uniword 1.0.10 → 1.2.4

data/lib/uniword/transformation.rb

@@ -34,5 +34,7 @@ module Uniword
              "#{__dir__}/transformation/mhtml_element_renderer"
     autoload :MhtmlMetadataBuilder,
              "#{__dir__}/transformation/mhtml_metadata_builder"
+    autoload :YamlCssGenerator,
+             "#{__dir__}/transformation/yaml_css_generator"
   end
 end

data/lib/uniword/validation/checkers/internal_link_checker.rb

@@ -133,10 +133,9 @@ module Uniword
           return [] unless document.is_a?(Uniword::Wordprocessingml::DocumentRoot)
 
           document.paragraphs.select do |p|
-            p.is_a?(Lutaml::Model::Serializable) && p.class.attributes.key?(:style) && p.style&.match?(/^Heading/)
+            p.style&.to_s&.match?(/^Heading/)
           end.filter_map do |p|
-            # Generate bookmark from heading text
-            p.is_a?(Lutaml::Model::Serializable) && p.class.attributes.key?(:text) ? heading_to_bookmark(p.text) : nil
+            heading_to_bookmark(p.text)
           end
         end
 

data/lib/uniword/version.rb

@@ -1,5 +1,5 @@
 # frozen_string_literal: true
 
 module Uniword
-  VERSION = "1.0.10"
+  VERSION = "1.2.4"
 end

data/lib/uniword/visitor/text_extractor.rb

@@ -103,7 +103,8 @@ module Uniword
       # @param run [Run] The run to visit
       # @return [void]
       def visit_run(run)
-        @text_parts << run.text if run.text
+        text = run.text_string
+        @text_parts << text unless text.empty?
       end
 
       # Visit an image element.

data/lib/uniword/watermark/manager.rb

@@ -108,11 +108,11 @@ module Uniword
         return false unless paragraph.properties
         return false unless paragraph.properties.is_a?(Uniword::Wordprocessingml::ParagraphProperties)
 
-        paragraph.properties.style.to_s == "UniwordWatermark"
+        paragraph.style == "UniwordWatermark"
       end
 
       def extract_watermark_text(paragraph)
-        paragraph.runs.map { |r| r.text.to_s }.join
+        paragraph.runs.map(&:text_string).join
       end
     end
   end

data/lib/uniword/wordprocessingml/allow_png.rb

@@ -0,0 +1,14 @@
+# frozen_string_literal: true
+
+require "lutaml/model"
+
+module Uniword
+  module Wordprocessingml
+    class AllowPng < Lutaml::Model::Serializable
+      xml do
+        element "allowPNG"
+        namespace Uniword::Ooxml::Namespaces::WordProcessingML
+      end
+    end
+  end
+end

data/lib/uniword/wordprocessingml/body.rb

@@ -66,35 +66,70 @@ module Uniword
         sync_element_order
       end
 
+      # Record a programmatically-added element in element_order.
+      # Called by DocumentBuilder to preserve insertion order when
+      # interleaving paragraphs, tables, and SDTs.
+      def append_to_element_order(element_name)
+        entry = Lutaml::Xml::Element.new("Element", element_name)
+        if element_order.nil? || element_order.empty?
+          self.element_order = [entry]
+        else
+          self.element_order = element_order + [entry]
+        end
+      end
+
       private
 
       def sync_element_order
         return if element_order.nil? || element_order.empty?
 
-        counts = element_order.each_with_object(Hash.new(0)) do |e, h|
-          h[e.name] += 1
+        # lutaml-model freezes element_order after XML parsing.
+        # Build a mutable copy to append missing elements.
+        new_entries = []
+
+        # Count how many p/tbl/sdt entries exist in element_order
+        ordered_p_count = element_order.count { |e| e.name == "p" }
+        ordered_tbl_count = element_order.count { |e| e.name == "tbl" }
+        ordered_sdt_count = element_order.count { |e| e.name == "sdt" }
+        ordered_bookmark_start_count = element_order.count do |e|
+          e.name == "bookmarkStart"
+        end
+        ordered_bookmark_end_count = element_order.count do |e|
+          e.name == "bookmarkEnd"
         end
 
-        append_missing_elements("p", paragraphs.size - counts["p"])
-        append_missing_elements("tbl", tables.size - counts["tbl"])
-        append_missing_elements("sdt",
-                                structured_document_tags.size - counts["sdt"])
-        append_missing_elements("bookmarkStart",
-                                bookmark_starts.size - counts["bookmarkStart"])
-        append_missing_elements("bookmarkEnd",
-                                bookmark_ends.size - counts["bookmarkEnd"])
+        # Collect missing paragraphs
+        (paragraphs.size - ordered_p_count).times do
+          new_entries << Lutaml::Xml::Element.new("Element", "p")
+        end
 
-        return unless section_properties && counts["sectPr"].zero?
+        # Collect missing tables
+        (tables.size - ordered_tbl_count).times do
+          new_entries << Lutaml::Xml::Element.new("Element", "tbl")
+        end
 
-        element_order << build_order_element("sectPr")
-      end
+        # Collect missing structured document tags
+        (structured_document_tags.size - ordered_sdt_count).times do
+          new_entries << Lutaml::Xml::Element.new("Element", "sdt")
+        end
 
-      def append_missing_elements(name, count)
-        count.times { element_order << build_order_element(name) }
-      end
+        # Collect missing bookmark starts
+        (bookmark_starts.size - ordered_bookmark_start_count).times do
+          new_entries << Lutaml::Xml::Element.new("Element", "bookmarkStart")
+        end
+
+        # Collect missing bookmark ends
+        (bookmark_ends.size - ordered_bookmark_end_count).times do
+          new_entries << Lutaml::Xml::Element.new("Element", "bookmarkEnd")
+        end
+
+        # Ensure section_properties is in element_order if present
+        if section_properties && element_order.none? { |e| e.name == "sectPr" }
+          new_entries << Lutaml::Xml::Element.new("Element", "sectPr")
+        end
 
-      def build_order_element(name)
-        Lutaml::Xml::Element.new("Element", name)
+        # Replace frozen array with mutable copy containing new entries
+        self.element_order = element_order + new_entries unless new_entries.empty?
       end
     end
   end

data/lib/uniword/wordprocessingml/body_div.rb

@@ -0,0 +1,17 @@
+# frozen_string_literal: true
+
+require "lutaml/model"
+
+module Uniword
+  module Wordprocessingml
+    class BodyDiv < Lutaml::Model::Serializable
+      attribute :val, :string
+
+      xml do
+        element "bodyDiv"
+        namespace Uniword::Ooxml::Namespaces::WordProcessingML
+        map_attribute "val", to: :val
+      end
+    end
+  end
+end

data/lib/uniword/wordprocessingml/div_border.rb

@@ -0,0 +1,23 @@
+# frozen_string_literal: true
+
+require "lutaml/model"
+
+module Uniword
+  module Wordprocessingml
+    class DivBorder < Lutaml::Model::Serializable
+      attribute :val, :string
+      attribute :sz, :string
+      attribute :space, :string
+      attribute :color, :string
+
+      xml do
+        element "border"
+        namespace Uniword::Ooxml::Namespaces::WordProcessingML
+        map_attribute "val", to: :val
+        map_attribute "sz", to: :sz
+        map_attribute "space", to: :space
+        map_attribute "color", to: :color
+      end
+    end
+  end
+end

data/lib/uniword/wordprocessingml/div_borders.rb

@@ -0,0 +1,24 @@
+# frozen_string_literal: true
+
+require "lutaml/model"
+
+module Uniword
+  module Wordprocessingml
+    class DivBorders < Lutaml::Model::Serializable
+      attribute :top, DivBorder
+      attribute :left, DivBorder
+      attribute :bottom, DivBorder
+      attribute :right, DivBorder
+
+      xml do
+        element "divBdr"
+        namespace Uniword::Ooxml::Namespaces::WordProcessingML
+        mixed_content
+        map_element "top", to: :top, render_nil: false
+        map_element "left", to: :left, render_nil: false
+        map_element "bottom", to: :bottom, render_nil: false
+        map_element "right", to: :right, render_nil: false
+      end
+    end
+  end
+end

data/lib/uniword/wordprocessingml/divs_child.rb

@@ -0,0 +1,17 @@
+# frozen_string_literal: true
+
+require "lutaml/model"
+
+module Uniword
+  module Wordprocessingml
+    # Defined as a stub first to break the WebDiv ↔ DivsChild mutual recursion.
+    # The :div attribute is added after WebDiv is defined.
+    class DivsChild < Lutaml::Model::Serializable
+      xml do
+        element "divsChild"
+        namespace Uniword::Ooxml::Namespaces::WordProcessingML
+        mixed_content
+      end
+    end
+  end
+end

data/lib/uniword/wordprocessingml/document_root.rb

@@ -2,9 +2,6 @@
 
 require "lutaml/model"
 
-require_relative "../document_input"
-require_relative "document_root/feature_facade"
-
 module Uniword
   module Wordprocessingml
     # Root element of a WordprocessingML document
@@ -12,6 +9,8 @@ module Uniword
     # Generated from OOXML schema: wordprocessingml.yml
     # Element: <w:document>
     class DocumentRoot < Lutaml::Model::Serializable
+      autoload :FeatureFacade, "#{__dir__}/document_root/feature_facade"
+
       include FeatureFacade
       include Uniword::DocumentInput
 
@@ -112,9 +111,13 @@ module Uniword
       attr_accessor :chart_parts
       # Bibliography sources for sources.xml
       attr_accessor :bibliography_sources
+      # OLE/embedded object binaries (word/embeddings/*)
+      attr_accessor :embeddings
       # Round-trip parts (copied from DocxPackage during load)
       attr_accessor :settings, :font_table, :web_settings, :document_rels, :theme_rels,
                     :package_rels, :content_types, :custom_properties, :custom_xml_items
+      # Central ID allocator — preserves IDs across build/save cycle
+      attr_accessor :allocator
 
       # Writers for properties that have lazy-initialized getters
       # (removing from attr_accessor to avoid shadowing custom getters)

data/lib/uniword/wordprocessingml/endnote.rb

@@ -19,7 +19,7 @@ module Uniword
         mixed_content
 
         map_attribute "id", to: :id
-        map_attribute "type", to: :type
+        map_attribute "type", to: :type, render_nil: false
         map_element "p", to: :paragraphs, render_nil: false
       end
     end

data/lib/uniword/wordprocessingml/endnotes.rb

@@ -9,6 +9,7 @@ module Uniword
     # Generated from OOXML schema: wordprocessingml.yml
     # Element: <w:endnotes>
     class Endnotes < Lutaml::Model::Serializable
+      attribute :mc_ignorable, Uniword::Ooxml::Types::McIgnorable
       attribute :endnote_entries, Endnote, collection: true,
                                            initialize_empty: true
 
@@ -17,6 +18,9 @@ module Uniword
         namespace Uniword::Ooxml::Namespaces::WordProcessingML
         mixed_content
 
+        namespace_scope Uniword::Ooxml::Namespaces::DOCUMENT_PART_SCOPES
+
+        map_attribute "Ignorable", to: :mc_ignorable, render_nil: false
         map_element "endnote", to: :endnote_entries, render_nil: false
       end
 

data/lib/uniword/wordprocessingml/footer.rb

@@ -9,6 +9,7 @@ module Uniword
     # Generated from OOXML schema: wordprocessingml.yml
     # Element: <w:ftr>
     class Footer < Lutaml::Model::Serializable
+      attribute :mc_ignorable, Uniword::Ooxml::Types::McIgnorable
       attribute :paragraphs, Paragraph, collection: true, initialize_empty: true
       attribute :tables, Table, collection: true, initialize_empty: true
 
@@ -17,6 +18,9 @@ module Uniword
         namespace Uniword::Ooxml::Namespaces::WordProcessingML
         mixed_content
 
+        namespace_scope Uniword::Ooxml::Namespaces::DOCUMENT_PART_SCOPES
+
+        map_attribute "Ignorable", to: :mc_ignorable, render_nil: false
         map_element "p", to: :paragraphs, render_nil: false
         map_element "tbl", to: :tables, render_nil: false
       end

data/lib/uniword/wordprocessingml/footer_reference.rb

@@ -1,7 +1,6 @@
 # frozen_string_literal: true
 
 require "lutaml/model"
-require_relative "../properties/relationship_id"
 
 module Uniword
   module Wordprocessingml

data/lib/uniword/wordprocessingml/footnote.rb

@@ -19,7 +19,7 @@ module Uniword
         mixed_content
 
         map_attribute "id", to: :id
-        map_attribute "type", to: :type
+        map_attribute "type", to: :type, render_nil: false
         map_element "p", to: :paragraphs, render_nil: false
       end
     end

data/lib/uniword/wordprocessingml/footnotes.rb

@@ -9,6 +9,7 @@ module Uniword
     # Generated from OOXML schema: wordprocessingml.yml
     # Element: <w:footnotes>
     class Footnotes < Lutaml::Model::Serializable
+      attribute :mc_ignorable, Uniword::Ooxml::Types::McIgnorable
       attribute :footnote_entries, Footnote, collection: true,
                                              initialize_empty: true
 
@@ -17,6 +18,9 @@ module Uniword
         namespace Uniword::Ooxml::Namespaces::WordProcessingML
         mixed_content
 
+        namespace_scope Uniword::Ooxml::Namespaces::DOCUMENT_PART_SCOPES
+
+        map_attribute "Ignorable", to: :mc_ignorable, render_nil: false
         map_element "footnote", to: :footnote_entries, render_nil: false
       end
 

data/lib/uniword/wordprocessingml/header.rb

@@ -9,6 +9,7 @@ module Uniword
     # Generated from OOXML schema: wordprocessingml.yml
     # Element: <w:hdr>
     class Header < Lutaml::Model::Serializable
+      attribute :mc_ignorable, Uniword::Ooxml::Types::McIgnorable
       attribute :paragraphs, Paragraph, collection: true, initialize_empty: true
       attribute :tables, Table, collection: true, initialize_empty: true
 
@@ -17,6 +18,9 @@ module Uniword
         namespace Uniword::Ooxml::Namespaces::WordProcessingML
         mixed_content
 
+        namespace_scope Uniword::Ooxml::Namespaces::DOCUMENT_PART_SCOPES
+
+        map_attribute "Ignorable", to: :mc_ignorable, render_nil: false
         map_element "p", to: :paragraphs, render_nil: false
         map_element "tbl", to: :tables, render_nil: false
       end

data/lib/uniword/wordprocessingml/header_reference.rb

@@ -1,7 +1,6 @@
 # frozen_string_literal: true
 
 require "lutaml/model"
-require_relative "../properties/relationship_id"
 
 module Uniword
   module Wordprocessingml

data/lib/uniword/wordprocessingml/hyperlink.rb

@@ -1,7 +1,7 @@
 # frozen_string_literal: true
 
 require "lutaml/model"
-require_relative "../has_run_position"
+require "uniword/ooxml/types/relationship_id"
 
 module Uniword
   module Wordprocessingml
@@ -12,7 +12,7 @@ module Uniword
     class Hyperlink < Lutaml::Model::Serializable
       include Uniword::HasRunPosition
 
-      attribute :id, :string
+      attribute :id, Ooxml::Types::RelationshipId
       attribute :anchor, :string
       attribute :tooltip, :string
       attribute :history, Uniword::Properties::HistoryValue

data/lib/uniword/wordprocessingml/instr_text.rb

@@ -9,7 +9,7 @@ module Uniword
     # Generated from OOXML schema: wordprocessingml.yml
     # Element: <w:instrText>
     class InstrText < Lutaml::Model::Serializable
-      attribute :text, :string
+      attribute :content, :string, collection: true, initialize_empty: true
       attribute :xml_space, :string
 
       xml do
@@ -17,17 +17,32 @@ module Uniword
         namespace Uniword::Ooxml::Namespaces::WordProcessingML
         mixed_content
 
+        map_content to: :content
         map_attribute "xml:space", to: :xml_space
-        map_element "text", to: :text, render_nil: false
       end
 
       def initialize(attrs = {})
-        if attrs[:text].is_a?(String) && (attrs[:text].start_with?(" ") || attrs[:text].end_with?(" ") || attrs[:text].include?("\t"))
-          attrs[:xml_space] =
-            "preserve"
+        if attrs.is_a?(Hash)
+          text_val = attrs.delete(:text) || attrs[:content]
+          if text_val.is_a?(String)
+            attrs[:xml_space] = "preserve" if Text.preserve_whitespace?(text_val)
+            attrs[:content] = [text_val] unless attrs.key?(:content)
+          end
         end
         super
       end
+
+      def text
+        content&.join
+      end
+
+      def text=(value)
+        self.content = value.is_a?(Array) ? value : [value.to_s]
+      end
+
+      def to_s
+        text.to_s
+      end
     end
   end
 end

data/lib/uniword/wordprocessingml/level.rb

@@ -66,6 +66,7 @@ module Uniword
     class Level < Lutaml::Model::Serializable
       attribute :ilvl, :integer
       attribute :tentative, :string
+      attribute :tplc, :string
       attribute :start, Start
       attribute :numFmt, NumFmt
       attribute :pStyle, PStyle
@@ -85,6 +86,7 @@ module Uniword
 
         map_attribute "ilvl", to: :ilvl
         map_attribute "tentative", to: :tentative, render_nil: false
+        map_attribute "tplc", to: :tplc, render_nil: false
         map_element "start", to: :start, render_nil: false
         map_element "numFmt", to: :numFmt, render_nil: false
         map_element "pStyle", to: :pStyle, render_nil: false

data/lib/uniword/wordprocessingml/mar_bottom.rb

@@ -0,0 +1,17 @@
+# frozen_string_literal: true
+
+require "lutaml/model"
+
+module Uniword
+  module Wordprocessingml
+    class MarBottom < Lutaml::Model::Serializable
+      attribute :val, :string
+
+      xml do
+        element "marBottom"
+        namespace Uniword::Ooxml::Namespaces::WordProcessingML
+        map_attribute "val", to: :val
+      end
+    end
+  end
+end

data/lib/uniword/wordprocessingml/mar_left.rb

@@ -0,0 +1,17 @@
+# frozen_string_literal: true
+
+require "lutaml/model"
+
+module Uniword
+  module Wordprocessingml
+    class MarLeft < Lutaml::Model::Serializable
+      attribute :val, :string
+
+      xml do
+        element "marLeft"
+        namespace Uniword::Ooxml::Namespaces::WordProcessingML
+        map_attribute "val", to: :val
+      end
+    end
+  end
+end

data/lib/uniword/wordprocessingml/mar_right.rb

@@ -0,0 +1,17 @@
+# frozen_string_literal: true
+
+require "lutaml/model"
+
+module Uniword
+  module Wordprocessingml
+    class MarRight < Lutaml::Model::Serializable
+      attribute :val, :string
+
+      xml do
+        element "marRight"
+        namespace Uniword::Ooxml::Namespaces::WordProcessingML
+        map_attribute "val", to: :val
+      end
+    end
+  end
+end

data/lib/uniword/wordprocessingml/mar_top.rb

@@ -0,0 +1,17 @@
+# frozen_string_literal: true
+
+require "lutaml/model"
+
+module Uniword
+  module Wordprocessingml
+    class MarTop < Lutaml::Model::Serializable
+      attribute :val, :string
+
+      xml do
+        element "marTop"
+        namespace Uniword::Ooxml::Namespaces::WordProcessingML
+        map_attribute "val", to: :val
+      end
+    end
+  end
+end

data/lib/uniword/wordprocessingml/num.rb

@@ -10,6 +10,7 @@ module Uniword
     # Element: <w:num>
     class Num < Lutaml::Model::Serializable
       attribute :numId, :integer
+      attribute :durable_id, :string
       attribute :abstractNumId, AbstractNumId
       attribute :lvlOverrides, LevelOverride, collection: true,
                                               initialize_empty: true
@@ -20,6 +21,9 @@ module Uniword
         mixed_content
 
         map_attribute "numId", to: :numId
+        map_attribute "durableId", to: :durable_id,
+                                   namespace: Uniword::Ooxml::Namespaces::Word2016Cid,
+                                   render_nil: false
         map_element "abstractNumId", to: :abstractNumId, render_nil: false
         map_element "lvlOverride", to: :lvlOverrides, render_nil: false
       end

data/lib/uniword/wordprocessingml/numbering.rb

@@ -18,28 +18,7 @@ module Uniword
         namespace Uniword::Ooxml::Namespaces::WordProcessingML
         mixed_content
 
-        # Force namespace declarations on root for mc:Ignorable prefixes
-        namespace_scope [
-          { namespace: Uniword::Ooxml::Namespaces::MarkupCompatibility,
-            declare: :always },
-          { namespace: Uniword::Ooxml::Namespaces::Word2010, declare: :always },
-          { namespace: Uniword::Ooxml::Namespaces::Word2012, declare: :always },
-          { namespace: Uniword::Ooxml::Namespaces::Word2015Symex,
-            declare: :always },
-          { namespace: Uniword::Ooxml::Namespaces::Word2016Cid,
-            declare: :always },
-          { namespace: Uniword::Ooxml::Namespaces::Word2018, declare: :always },
-          { namespace: Uniword::Ooxml::Namespaces::Word2018Cex,
-            declare: :always },
-          { namespace: Uniword::Ooxml::Namespaces::Word2020SdtDataHash,
-            declare: :always },
-          { namespace: Uniword::Ooxml::Namespaces::Word2024SdtFormatLock,
-            declare: :always },
-          { namespace: Uniword::Ooxml::Namespaces::Word2023Du,
-            declare: :always },
-          { namespace: Uniword::Ooxml::Namespaces::Word2010Drawing,
-            declare: :always },
-        ]
+        namespace_scope Uniword::Ooxml::Namespaces::DOCUMENT_PART_SCOPES
 
         map_element "abstractNum", to: :abstractNums, render_nil: false
         map_element "num", to: :nums, render_nil: false

data/lib/uniword/wordprocessingml/optimize_for_browser.rb

@@ -0,0 +1,14 @@
+# frozen_string_literal: true
+
+require "lutaml/model"
+
+module Uniword
+  module Wordprocessingml
+    class OptimizeForBrowser < Lutaml::Model::Serializable
+      xml do
+        element "optimizeForBrowser"
+        namespace Uniword::Ooxml::Namespaces::WordProcessingML
+      end
+    end
+  end
+end

data/lib/uniword/wordprocessingml/page_defaults.rb

@@ -0,0 +1,39 @@
+# frozen_string_literal: true
+
+module Uniword
+  module Wordprocessingml
+    # Shared page layout defaults.
+    #
+    # Provides canonical values for page size, margins, columns, and docGrid
+    # used by both builder and reconciler.
+    module PageDefaults
+      LETTER_WIDTH  = 12_240
+      LETTER_HEIGHT = 15_840
+
+      DEFAULT_MARGINS = {
+        top: 1440, right: 1440, bottom: 1440, left: 1440,
+        header: 720, footer: 720, gutter: 0,
+      }.freeze
+
+      DEFAULT_COL_SPACE = 720
+      DEFAULT_LINE_PITCH = 360
+
+      def self.default_page_size
+        Wordprocessingml::PageSize.new(width: LETTER_WIDTH,
+                                       height: LETTER_HEIGHT)
+      end
+
+      def self.default_page_margins
+        Wordprocessingml::PageMargins.new(**DEFAULT_MARGINS)
+      end
+
+      def self.default_columns
+        Wordprocessingml::Columns.new(space: DEFAULT_COL_SPACE)
+      end
+
+      def self.default_doc_grid
+        Wordprocessingml::DocGrid.new(line_pitch: DEFAULT_LINE_PITCH)
+      end
+    end
+  end
+end