coradoc-adoc 2.0.17 → 2.0.19

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 99b9b654a8eebde5418dfbee6f7b184e65bbbcd830b24f0211963d61ed6c1201
-  data.tar.gz: 93e69befaf29a65358e5b6655b7b6cf63d0fb9baab9a828848d7468290c87e3b
+  metadata.gz: 817eba65c55a5dea765ab8e80ee5542a1353741659c8a32ef4eb80e39f052dcb
+  data.tar.gz: 25a07bc63696b7ba27933c9724c0a5ada466dc7cd0f37350ae4c9a4a475afbec
 SHA512:
-  metadata.gz: 1480db996c2768544ae15657c7f394bcaed954c31c79bdf1bfd3f3493b608aad77dc078ed44058ac0964632626585f096a7c96d6047a256ed07f902291ac204a
-  data.tar.gz: b3d0d2044b40702ad13d2f500e06e4e5881e865eda9ac0bc3f01fba94c82175b9b1752e3076b05781d7fd03075af7d276b8af55ef18b4be08576ea5018ea07b0
+  metadata.gz: c3d26bcc8f91f799f455f051109933875545219b14e9556cc4d5774dcda16e9abdcb1f0e6a090f1d9aea35c49d4d7f5ed8c6bae2b1e47a76cc75a86270e47cd7
+  data.tar.gz: 50a575b3666d1d6df0a8813a9761345ced5d895b2c73bf80347d8fb24e36c396389d4901fa22a6f34e1adcea6cb4d5b697bdab121dc495603da88740753e821d

data/lib/coradoc/asciidoc/parser/attribute_list.rb

@@ -44,7 +44,7 @@ module Coradoc
         end
 
         def positional_value_unquoted
-          match('[^\]\s,]').repeat(1) >> space.absent?
+          match('[^\],]').repeat(1)
         end
 
         def positional_value_single_quote

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

@@ -22,7 +22,8 @@ module Coradoc
         # - "|" separates cells within the table
 
         def block(n_deep = 3)
-          (example_block(n_deep) |
+          (markdown_code_block(n_deep) |
+          example_block(n_deep) |
           sidebar_block(n_deep) |
           source_block(n_deep) |
           quote_block(n_deep) |
@@ -60,6 +61,38 @@ module Coradoc
           block_style_exact(n_deep, '-', 2)
         end
 
+        # Markdown-style fenced code block: triple-backtick (or longer)
+        # fence with an optional language tag on the opening line. Behaves
+        # as a verbatim source block — same model as `[source,lang]\n----`.
+        # Pragmatic permissiveness for content that originates from (or is
+        # edited alongside) Markdown; not standard AsciiDoc but widely
+        # accepted (GitHub's renderer treats ``` as a listing delimiter).
+        def markdown_code_block(n_deep = 3)
+          capture_key = :"md_fence_#{n_deep}"
+          opening_fence = str('`').repeat(3).capture(capture_key)
+          closing_fence = dynamic do |_s, c|
+            str(c.captures[capture_key].to_s.strip)
+          end
+          language = (space? >> match("[A-Za-z0-9_+.-]").repeat(1).as(:language)).maybe
+
+          block_content_with_closing = dynamic do |_s, c|
+            fence_str = c.captures[capture_key].to_s.strip
+            closing_pattern = closing_fence >> space? >> newline
+
+            content = text_line(false, unguarded: true, verbatim: true) |
+                        empty_line.as(:line_break)
+
+            (closing_pattern.absent? >> content).repeat(1)
+          end
+
+          block_header >>
+            line_start? >>
+            opening_fence.as(:delimiter) >> language >> newline >>
+            block_content_with_closing.as(:lines) >>
+            line_start? >>
+            closing_fence >> space? >> newline
+        end
+
         def block_title
           (line_start? >> block_delimiter.absent?) >>
             str('.') >> space.absent? >> text.as(:title) >> newline
@@ -80,8 +113,9 @@ module Coradoc
           c.repeat(1)
         end
 
-        # Block delimiter: 4+ identical characters (or 2 for open block)
-        # Used by paragraph.rb to reject lines that look like block delimiters.
+        # Block delimiter: 4+ identical characters, or 2 dashes for open
+        # blocks, or 3+ backticks for Markdown-style code fences. Used by
+        # paragraph.rb to reject lines that look like block delimiters.
         # NOTE: repeat(4,) means 4 or more (not exactly 4)
         def block_delimiter
           line_start? >>
@@ -91,7 +125,8 @@ module Coradoc
               str('+') |
               str('.') |
               str('-')).repeat(4) | # 4+ characters for most blocks
-              str('-').repeat(2, 2)) >> # Exactly 2 for open block
+              str('-').repeat(2, 2) | # Exactly 2 for open block
+              str('`').repeat(3)) >> # 3+ for Markdown code fences
             newline
         end
 

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

@@ -118,6 +118,7 @@ module Coradoc
 
         def inline_image
           (str('image:').present? >> str('image:') >>
+            str(':').absent? >>
             match('[A-Za-z0-9_.\\-:/&?=+,%#~;]+').repeat(1).as(:path) >>
             (str('[') >> match('[^\\]]').repeat(1).as(:text) >> str(']')).maybe
           ).as(:inline_image)

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

@@ -72,11 +72,35 @@ module Coradoc
         def ulist_marker(nesting_level = 1)
           line_start? >>
             (nesting_level > 1 ? literal_space.maybe : str('')) >>
-            str('*' * nesting_level) >>
+            (
+              asterisk_marker(nesting_level) |
+              dash_marker(nesting_level)
+            )
+        end
+
+        # AsciiDoc standard bullet: `*`, `**`, `***`, ... matching the
+        # nesting level. Excludes table delimiters (`|===`) and deeper
+        # asterisk runs that belong to a sibling level.
+        def asterisk_marker(nesting_level)
+          str('*' * nesting_level) >>
             str('*').absent? >>
             str('===').absent?
         end
 
+        # Markdown-style dash bullet: `-`. Accepted only at the top level
+        # because Markdown nests via indentation rather than multi-char
+        # markers — deeper levels stay on the AsciiDoc `*` form. Guards
+        # exclude em-dashes (`--`), delimited-block fences (`----`),
+        # and negative-number runs (`-1`, `-42`) which are not list
+        # markers in any common dialect.
+        def dash_marker(nesting_level)
+          return match('').absent? unless nesting_level == 1
+
+          str('-') >>
+            str('-').absent? >>
+            match('[0-9]').absent?
+        end
+
         def ulist_item(nesting_level = 1)
           item = ulist_marker(nesting_level).as(:marker) >>
                  str(' [[[').absent? >>

data/lib/coradoc/asciidoc/transform/element_transformers/block_transformer.rb

@@ -128,7 +128,7 @@ module Coradoc
 
             def transform_typed_block(block, klass, extra_attrs = {})
               raw_lines = Array(block.lines)
-              has_nested_blocks = raw_lines.any?(Coradoc::AsciiDoc::Model::Block::Core)
+              has_nested_blocks = raw_lines.any? { |line| block_level_child?(line) }
 
               if has_nested_blocks
                 children = raw_lines.reject do |line|
@@ -195,6 +195,17 @@ module Coradoc
               groups << current if current.any?
               groups
             end
+
+            # A line that should be emitted as a direct child of the
+            # enclosing block rather than joined into a paragraph sibling.
+            # The AsciiDoc model layer declares level via `block_level?`
+            # (true for Block::Core delimited blocks, BlockImage, Table,
+            # List::Core, Section, CommentBlock, Attached). Inline content
+            # (String, TextElement, LineBreak, PageBreak) returns false and
+            # stays inside paragraphs.
+            def block_level_child?(line)
+              line.is_a?(Coradoc::AsciiDoc::Model::Base) && line.block_level?
+            end
           end
         end
       end

data/lib/coradoc/asciidoc/transform/element_transformers/include_transformer.rb

@@ -0,0 +1,56 @@
+# frozen_string_literal: true
+
+module Coradoc
+  module AsciiDoc
+    module Transform
+      module ElementTransformers
+        # Transforms AsciiDoc {Model::Include} into the canonical
+        # {CoreModel::Include}. The CoreModel node carries the parsed
+        # options as a typed {CoreModel::IncludeOptions} instance plus
+        # the raw bracket body for verbatim round-trip.
+        class IncludeTransformer
+          class << self
+            # @param model [Coradoc::AsciiDoc::Model::Include]
+            # @return [Coradoc::CoreModel::Include]
+            def transform_include(model)
+              options_hash = extract_options_hash(model.attributes)
+              options = CoreModel::IncludeOptions.from_hash(options_hash)
+              raw = extract_raw_options(model.attributes)
+
+              CoreModel::Include.new(
+                target: model.path.to_s,
+                options: options,
+                raw_options: raw,
+                line_break: model.line_break.to_s
+              )
+            end
+
+            private
+
+            def extract_options_hash(attrs)
+              return {} unless attrs.is_a?(Coradoc::AsciiDoc::Model::AttributeList)
+
+              attrs.named.each_with_object({}) do |attr, hash|
+                hash[attr.name.to_s] = serialize_named_value(attr.value)
+              end
+            end
+
+            def serialize_named_value(value)
+              case value
+              when Array then value.map(&:to_s).join(';')
+              else value.to_s
+              end
+            end
+
+            def extract_raw_options(attrs)
+              return '' unless attrs.is_a?(Coradoc::AsciiDoc::Model::AttributeList)
+
+              adoc = attrs.to_adoc(show_empty: false).to_s
+              adoc.gsub(/\A\[|\]\z/, '')
+            end
+          end
+        end
+      end
+    end
+  end
+end

data/lib/coradoc/asciidoc/transform/element_transformers/other_transformer.rb

@@ -27,9 +27,19 @@ module Coradoc
             def transform_image(image)
               src = image.src.to_s
               src = src[1..] if src.start_with?(':')
+              positional = image.attributes&.positional || []
+              positional_alt = positional.first&.value&.to_s || ''
+              caption = positional[1]&.value&.to_s
+              title = image.title&.to_s
+              # AsciiDoc block-title (`.Caption`) is semantically a caption,
+              # but for compatibility with the existing pipeline that treated
+              # it as alt when no positional alt was supplied, fall back to it.
+              alt = positional_alt.empty? ? title : positional_alt
               Coradoc::CoreModel::Image.new(
                 src: src,
-                alt: image.title&.to_s,
+                alt: alt,
+                caption: caption,
+                title: title,
                 width: image.attributes&.[]('width'),
                 height: image.attributes&.[]('height')
               )

data/lib/coradoc/asciidoc/transform/element_transformers.rb

@@ -10,6 +10,7 @@ module Coradoc
         autoload :InlineTransformer, "#{__dir__}/element_transformers/inline_transformer"
         autoload :TableTransformer, "#{__dir__}/element_transformers/table_transformer"
         autoload :OtherTransformer, "#{__dir__}/element_transformers/other_transformer"
+        autoload :IncludeTransformer, "#{__dir__}/element_transformers/include_transformer"
       end
     end
   end

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

@@ -388,8 +388,51 @@ module Coradoc
             )
           end
 
+          def transform_include(include)
+            Coradoc::AsciiDoc::Model::Include.new(
+              path: include.target.to_s,
+              attributes: build_include_attributes(include),
+              line_break: include.line_break.to_s
+            )
+          end
+
           private
 
+          def build_include_attributes(include)
+            list = Coradoc::AsciiDoc::Model::AttributeList.new
+            options = include.options
+            return list if options.nil?
+
+            add_tag_attribute(list, options)
+            add_simple_attribute(list, 'lines', options.lines_spec)
+            add_leveloffset_attribute(list, options)
+            add_simple_attribute(list, 'indent', options.indent&.to_s)
+            add_simple_attribute(list, 'encoding', options.file_encoding)
+            list
+          end
+
+          def add_simple_attribute(list, name, value)
+            return if value.nil? || value.to_s.empty?
+
+            list.add_named(name, value.to_s)
+          end
+
+          def add_tag_attribute(list, options)
+            if options.tags_wildcard
+              list.add_named('tags', '*')
+            elsif options.tags_inverted
+              list.add_named('tags', '**')
+            elsif options.tags.any?
+              list.add_named('tags', options.tags.join(';'))
+            end
+          end
+
+          def add_leveloffset_attribute(list, options)
+            return if options.leveloffset.nil?
+
+            list.add_named('leveloffset', options.leveloffset.to_s)
+          end
+
           # If the first CoreModel child is a FrontmatterBlock, serialize
           # it to YAML text via Codec (single source of truth) and pop it
           # from the children list. Returns [remaining_children,

data/lib/coradoc/asciidoc/transform/from_core_model_registrations.rb

@@ -120,6 +120,11 @@ module Coradoc
               Coradoc::CoreModel::CommentLine,
               ->(model) { FromCoreModel.transform_comment_line(model) }
             )
+
+            Registry.register(
+              Coradoc::CoreModel::Include,
+              ->(model) { FromCoreModel.transform_include(model) }
+            )
           end
         end
       end

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

@@ -10,6 +10,7 @@ module Coradoc
         Inl = ElementTransformers::InlineTransformer
         Tbl = ElementTransformers::TableTransformer
         Oth = ElementTransformers::OtherTransformer
+        Inc = ElementTransformers::IncludeTransformer
 
         class << self
           def register_all!
@@ -231,8 +232,16 @@ module Coradoc
               ->(model) { Oth.transform_bibliography_entry(model) }
             )
 
-            [
+            # Include directives become first-class CoreModel::Include nodes
+            # (link edges in a text graph). To splice their content inline,
+            # call +Coradoc.resolve_includes(doc, base_dir:)+ on the parsed
+            # document — that step is intentionally separate from parse.
+            Registry.register(
               Coradoc::AsciiDoc::Model::Include,
+              ->(model) { Inc.transform_include(model) }
+            )
+
+            [
               Coradoc::AsciiDoc::Model::Audio,
               Coradoc::AsciiDoc::Model::Video,
               Coradoc::AsciiDoc::Model::ContentList,

data/lib/coradoc/asciidoc/transformer/block_rules.rb

@@ -25,6 +25,10 @@ module Coradoc
                 lines: lines,
                 ordering: ordering
               }
+              # Markdown fences carry the language tag inline (```ruby);
+              # pass it through so the SourceCode classifier entry can set
+              # block.lang directly, which extract_block_language prefers.
+              opts[:lang] = block[:language].to_s if block.key?(:language) && !block[:language].nil?
               BlockTypeClassifier.classify(delimiter, opts, attribute_list)
             end
 
@@ -46,7 +50,7 @@ module Coradoc
               id = block_image[:id]
               title = block_image[:title]
               path = block_image[:path]
-              attrs = AttributeListNormalizer.coerce(block_image[:attribute_list])
+              attrs = AttributeListNormalizer.coerce(block_image[:attribute_list_macro])
               Model::Image::BlockImage.new(
                 title: title,
                 id: id,

data/lib/coradoc/asciidoc/transformer/block_type_classifier.rb

@@ -28,7 +28,15 @@ module Coradoc
           ['.', 4, nil, ->(opts, attrs) { Model::Block::Literal.new(**opts.merge(attributes: attrs)) }],
           ['_', 4, nil, ->(opts, attrs) { Model::Block::Quote.new(**opts.merge(attributes: attrs)) }],
           ['-', 4, nil, ->(opts, attrs) { Model::Block::SourceCode.new(**opts.merge(attributes: attrs)) }],
-          ['-', 2, 2,  ->(opts, attrs) { Model::Block::Open.new(**opts.merge(attributes: attrs)) }]
+          ['-', 2, 2,  ->(opts, attrs) { Model::Block::Open.new(**opts.merge(attributes: attrs)) }],
+          # Markdown-style triple-backtick fence: behaves as a SourceCode
+          # block. The language tag parsed from the opening fence is passed
+          # through opts[:lang]; extract_block_language prefers block.lang.
+          ['`', 3, nil, ->(opts, attrs) {
+            model_opts = opts.merge(attributes: attrs, delimiter_char: '`')
+            model_opts[:lang] = opts[:lang] if opts.key?(:lang)
+            Model::Block::SourceCode.new(**model_opts)
+          }]
         ].freeze
 
         module_function

data/lib/coradoc/asciidoc/version.rb

@@ -2,6 +2,6 @@
 
 module Coradoc
   module AsciiDoc
-    VERSION = '2.0.17'
+    VERSION = '2.0.19'
   end
 end

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: coradoc-adoc
 version: !ruby/object:Gem::Version
-  version: 2.0.17
+  version: 2.0.19
 platform: ruby
 authors:
 - Ribose Inc.
@@ -319,6 +319,7 @@ files:
 - lib/coradoc/asciidoc/transform/element_transformers.rb
 - lib/coradoc/asciidoc/transform/element_transformers/block_transformer.rb
 - lib/coradoc/asciidoc/transform/element_transformers/document_transformer.rb
+- lib/coradoc/asciidoc/transform/element_transformers/include_transformer.rb
 - lib/coradoc/asciidoc/transform/element_transformers/inline_transformer.rb
 - lib/coradoc/asciidoc/transform/element_transformers/list_transformer.rb
 - lib/coradoc/asciidoc/transform/element_transformers/other_transformer.rb