coradoc 2.0.5 → 2.0.6

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: ba4d5d05249449db87eabe01151f28977069d451c5d8b3069d8c7c59238e05df
-  data.tar.gz: 225bfd2be3033634b97443c5949ca0a4c9851c4a2cdfdbfe7059edbdd6d44d1d
+  metadata.gz: 0e11ee6694907b37fdbd6fd3e19fcb999f418a7e651ad75905e4a1624cb4b90f
+  data.tar.gz: 52d69f86c727cbd25531a0849f5ac45df26c22d7adeb7d3cf7af6dcd6ec04cbb
 SHA512:
-  metadata.gz: e5fce433e6b057839f9cb93efc4ac9722dcf812046667082121586a975f273da925d5a87c2cd3c46cd0456dd06f104c36df51a3f9012a043f8ec04674c7fa7ea
-  data.tar.gz: df5ee8e79dad51920ee9a5bc89ddd90f159396d2cd76c8956323b1439129ea6f9f56c2a6b2f5b2cad518c5c93fd0c04a0668006126deb1cda33d7651b6d0137a
+  metadata.gz: d5b89c11f9d99a1a5791e8c2cad44437dbb9af3adad2d5c5600b64d6d0cd222282e577367c2a09035baa69095a43468535cf93092759d788439a65ad33ae3d15
+  data.tar.gz: 26174aca95a3d2553919b31b65718ac0376f6c8e46744772201bf5869796226df2223353808ffdcaf871ff4b698cda6effd4cbab49013fc306d3d181af510bff

data/.rubocop_todo.yml

@@ -1,6 +1,6 @@
 # This configuration was generated by
 # `rubocop --auto-gen-config`
-# on 2026-05-08 15:00:33 UTC using RuboCop version 1.75.8.
+# on 2026-05-09 04:47:12 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
@@ -15,33 +15,56 @@ Gemspec/RequiredRubyVersion:
     - 'coradoc-html/coradoc-html.gemspec'
     - 'coradoc.gemspec'
 
-# Offense count: 1
+# Offense count: 2
 # This cop supports safe autocorrection (--autocorrect).
 # Configuration parameters: EnforcedStyle, IndentationWidth.
 # SupportedStyles: with_first_argument, with_fixed_indentation
 Layout/ArgumentAlignment:
   Exclude:
-    - 'coradoc-adoc/lib/coradoc/asciidoc/serializer/serializers/document.rb'
+    - 'coradoc-adoc/lib/coradoc/asciidoc/transformer/structural_rules.rb'
 
 # Offense count: 1
 # This cop supports safe autocorrection (--autocorrect).
-Layout/ClosingParenthesisIndentation:
+# Configuration parameters: EnforcedStyleAlignWith, Severity.
+# SupportedStylesAlignWith: start_of_line, def
+Layout/DefEndAlignment:
   Exclude:
-    - 'lib/coradoc/document_manipulator.rb'
+    - 'lib/coradoc/core_model/block.rb'
 
-# Offense count: 1
+# Offense count: 2
 # This cop supports safe autocorrection (--autocorrect).
 Layout/EmptyLineAfterGuardClause:
   Exclude:
-    - 'coradoc-adoc/spec/coradoc/asciidoc/integration_pipeline_spec.rb'
+    - 'coradoc-adoc/lib/coradoc/asciidoc/transform/to_core_model.rb'
+    - 'coradoc-html/lib/coradoc/html/renderer.rb'
+
+# Offense count: 4
+# This cop supports safe autocorrection (--autocorrect).
+Layout/EmptyLines:
+  Exclude:
+    - 'coradoc-markdown/lib/coradoc/markdown/model/extension.rb'
+    - 'coradoc-markdown/lib/coradoc/markdown/model/highlight.rb'
+    - 'coradoc-markdown/lib/coradoc/markdown/model/math.rb'
+    - 'coradoc-markdown/lib/coradoc/markdown/model/strikethrough.rb'
+
+# Offense count: 12
+# This cop supports safe autocorrection (--autocorrect).
+# Configuration parameters: AllowMultipleStyles, EnforcedHashRocketStyle, EnforcedColonStyle, EnforcedLastArgumentHashStyle.
+# SupportedHashRocketStyles: key, separator, table
+# SupportedColonStyles: key, separator, table
+# SupportedLastArgumentHashStyles: always_inspect, always_ignore, ignore_implicit, ignore_explicit
+Layout/HashAlignment:
+  Exclude:
+    - 'coradoc-adoc/lib/coradoc/asciidoc/parser/block.rb'
+    - 'coradoc-adoc/lib/coradoc/asciidoc/transformer/structural_rules.rb'
+    - 'spec/core_model/typed_blocks_spec.rb'
 
 # Offense count: 1
 # This cop supports safe autocorrection (--autocorrect).
-# Configuration parameters: EnforcedStyle, IndentationWidth.
-# SupportedStyles: consistent, consistent_relative_to_receiver, special_for_inner_method_call, special_for_inner_method_call_in_parentheses
-Layout/FirstArgumentIndentation:
+# Configuration parameters: Width, AllowedPatterns.
+Layout/IndentationWidth:
   Exclude:
-    - 'lib/coradoc/document_manipulator.rb'
+    - 'lib/coradoc/core_model/block.rb'
 
 # Offense count: 6
 # This cop supports safe autocorrection (--autocorrect).
@@ -50,26 +73,24 @@ Layout/FirstArgumentIndentation:
 Layout/LineLength:
   Max: 235
 
-# Offense count: 1
-# This cop supports safe autocorrection (--autocorrect).
-# Configuration parameters: EnforcedStyle.
-# SupportedStyles: symmetrical, new_line, same_line
-Layout/MultilineMethodCallBraceLayout:
-  Exclude:
-    - 'lib/coradoc/document_manipulator.rb'
-
-# Offense count: 1
+# Offense count: 2
 # This cop supports safe autocorrection (--autocorrect).
 # Configuration parameters: AllowInHeredoc.
 Layout/TrailingWhitespace:
   Exclude:
-    - 'coradoc-adoc/lib/coradoc/asciidoc/serializer/serializers/document.rb'
+    - 'coradoc-adoc/lib/coradoc/asciidoc/parser/block.rb'
+    - 'coradoc-adoc/lib/coradoc/asciidoc/transformer/structural_rules.rb'
 
 # Offense count: 1
 Lint/BinaryOperatorWithIdenticalOperands:
   Exclude:
     - 'coradoc-html/lib/coradoc/html/converters/attribute_reference.rb'
 
+# Offense count: 1
+Lint/DuplicateCaseCondition:
+  Exclude:
+    - 'coradoc-html/lib/coradoc/html/converters/base.rb'
+
 # Offense count: 3
 # This cop supports safe autocorrection (--autocorrect).
 # Configuration parameters: AutoCorrect, AllowComments.
@@ -98,6 +119,13 @@ Lint/MissingSuper:
     - 'coradoc-html/lib/coradoc/html/converters/template_html_converter.rb'
     - 'lib/coradoc/visitor.rb'
 
+# Offense count: 2
+# This cop supports safe autocorrection (--autocorrect).
+Lint/RedundantCopEnableDirective:
+  Exclude:
+    - 'coradoc-adoc/spec/coradoc/asciidoc/round_trip_spec.rb'
+    - 'coradoc-adoc/spec/coradoc/developer_experience_spec.rb'
+
 # Offense count: 29
 # Configuration parameters: AllowKeywordBlockArguments.
 Lint/UnderscorePrefixedVariableName:
@@ -142,7 +170,7 @@ Lint/Void:
 # Offense count: 237
 # Configuration parameters: AllowedMethods, AllowedPatterns, CountRepeatedAttributes.
 Metrics/AbcSize:
-  Max: 178
+  Max: 190
 
 # Offense count: 15
 # Configuration parameters: CountComments, CountAsOne, AllowedMethods, AllowedPatterns, inherit_mode.
@@ -160,15 +188,15 @@ Metrics/BlockNesting:
 Metrics/ClassLength:
   Max: 556
 
-# Offense count: 180
+# Offense count: 183
 # Configuration parameters: AllowedMethods, AllowedPatterns.
 Metrics/CyclomaticComplexity:
-  Max: 32
+  Max: 33
 
-# Offense count: 385
+# Offense count: 388
 # Configuration parameters: CountComments, CountAsOne, AllowedMethods, AllowedPatterns.
 Metrics/MethodLength:
-  Max: 246
+  Max: 245
 
 # Offense count: 21
 # Configuration parameters: CountComments, CountAsOne.
@@ -253,7 +281,7 @@ RSpec/BeforeAfterAll:
     - '**/spec/support/**/*.rb'
     - 'spec/benchmark/performance_spec.rb'
 
-# Offense count: 21
+# Offense count: 24
 # Configuration parameters: IgnoredMetadata.
 RSpec/DescribeClass:
   Enabled: false
@@ -309,7 +337,7 @@ RSpec/MessageSpies:
 RSpec/MultipleDescribes:
   Enabled: false
 
-# Offense count: 628
+# Offense count: 629
 RSpec/MultipleExpectations:
   Max: 12
 
@@ -388,19 +416,23 @@ Style/ClassVars:
     - 'coradoc-adoc/lib/coradoc/asciidoc/serializer/element_registry.rb'
     - 'coradoc-html/lib/coradoc/html/input/converters.rb'
 
-# Offense count: 292
+# Offense count: 305
 # Configuration parameters: AllowedConstants.
 Style/Documentation:
   Enabled: false
 
-# Offense count: 3
+# Offense count: 7
 # This cop supports safe autocorrection (--autocorrect).
 # Configuration parameters: AutoCorrect, EnforcedStyle, AllowComments.
 # SupportedStyles: empty, nil, both
 Style/EmptyElse:
   Exclude:
     - 'coradoc-docx/lib/coradoc/docx/transform/rules/bookmark_rule.rb'
+    - 'coradoc-html/lib/coradoc/html/converters/base.rb'
     - 'coradoc-html/lib/coradoc/html/input/converters/text.rb'
+    - 'coradoc-markdown/lib/coradoc/markdown/transform/from_core_model.rb'
+    - 'lib/coradoc/core_model/block.rb'
+    - 'lib/coradoc/core_model/builder/block_builder.rb'
     - 'lib/coradoc/core_model/toc_generator.rb'
 
 # Offense count: 3
@@ -416,25 +448,25 @@ Style/GlobalVars:
   Exclude:
     - 'utils/parser_analyzer.rb'
 
-# Offense count: 5
+# Offense count: 6
 # Configuration parameters: MinBranchesCount.
 Style/HashLikeCase:
   Exclude:
     - 'coradoc-adoc/lib/coradoc/asciidoc/model/table_cell.rb'
     - 'coradoc-html/lib/coradoc/html/input/converters/table.rb'
     - 'coradoc-html/lib/coradoc/html/input/converters/td.rb'
+    - 'lib/coradoc/core_model/builder/block_builder.rb'
 
-# Offense count: 5
+# Offense count: 1
 # 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:
+# Offense count: 2
+Style/MultilineBlockChain:
   Exclude:
-    - 'coradoc-adoc/lib/coradoc/asciidoc/serializer/serializers/document.rb'
+    - 'coradoc-adoc/lib/coradoc/asciidoc/transform/to_core_model.rb'
 
 # Offense count: 1
 # This cop supports unsafe autocorrection (--autocorrect-all).
@@ -451,6 +483,20 @@ Style/OptionalBooleanParameter:
   Exclude:
     - 'coradoc-adoc/lib/coradoc/asciidoc/parser/content.rb'
 
+# Offense count: 1
+# This cop supports safe autocorrection (--autocorrect).
+Style/RedundantParentheses:
+  Exclude:
+    - 'lib/coradoc/core_model/block.rb'
+
+# Offense count: 1
+# This cop supports unsafe autocorrection (--autocorrect-all).
+# Configuration parameters: ConvertCodeThatCanStartToReturnNil, AllowedMethods, MaxChainLength.
+# AllowedMethods: present?, blank?, presence, try, try!
+Style/SafeNavigation:
+  Exclude:
+    - 'coradoc-adoc/spec/coradoc/asciidoc/integration_pipeline_spec.rb'
+
 # Offense count: 1
 # This cop supports unsafe autocorrection (--autocorrect-all).
 # Configuration parameters: Mode.

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

@@ -1,10 +1,10 @@
 # frozen_string_literal: true
 
-require_relative 'resolver'
-
 module Coradoc
   module AsciiDoc
     module Model
+      autoload :Resolver, "#{__dir__}/resolver"
+
       # Document model representing an AsciiDoc document.
       #
       # The Document class is the main container for parsed AsciiDoc content.

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

@@ -1,10 +1,10 @@
 # frozen_string_literal: true
 
-require_relative 'resolvable'
-
 module Coradoc
   module AsciiDoc
     module Model
+      autoload :Resolvable, "#{__dir__}/resolvable"
+
       # Include directive element for AsciiDoc documents.
       #
       # Include directives allow incorporating content from external files

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

@@ -1,14 +1,16 @@
 # frozen_string_literal: true
 
-# IMPORTANT: Load Base class and Serialization module FIRST
-# These register the asciidoc format with Lutaml::Model, which is needed
-# before any model class can use the asciidoc DSL
-require_relative 'model/base'
+# IMPORTANT: Load Serialization module eagerly — it registers the asciidoc
+# format with Lutaml::Model, which is needed before any model class can
+# use the asciidoc DSL.
 require_relative 'model/serialization'
 
 module Coradoc
   module AsciiDoc
     module Model
+      # Base class for all AsciiDoc model classes (no side effects)
+      autoload :Base, "#{__dir__}/model/base"
+
       # Layer 0: Mixins (no dependencies)
       autoload :Anchorable, "#{__dir__}/model/anchorable"
       autoload :Attached, "#{__dir__}/model/attached"

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

@@ -55,7 +55,7 @@ module Coradoc
         end
 
         def source_block(n_deep)
-          block_style(n_deep, '-', 4)
+          block_style(n_deep, '-', 4, verbatim: true)
         end
 
         # Open block: exactly 2 dashes (cannot nest within itself)
@@ -110,7 +110,7 @@ module Coradoc
         # @param delimiter [String] The delimiter character ("=", "-", "_", "*", "+")
         # @param repeater [Integer] Minimum number of delimiter characters (default: 4)
         # @param type [Symbol] Block type for special handling (e.g., :pass)
-        def block_style(n_deep = 3, delimiter = '*', repeater = 4, type = nil)
+        def block_style(n_deep = 3, delimiter = '*', repeater = 4, type = nil, verbatim: false)
           # repeat(repeater,) means repeater or more characters
           current_delimiter = str(delimiter).repeat(repeater).capture(:delimit)
           closing_delimiter = dynamic do |_s, c|
@@ -124,7 +124,8 @@ module Coradoc
             closing_pattern = str(delim_str) >> newline
 
             # Build content that doesn't match the closing delimiter
-            content = block_image | list | text_line(false, unguarded: true) | empty_line.as(:line_break)
+            content = block_image | list | text_line(false, unguarded: true,
+                                                            verbatim: verbatim) | 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 +140,7 @@ module Coradoc
             line_start? >>
             current_delimiter.as(:delimiter) >> newline >>
             if type == :pass
-              (text_line(false, unguarded: true) | empty_line.as(:line_break)).repeat(1).as(:lines)
+              (text_line(false, unguarded: true, verbatim: verbatim) | empty_line.as(:line_break)).repeat(1).as(:lines)
             else
               # Use dynamic block content that respects closing delimiter
               block_content_with_closing.as(:lines)

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

@@ -22,8 +22,10 @@ module Coradoc
 
         # Text
         # :zero :one :many
-        def text_line(many_breaks = false, unguarded: false)
-          tl = if unguarded
+        def text_line(many_breaks = false, unguarded: false, verbatim: false)
+          tl = if verbatim
+                 text_any.as(:text)
+               elsif unguarded
                  literal_space? >> text_any.as(:text)
                else
                  (asciidoc_char_with_id.absent? | element_id_inline) >>

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

@@ -103,14 +103,15 @@ module Coradoc
           def transform_block(block)
             content = block.renderable_content
 
-            if block.element_type == 'paragraph'
+            semantic = resolve_semantic_type(block)
+
+            case semantic
+            when :paragraph
               return Coradoc::AsciiDoc::Model::Paragraph.new(
                 id: block.id,
                 content: create_text_elements(content)
               )
-            end
-
-            if block.element_type == 'comment'
+            when :comment
               return Coradoc::AsciiDoc::Model::CommentBlock.new(
                 text: safe_content_to_string(content)
               )
@@ -118,42 +119,67 @@ module Coradoc
 
             content_text = safe_content_to_string(content)
 
-            case block.delimiter_type
-            when 'source'
+            case semantic
+            when :source_code
               Coradoc::AsciiDoc::Model::Block::SourceCode.new(
                 id: block.id,
                 title: block.title,
                 lines: content_text.split("\n"),
                 attributes: build_attributes(block)
               )
-            when 'quote'
+            when :quote
               Coradoc::AsciiDoc::Model::Block::Quote.new(
                 id: block.id,
                 title: block.title,
                 lines: content_text.split("\n")
               )
-            when 'example'
+            when :example
               Coradoc::AsciiDoc::Model::Block::Example.new(
                 id: block.id,
                 title: block.title,
                 lines: content_text.split("\n")
               )
-            when 'sidebar'
+            when :sidebar
               Coradoc::AsciiDoc::Model::Block::Side.new(
                 id: block.id,
                 title: block.title,
                 lines: content_text.split("\n")
               )
-            when 'literal'
+            when :literal
               Coradoc::AsciiDoc::Model::Block::Literal.new(
                 id: block.id,
                 title: block.title,
                 lines: content_text.split("\n")
               )
-            when 'paragraph'
-              Coradoc::AsciiDoc::Model::Paragraph.new(
+            when :open
+              Coradoc::AsciiDoc::Model::Block::Open.new(
                 id: block.id,
-                content: create_text_elements(content)
+                title: block.title,
+                lines: content_text.split("\n")
+              )
+            when :pass
+              Coradoc::AsciiDoc::Model::Block::Pass.new(
+                id: block.id,
+                title: block.title,
+                lines: content_text.split("\n")
+              )
+            when :listing
+              Coradoc::AsciiDoc::Model::Block::Listing.new(
+                id: block.id,
+                title: block.title,
+                lines: content_text.split("\n")
+              )
+            when :verse
+              Coradoc::AsciiDoc::Model::Block::Quote.new(
+                id: block.id,
+                title: block.title,
+                lines: content_text.split("\n")
+              )
+            when :reviewer
+              Coradoc::AsciiDoc::Model::Block::ReviewerComment.new(
+                id: block.id,
+                title: block.title,
+                lines: content_text.split("\n")
               )
             else
               delim = block.delimiter_type.to_s
@@ -350,6 +376,51 @@ module Coradoc
 
           private
 
+          # Reverse mapping: semantic type → AsciiDoc delimiter generation
+          SEMANTIC_TO_ADOC_BLOCK = {
+            source_code: :source_code,
+            quote: :quote,
+            example: :example,
+            sidebar: :sidebar,
+            literal: :literal,
+            pass: :pass,
+            open: :open,
+            verse: :open,
+            paragraph: :paragraph,
+            comment: :comment
+          }.freeze
+
+          # Map raw delimiter_type string → semantic type (backward compat)
+          DELIMITER_CHAR_TO_SEMANTIC = {
+            '-' => :source_code,
+            '=' => :example,
+            '_' => :quote,
+            '*' => :sidebar,
+            '.' => :literal,
+            '+' => :pass
+          }.freeze
+
+          def resolve_semantic_type(block)
+            # Polymorphic dispatch: typed classes override semantic_type
+            semantic = block.resolve_semantic_type
+            return semantic if semantic
+
+            # Format-specific fallback from delimiter_type
+            delim = block.delimiter_type
+            return nil unless delim && !delim.empty?
+
+            case delim
+            when 'comment' then :comment
+            when 'paragraph' then :paragraph
+            when '[verse]' then :verse
+            when '>' then :quote
+            when "'''", '---', '___', '***' then :horizontal_rule
+            else
+              char = delim[0]
+              DELIMITER_CHAR_TO_SEMANTIC[char] || nil
+            end
+          end
+
           def safe_content_to_string(content)
             case content
             when String

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

@@ -31,21 +31,23 @@ module Coradoc
             when Coradoc::AsciiDoc::Model::Paragraph
               transform_paragraph(model)
             when Coradoc::AsciiDoc::Model::Block::SourceCode
-              transform_block(model, 'source')
+              transform_source_block(model)
             when Coradoc::AsciiDoc::Model::Block::Quote
-              transform_block(model, 'quote')
+              transform_typed_block(model, Coradoc::CoreModel::QuoteBlock)
             when Coradoc::AsciiDoc::Model::Block::Example
-              transform_block(model, 'example')
+              transform_typed_block(model, Coradoc::CoreModel::ExampleBlock)
             when Coradoc::AsciiDoc::Model::Block::Side
-              transform_block(model, 'sidebar')
+              transform_typed_block(model, Coradoc::CoreModel::SidebarBlock)
             when Coradoc::AsciiDoc::Model::Block::Literal
-              transform_block(model, 'literal')
+              transform_typed_block(model, Coradoc::CoreModel::LiteralBlock)
             when Coradoc::AsciiDoc::Model::Block::Open
-              transform_block(model, 'open')
+              transform_typed_block(model, Coradoc::CoreModel::OpenBlock)
             when Coradoc::AsciiDoc::Model::Block::Pass
-              transform_block(model, 'pass')
+              transform_typed_block(model, Coradoc::CoreModel::PassBlock)
+            when Coradoc::AsciiDoc::Model::Block::Listing
+              transform_typed_block(model, Coradoc::CoreModel::ListingBlock)
             when Coradoc::AsciiDoc::Model::Block::Core
-              transform_block(model, model.delimiter)
+              transform_block(model, model.delimiter.to_s)
             when Coradoc::AsciiDoc::Model::Table
               transform_table(model)
             when Coradoc::AsciiDoc::Model::TableRow
@@ -125,33 +127,25 @@ module Coradoc
 
             Coradoc::CoreModel::Block.new(
               element_type: 'paragraph',
+              block_semantic_type: :paragraph,
               id: para.id,
               content: extract_text_content(para.content),
               children: children
             )
           end
 
-          def transform_block(block, delimiter_type)
-            content_lines = Array(block.lines).map do |line|
-              case line
-              when Coradoc::AsciiDoc::Model::Base
-                transformed = transform(line)
-                if transformed.is_a?(Coradoc::CoreModel::Base)
-                  extract_core_model_text(transformed)
-                else
-                  transformed.to_s
-                end
-              else
-                line.to_s
-              end
+          def transform_source_block(block)
+            content_lines = Array(block.lines).reject do |line|
+              line.is_a?(Coradoc::AsciiDoc::Model::LineBreak) ||
+                line.is_a?(Coradoc::AsciiDoc::Model::Break::PageBreak)
+            end.map do |line|
+              extract_text_content(line)
             end.join("\n")
 
-            language = block.lang || block.attributes&.[]('language') ||
-                       block.attributes&.positional&.first
+            language = extract_block_language(block)
 
-            Coradoc::CoreModel::Block.new(
+            Coradoc::CoreModel::SourceBlock.new(
               element_type: 'block',
-              delimiter_type: delimiter_type,
               id: block.id,
               title: extract_title_text(block.title),
               content: content_lines,
@@ -159,6 +153,47 @@ module Coradoc
             )
           end
 
+          def transform_block(block, semantic_type_or_delimiter)
+            content_lines = extract_block_lines(block)
+            semantic_type = if semantic_type_or_delimiter.is_a?(Symbol)
+                              semantic_type_or_delimiter
+                            else
+                              asciidoc_delimiter_to_semantic(semantic_type_or_delimiter)
+                            end
+
+            Coradoc::CoreModel::Block.new(
+              element_type: 'block',
+              block_semantic_type: semantic_type,
+              delimiter_type: semantic_type_or_delimiter.is_a?(String) ? semantic_type_or_delimiter : nil,
+              id: block.id,
+              title: extract_title_text(block.title),
+              content: content_lines,
+              language: extract_block_language(block)
+            )
+          end
+
+          def transform_typed_block(block, klass, extra_attrs = {})
+            content_lines = extract_block_lines(block)
+
+            klass.new(
+              element_type: 'block',
+              id: block.id,
+              title: extract_title_text(block.title),
+              content: content_lines,
+              language: extract_block_language(block),
+              **extra_attrs
+            )
+          end
+
+          def extract_block_lines(block)
+            Array(block.lines).reject do |line|
+              line.is_a?(Coradoc::AsciiDoc::Model::LineBreak) ||
+                line.is_a?(Coradoc::AsciiDoc::Model::Break::PageBreak)
+            end.map do |line|
+              extract_text_content(line)
+            end.join("\n")
+          end
+
           def transform_table(table)
             rows = Array(table.rows).map do |row|
               transform_table_row(row)
@@ -322,11 +357,45 @@ module Coradoc
 
           private
 
+          # AsciiDoc delimiters are any length of the same character (4+ for most, 2+ for open).
+          # We map by the first character to handle all lengths correctly.
+          ADOC_DELIMITER_CHAR_TO_SEMANTIC = {
+            '-' => :source_code,
+            '=' => :example,
+            '_' => :quote,
+            '*' => :sidebar,
+            '.' => :literal,
+            '+' => :pass
+          }.freeze
+
+          def asciidoc_delimiter_to_semantic(delimiter)
+            return :open if delimiter && delimiter.length < 4
+
+            char = delimiter&.[](0)
+            ADOC_DELIMITER_CHAR_TO_SEMANTIC[char] || :open
+          end
+
           def extract_document_attributes(doc)
             return {} unless doc.document_attributes
+
             doc.document_attributes.to_hash
           end
 
+          def extract_block_language(block)
+            lang = block.lang
+            return lang if lang.is_a?(String) && !lang.empty?
+
+            attrs = block.attributes
+            return nil unless attrs.is_a?(Coradoc::AsciiDoc::Model::AttributeList)
+
+            named_lang = attrs['language']
+            return named_lang.to_s if named_lang
+
+            # For [source,yaml], the language is the second positional attribute
+            positional = attrs.positional
+            positional[1]&.value&.to_s if positional.length > 1
+          end
+
           def transform_inline_content(content)
             return [] if content.nil?