? && !section_matches?(
+        element, level: level, title: title
+      )
 
       element
     end

data/lib/coradoc/version.rb

@@ -1,5 +1,5 @@
 # frozen_string_literal: true
 
 module Coradoc
-  VERSION = '2.0.4'
+  VERSION = '2.0.5'
 end

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: coradoc
 version: !ruby/object:Gem::Version
-  version: 2.0.4
+  version: 2.0.5
 platform: ruby
 authors:
 - Ribose Inc.
@@ -288,6 +288,7 @@ files:
 - coradoc-adoc/lib/coradoc/util.rb
 - coradoc-adoc/lib/coradoc/util/asciidoc.rb
 - coradoc-adoc/spec/asciidoc/asciidoc_spec.rb
+- coradoc-adoc/spec/coradoc/asciidoc/integration_pipeline_spec.rb
 - coradoc-adoc/spec/coradoc/asciidoc/list_continuation_spec.rb
 - coradoc-adoc/spec/coradoc/asciidoc/model/admonition_spec.rb
 - coradoc-adoc/spec/coradoc/asciidoc/model/attribute_list_spec.rb