uniword 1.0.11 → 1.2.4

data/lib/uniword/builder/table_builder.rb

@@ -19,22 +19,19 @@ module Uniword
         Wordprocessingml::Table
       end
 
-      # Create and add a row to the table
-      #
-      # @yield [TableRowBuilder] Builder for row configuration
-      # @return [TableRowBuilder] The row builder
+      def initialize(model = nil)
+        super
+        ensure_table_structure
+      end
+
       def row(&block)
         r = TableRowBuilder.new
         yield(r) if block
         @model.rows << r.build
+        finalize_table_structure
         r
       end
 
-      # Set table width
-      #
-      # @param value [Integer] Width in twips
-      # @param rule [String] Width rule ('auto', 'exact', 'pct')
-      # @return [self]
       def width(value, rule: nil)
         ensure_properties
         @model.properties.table_width ||= Properties::TableWidth.new
@@ -43,10 +40,6 @@ module Uniword
         self
       end
 
-      # Set table indentation
-      #
-      # @param value [Integer] Indent in twips
-      # @return [self]
       def indent(value)
         ensure_properties
         @model.properties.table_indent ||= Properties::TableIndent.new
@@ -62,10 +55,36 @@ module Uniword
 
       private
 
+      def properties_tag
+        "tblPr"
+      end
+
       def ensure_properties
         @model.properties ||= Wordprocessingml::TableProperties.new
+        ensure_properties_in_order
         @model.properties
       end
+
+      def ensure_table_structure
+        ensure_properties
+        @model.properties.table_width ||= Properties::TableWidth.new(w: 0, type: "auto")
+        @model.properties.table_look ||= Wordprocessingml::TableDefaults.default_table_look
+      end
+
+      def finalize_table_structure
+        col_count = @model.rows.map(&:cell_count).max || 0
+        return if col_count.zero?
+
+        @model.grid ||= Wordprocessingml::TableGrid.new
+        existing = @model.grid.columns.size
+        if existing < col_count
+          (col_count - existing).times do
+            @model.grid.columns << Wordprocessingml::GridCol.new
+          end
+        elsif existing > col_count
+          @model.grid.columns = @model.grid.columns.first(col_count)
+        end
+      end
     end
   end
 end

data/lib/uniword/builder/table_cell_builder.rb

@@ -17,6 +17,11 @@ module Uniword
         Wordprocessingml::TableCell
       end
 
+      def initialize(model = nil)
+        super
+        ensure_cell_structure
+      end
+
       # Append content to the cell. Routes by type:
       # - String -> creates a Paragraph with a Run
       # - Paragraph -> appends to paragraphs
@@ -30,12 +35,16 @@ module Uniword
           para = Wordprocessingml::Paragraph.new
           para.runs << Wordprocessingml::Run.new(text: element)
           @model.paragraphs << para
+          track_element_order("p")
         when Wordprocessingml::Paragraph
           @model.paragraphs << element
+          track_element_order("p")
         when Wordprocessingml::Table
           @model.tables << element
+          track_element_order("tbl")
         when ParagraphBuilder
           @model.paragraphs << element.build
+          track_element_order("p")
         else
           raise ArgumentError, "Cannot add #{element.class} to table cell"
         end
@@ -85,10 +94,20 @@ module Uniword
 
       private
 
+      def properties_tag
+        "tcPr"
+      end
+
       def ensure_properties
         @model.properties ||= Wordprocessingml::TableCellProperties.new
+        ensure_properties_in_order
         @model.properties
       end
+
+      def ensure_cell_structure
+        ensure_properties
+        @model.properties.cell_width ||= Properties::CellWidth.new(w: 0, type: "auto")
+      end
     end
   end
 end

data/lib/uniword/builder.rb

@@ -39,6 +39,7 @@ module Uniword
     autoload :SdtBuilder, "uniword/builder/sdt_builder"
     autoload :BibliographyBuilder, "uniword/builder/bibliography_builder"
     autoload :ChartBuilder, "uniword/builder/chart_builder"
+    autoload :DeterministicId, "uniword/builder/deterministic_id"
 
     # Factory: creates a Run with optional formatting
     #
@@ -216,28 +217,35 @@ module Uniword
       def build_field_paragraph(instruction)
         para = Wordprocessingml::Paragraph.new
 
-        # Begin character (inside a run wrapper)
-        fc_begin = Wordprocessingml::FieldChar.new
-        fc_begin.fldCharType = "begin"
-        para.field_chars << fc_begin
-
-        # Instruction text
-        it = Wordprocessingml::InstrText.new
-        it.text = instruction
-        para.instr_text << it
-
-        # Separate character
-        fc_sep = Wordprocessingml::FieldChar.new
-        fc_sep.fldCharType = "separate"
-        para.field_chars << fc_sep
-
-        # End character
-        fc_end = Wordprocessingml::FieldChar.new
-        fc_end.fldCharType = "end"
-        para.field_chars << fc_end
+        # Begin character inside a run
+        run_begin = Wordprocessingml::Run.new
+        run_begin.field_char = Wordprocessingml::FieldChar.new
+        run_begin.field_char.fldCharType = "begin"
+        para.runs << run_begin
+
+        # Instruction text inside a run
+        run_instr = Wordprocessingml::Run.new
+        run_instr.instr_text = Wordprocessingml::InstrText.new
+        run_instr.instr_text.text = instruction
+        para.runs << run_instr
+
+        # Separate character inside a run
+        run_sep = Wordprocessingml::Run.new
+        run_sep.field_char = Wordprocessingml::FieldChar.new
+        run_sep.field_char.fldCharType = "separate"
+        para.runs << run_sep
+
+        # End character inside a run
+        run_end = Wordprocessingml::Run.new
+        run_end.field_char = Wordprocessingml::FieldChar.new
+        run_end.field_char.fldCharType = "end"
+        para.runs << run_end
 
         para
       end
     end
+
+    # Shared run utilities
+    autoload :RunUtils, "uniword/builder/run_utils"
   end
 end

data/lib/uniword/cli/diff_cli.rb

@@ -2,7 +2,6 @@
 
 require "thor"
 require "json"
-require_relative "helpers"
 
 module Uniword
   # Diff subcommands for Uniword CLI.

data/lib/uniword/cli/generate_cli.rb

@@ -1,7 +1,6 @@
 # frozen_string_literal: true
 
 require "thor"
-require_relative "helpers"
 
 module Uniword
   # Generate subcommand for Uniword CLI.

data/lib/uniword/cli/headers_cli.rb

@@ -1,7 +1,6 @@
 # frozen_string_literal: true
 
 require "thor"
-require_relative "helpers"
 
 module Uniword
   # Headers/Footers subcommand for Uniword CLI.

data/lib/uniword/cli/images_cli.rb

@@ -2,7 +2,6 @@
 
 require "thor"
 require "json"
-require_relative "helpers"
 
 module Uniword
   # Images subcommands for Uniword CLI.

data/lib/uniword/cli/main.rb

@@ -2,7 +2,6 @@
 
 require "thor"
 require "rainbow"
-require_relative "helpers"
 require_relative "styleset_cli"
 require_relative "resources_cli"
 require_relative "theme_cli"
@@ -501,17 +500,17 @@ module Uniword
 
     def apply_metadata_updates(core_properties)
       setters = {
-        "set-title" => :title=,
-        "set-author" => :creator=,
-        "set-subject" => :subject=,
-        "set-keywords" => :keywords=,
-        "set-description" => :description=,
+        "set-title" => ->(cp, v) { cp.title = v },
+        "set-author" => ->(cp, v) { cp.creator = v },
+        "set-subject" => ->(cp, v) { cp.subject = v },
+        "set-keywords" => ->(cp, v) { cp.keywords = v },
+        "set-description" => ->(cp, v) { cp.description = v },
       }
       updated = false
       setters.each do |opt, setter|
         next unless options[opt]
 
-        core_properties&.public_send(setter, options[opt])
+        setter.call(core_properties, options[opt]) if core_properties
         updated = true
       end
       updated

data/lib/uniword/cli/protect_cli.rb

@@ -1,7 +1,6 @@
 # frozen_string_literal: true
 
 require "thor"
-require_relative "helpers"
 
 module Uniword
   # Protect subcommand for Uniword CLI.

data/lib/uniword/cli/resources_cli.rb

@@ -1,7 +1,6 @@
 # frozen_string_literal: true
 
 require "thor"
-require_relative "helpers"
 
 module Uniword
   # Word Resources subcommands for Uniword CLI.

data/lib/uniword/cli/review_cli.rb

@@ -1,7 +1,6 @@
 # frozen_string_literal: true
 
 require "thor"
-require_relative "helpers"
 
 module Uniword
   # Review subcommands for Uniword CLI.

data/lib/uniword/cli/spellcheck_cli.rb

@@ -2,7 +2,6 @@
 
 require "thor"
 require "json"
-require_relative "helpers"
 
 module Uniword
   # Spellcheck subcommands for Uniword CLI.

data/lib/uniword/cli/styleset_cli.rb

@@ -1,7 +1,6 @@
 # frozen_string_literal: true
 
 require "thor"
-require_relative "helpers"
 
 module Uniword
   # StyleSet subcommands for Uniword CLI.

data/lib/uniword/cli/template_cli.rb

@@ -3,7 +3,6 @@
 require "thor"
 require "json"
 require "yaml"
-require_relative "helpers"
 
 module Uniword
   # Template subcommands for Uniword CLI.

data/lib/uniword/cli/theme_cli.rb

@@ -1,7 +1,6 @@
 # frozen_string_literal: true
 
 require "thor"
-require_relative "helpers"
 
 module Uniword
   # Theme subcommands for Uniword CLI.

data/lib/uniword/cli/toc_cli.rb

@@ -2,7 +2,6 @@
 
 require "thor"
 require "json"
-require_relative "helpers"
 
 module Uniword
   # TOC subcommands for Uniword CLI.

data/lib/uniword/cli/watermark_cli.rb

@@ -1,7 +1,6 @@
 # frozen_string_literal: true
 
 require "thor"
-require_relative "helpers"
 
 module Uniword
   # Watermark subcommand for Uniword CLI.

data/lib/uniword/cli.rb

@@ -1,20 +1,9 @@
 # frozen_string_literal: true
 
-# CLI entry point — loads decomposed classes from lib/uniword/cli/.
-# Uniword::CLI is the main Thor class. Subcommands are in separate files.
+# CLI namespace. All CLI classes are autoloaded from lib/uniword.rb.
+# This file exists for backwards compatibility with code that
+# requires "uniword/cli" to load the CLI namespace.
 
-require_relative "cli/helpers"
-require_relative "cli/styleset_cli"
-require_relative "cli/resources_cli"
-require_relative "cli/theme_cli"
-require_relative "cli/generate_cli"
-require_relative "cli/review_cli"
-require_relative "cli/diff_cli"
-require_relative "cli/toc_cli"
-require_relative "cli/images_cli"
-require_relative "cli/spellcheck_cli"
-require_relative "cli/template_cli"
-require_relative "cli/headers_cli"
-require_relative "cli/watermark_cli"
-require_relative "cli/protect_cli"
-require_relative "cli/main"
+module Uniword
+  # All CLI autoload entries live in lib/uniword.rb.
+end

data/lib/uniword/diff/document_differ.rb

@@ -1,7 +1,5 @@
 # frozen_string_literal: true
 
-require_relative "diff_result"
-
 module Uniword
   module Diff
     # Compares two DocumentRoot instances and produces a DiffResult.
@@ -442,7 +440,10 @@ module Uniword
       def format_value(props, method)
         return nil unless props
 
-        val = props.public_send(method)
+        val = case method
+              when :alignment then props.alignment
+              when :style then Array(props.style).first
+              end
         unwrap_value(val)
       end
 
@@ -454,7 +455,14 @@ module Uniword
       def run_prop_value(props, method)
         return nil unless props
 
-        val = props.public_send(method)
+        val = case method
+              when :bold then props.bold
+              when :italic then props.italic
+              when :underline then props.underline
+              when :fonts then props.fonts
+              when :size then props.size
+              when :color then props.color
+              end
         return nil if val.nil?
 
         unwrap_value(val)
@@ -468,7 +476,17 @@ module Uniword
       def metadata_value(cp, field)
         return nil unless cp
 
-        val = cp.public_send(field)
+        val = case field
+              when :title then cp.title
+              when :creator then cp.creator
+              when :subject then cp.subject
+              when :keywords then cp.keywords
+              when :description then cp.description
+              when :last_modified_by then cp.last_modified_by
+              when :revision then cp.revision
+              when :created then cp.created
+              when :modified then cp.modified
+              end
         return nil if val.nil?
 
         unwrap_value(val).to_s

data/lib/uniword/diff/package_differ.rb

@@ -3,7 +3,6 @@
 require "zip"
 require "nokogiri"
 require "canon"
-require_relative "package_diff_result"
 
 module Uniword
   module Diff

data/lib/uniword/diff.rb

@@ -14,6 +14,10 @@ module Uniword
     autoload :DocumentDiffer, "uniword/diff/document_differ"
     autoload :Formatter, "uniword/diff/formatter"
     autoload :PackageDiffer, "uniword/diff/package_differ"
+    autoload :XmlChange, "uniword/diff/package_diff_result"
+    autoload :ZipMetadataChange, "uniword/diff/package_diff_result"
+    autoload :OpcIssue, "uniword/diff/package_diff_result"
+    autoload :PartChange, "uniword/diff/package_diff_result"
     autoload :PackageDiffResult, "uniword/diff/package_diff_result"
   end
 end

data/lib/uniword/document_factory.rb

@@ -169,27 +169,6 @@ module Uniword
         detector.detect(path)
       end
 
-      # Mapping from Docx::Package attribute names to DocumentRoot
-      # attribute names. Keys where names differ use explicit mapping.
-      PACKAGE_PART_MAPPINGS = {
-        styles: :styles_configuration,
-        numbering: :numbering_configuration,
-        settings: :settings,
-        font_table: :font_table,
-        web_settings: :web_settings,
-        theme: :theme,
-        core_properties: :core_properties,
-        app_properties: :app_properties,
-        document_rels: :document_rels,
-        theme_rels: :theme_rels,
-        package_rels: :package_rels,
-        content_types: :content_types,
-        custom_properties: :custom_properties,
-        custom_xml_items: :custom_xml_items,
-        footnotes: :footnotes,
-        endnotes: :endnotes,
-        embeddings: :embeddings,
-      }.freeze
 
       # Copy package parts to document for round-trip preservation
       #
@@ -199,10 +178,22 @@ module Uniword
       def copy_package_parts_to_document(package, document)
         return unless document.is_a?(Uniword::Wordprocessingml::DocumentRoot)
 
-        PACKAGE_PART_MAPPINGS.each do |pkg_attr, doc_attr|
-          value = package.public_send(pkg_attr)
-          document.public_send(:"#{doc_attr}=", value) if value
-        end
+        document.styles_configuration = package.styles if package.styles
+        document.numbering_configuration = package.numbering if package.numbering
+        document.settings = package.settings if package.settings
+        document.font_table = package.font_table if package.font_table
+        document.web_settings = package.web_settings if package.web_settings
+        document.theme = package.theme if package.theme
+        document.core_properties = package.core_properties if package.core_properties
+        document.app_properties = package.app_properties if package.app_properties
+        document.document_rels = package.document_rels if package.document_rels
+        document.theme_rels = package.theme_rels if package.theme_rels
+        document.package_rels = package.package_rels if package.package_rels
+        document.content_types = package.content_types if package.content_types
+        document.custom_properties = package.custom_properties if package.custom_properties
+        document.custom_xml_items = package.custom_xml_items if package.custom_xml_items
+        document.footnotes = package.footnotes if package.footnotes
+        document.endnotes = package.endnotes if package.endnotes
       end
 
       private

data/lib/uniword/docx/document_statistics.rb

@@ -2,23 +2,8 @@
 
 module Uniword
   module Docx
-    # Calculates document statistics that appear in docProps/app.xml.
-    #
-    # Verified against Microsoft Word 2024. Rules:
-    #   - Words: whitespace-separated tokens; each CJK char = 1 word
-    #   - Characters: total chars minus whitespace (NO paragraph marks)
-    #   - CharactersWithSpaces: total chars including spaces (NO paragraph marks)
-    #   - Paragraphs: count of non-empty paragraphs (empty paras excluded)
-    #   - Lines: same as non-empty paragraph count (no page layout engine)
-    #   - Pages: simple approximation (no page layout engine)
-    #
-    # Known limitations:
-    #   - Pages and Lines use paragraph-based approximation (no page layout engine)
-    #   - Footnote/endnote text is not included (unclear if Word includes it)
-    #   - Header/footer text is not included (Word likely excludes it)
     class DocumentStatistics
-      # CJK Unified Ideographs and extension ranges
-      CJK_REGEX = /[\u4E00-\u9FFF\u3400-\u4DBF\uF900-\uFAFF\u2F00-\u2FDF\u2E80-\u2EFF]/
+      CJK_REGEX = /[一-鿿㐀-䶿豈-﫿⼀-⿟⺀-⻿]/
       WHITESPACE_REGEX = /[ \t\r\n]/
 
       def initialize(package)
@@ -32,8 +17,9 @@ module Uniword
 
         text_per_paragraph = []
         collect_text(body, text_per_paragraph)
+        collect_notes(text_per_paragraph)
+        collect_headers_footers(text_per_paragraph)
 
-        # Word only counts non-empty paragraphs
         non_empty = text_per_paragraph.reject { |t| t.strip.empty? }
 
         {
@@ -48,45 +34,58 @@ module Uniword
 
       private
 
-      # Recursively collect text from paragraphs, tables, and SDTs.
-      def collect_text(container, text_per_paragraph)
-        if container.is_a?(Uniword::Wordprocessingml::Body) || container.is_a?(Uniword::Wordprocessingml::DocumentRoot) || container.is_a?(Uniword::Wordprocessingml::TableCell)
-          container.paragraphs.each do |para|
-            text_per_paragraph << para.text
-          end
-        end
+      def collect_text(body, text_per_paragraph)
+        collect_paragraphs(body.paragraphs, text_per_paragraph)
+        collect_table_text(body.tables, text_per_paragraph)
+        collect_sdt_text(body.structured_document_tags, text_per_paragraph)
+      end
+
+      def collect_paragraphs(paragraphs, text_per_paragraph)
+        paragraphs.each { |para| text_per_paragraph << para.text }
+      end
 
-        if container.is_a?(Uniword::Wordprocessingml::Body) || container.is_a?(Uniword::Wordprocessingml::DocumentRoot)
-          container.tables.each do |table|
-            table.rows.each do |row|
-              row.cells.each do |cell|
-                collect_text(cell, text_per_paragraph)
-              end
+      def collect_table_text(tables, text_per_paragraph)
+        tables.each do |table|
+          table.rows.each do |row|
+            row.cells.each do |cell|
+              collect_paragraphs(cell.paragraphs, text_per_paragraph)
+              collect_table_text(cell.tables, text_per_paragraph)
             end
           end
         end
+      end
 
-        if container.is_a?(Uniword::Wordprocessingml::Body)
-          container.structured_document_tags.each do |sdt|
-            sdt.content&.paragraphs&.each do |para|
-              text = para.text
-              text_per_paragraph << text
-            end
-          end
+      def collect_sdt_text(sdts, text_per_paragraph)
+        sdts.each do |sdt|
+          next unless sdt.content
+          collect_paragraphs(sdt.content.paragraphs, text_per_paragraph)
+        end
+      end
+
+      def collect_notes(text_per_paragraph)
+        @package.footnotes&.footnote_entries&.each do |entry|
+          entry.paragraphs.each { |p| text_per_paragraph << p.text }
+        end
+        @package.endnotes&.endnote_entries&.each do |entry|
+          entry.paragraphs.each { |p| text_per_paragraph << p.text }
+        end
+      end
+
+      def collect_headers_footers(text_per_paragraph)
+        @package.document&.headers&.each_value do |header|
+          header.paragraphs.each { |p| text_per_paragraph << p.text }
+        end
+        @package.document&.footers&.each_value do |footer|
+          footer.paragraphs.each { |p| text_per_paragraph << p.text }
         end
       end
 
-      # Count words using Word's algorithm:
-      # - Whitespace-separated tokens for Latin text
-      # - Each CJK character counts as 1 word individually
-      # - Mixed CJK/Latin segments are counted appropriately
       def count_words(text_per_paragraph)
         total = 0
         text_per_paragraph.each do |text|
           next if text.strip.empty?
 
-          # Split text into CJK and non-CJK segments
-          segments = text.scan(/[\u4E00-\u9FFF\u3400-\u4DBF\uF900-\uFAFF\u2F00-\u2FDF\u2E80-\u2EFF]+|[^ \t\r\n]+/)
+          segments = text.scan(/[一-鿿㐀-䶿豈-﫿⼀-⿟⺀-⻿]+|[^ \t\r\n]+/)
           segments.each do |seg|
             total += if seg.match?(CJK_REGEX)
                        seg.length
@@ -98,31 +97,20 @@ module Uniword
         total
       end
 
-      # Characters (no spaces) = total chars minus whitespace.
-      #
-      # Verified against Word 2024:
-      #   "Hello World" → 10 (not 11, no paragraph marks)
       def count_characters_no_spaces(text_per_paragraph)
         total_chars = text_per_paragraph.sum(&:length)
         whitespace_count = text_per_paragraph.sum { |t| t.count(" \t") }
         total_chars - whitespace_count
       end
 
-      # Characters (with spaces) = total chars including spaces.
-      #
-      # Verified against Word 2024:
-      #   "Hello World" → 11
       def count_characters_with_spaces(text_per_paragraph)
         text_per_paragraph.sum(&:length)
       end
 
-      # Page estimation without page layout engine.
-      # Uses a simple heuristic: ~45 paragraphs per page.
       def estimate_pages(paragraph_count)
         [1, (paragraph_count / 45.0).ceil].max
       end
 
-      # Line estimation without page layout engine.
       def estimate_lines(paragraph_count)
         [1, paragraph_count].max
       end