asciidoctor-pdf 1.5.0.beta.7 → 1.5.0.beta.8

data/lib/asciidoctor/pdf/ext/asciidoctor.rb

@@ -1,9 +1,10 @@
 # frozen_string_literal: true
 # NOTE these are either candidates for inclusion in Asciidoctor core or backports
+require_relative 'asciidoctor/logging_shim' unless defined? Asciidoctor::Logging
+require_relative 'asciidoctor/abstract_node'
 require_relative 'asciidoctor/abstract_block'
 require_relative 'asciidoctor/document'
 require_relative 'asciidoctor/section'
 require_relative 'asciidoctor/list'
 require_relative 'asciidoctor/list_item'
-require_relative 'asciidoctor/logging_shim' unless defined? Asciidoctor::Logging
 require_relative 'asciidoctor/image'

data/lib/asciidoctor/pdf/ext/asciidoctor/abstract_node.rb

@@ -0,0 +1,6 @@
+# frozen_string_literal: true
+class Asciidoctor::AbstractNode
+  def remove_attr name
+    @attributes.delete name
+  end unless method_defined? :remove_attr
+end

data/lib/asciidoctor/pdf/ext/core/array.rb

@@ -3,4 +3,8 @@ class Array
   def delete_all *entries
     entries.map {|entry| delete entry }.compact
   end unless method_defined? :delete_all
+
+  def sum
+    reduce(&:+)
+  end unless method_defined? :sum
 end

data/lib/asciidoctor/pdf/ext/core/object.rb

@@ -1,7 +1,7 @@
 # frozen_string_literal: true
 class Object
   # Convert the object to a serialized PDF object.
-  def to_pdf
+  def to_pdf_object
     ::PDF::Core.pdf_object self
   end
 end

data/lib/asciidoctor/pdf/ext/core/string.rb

@@ -12,16 +12,12 @@ class String
 
   # If the string is ASCII only, convert it to a PDF LiteralString object. Otherwise, return self.
   def as_pdf
-    if ascii_only?
-      ::PDF::Core::LiteralString.new(encode ::Encoding::ASCII_8BIT)
-    else
-      self
-    end
+    ascii_only? ? (::PDF::Core::LiteralString.new encode ::Encoding::ASCII_8BIT) : self
   end
 
   # Convert the string to a serialized PDF object. If the string can be encoded as ASCII-8BIT, first convert it to a PDF
   # LiteralString object.
-  def to_pdf
+  def to_pdf_object
     ::PDF::Core.pdf_object as_pdf
   end
 end

data/lib/asciidoctor/pdf/ext/prawn-table/cell/asciidoc.rb

@@ -55,9 +55,10 @@ module Prawn; class Table; class Cell
 
     def draw_content
       pdf = @pdf
-      # NOTE draw_bounded_content adds FPTolerance to width and height, which causes content to overflow
+      # NOTE draw_bounded_content automatically adds FPTolerance to width and height
       pdf.bounds.instance_variable_set :@width, spanned_content_width
-      pdf.bounds.instance_variable_set :@height, spanned_content_height
+      # NOTE we've already reserved the space, so just let the box stretch to the bottom of the page to avoid overflow
+      pdf.bounds.instance_variable_set :@height, pdf.y
       if @valign != :top && (excess_y = spanned_content_height - natural_content_height) > 0
         pdf.move_down(@valign == :center ? (excess_y.fdiv 2) : excess_y)
       end

data/lib/asciidoctor/pdf/ext/prawn.rb

@@ -2,5 +2,6 @@
 # the following are organized under the Asciidoctor::Prawn namespace
 require_relative 'prawn/font/afm'
 require_relative 'prawn/images'
+require_relative 'prawn/formatted_text/box'
 require_relative 'prawn/formatted_text/fragment'
 require_relative 'prawn/extensions'

data/lib/asciidoctor/pdf/ext/prawn/extensions.rb

@@ -10,6 +10,7 @@ module Prawn
 module Extensions
   include ::Asciidoctor::PDF::Measurements
   include ::Asciidoctor::PDF::Sanitizer
+  include ::Asciidoctor::PDF::TextTransformer
 
   FontAwesomeIconSets = %w(fab far fas)
   IconSets = %w(fab far fas fi pf).to_set
@@ -250,7 +251,7 @@ module Extensions
       elsif points.end_with? 'em'
         super(@font_size * points.to_f)
       elsif points.end_with? '%'
-        super(@font_size * (points.to_f / 100.0))
+        super(@font_size * (points.to_f / 100))
       else
         super points.to_f
       end
@@ -429,12 +430,18 @@ module Extensions
     when :uppercase, 'uppercase'
       uppercase_pcdata text
     when :lowercase, 'lowercase'
-      lowercase_mb text
+      lowercase_pcdata text
+    when :capitalize, 'capitalize'
+      capitalize_words_pcdata text
     else
       text
     end
   end
 
+  def hyphenate_text text, hyphenator
+    hyphenate_words_pcdata text, hyphenator
+  end
+
   # Cursor
 
   # Short-circuits the call to the built-in move_up operation

data/lib/asciidoctor/pdf/ext/prawn/formatted_text/box.rb

@@ -0,0 +1,16 @@
+# frozen_string_literal: true
+Prawn::Text::Formatted::Box.prepend (Module.new do
+  def draw_fragment_overlay_styles fragment
+    if (underline = (styles = fragment.styles).include? :underline) || (styles.include? :strikethrough)
+      (doc = fragment.document).save_graphics_state do
+        if (text_decoration_width = (fs = fragment.format_state)[:text_decoration_width])
+          doc.line_width = text_decoration_width
+        end
+        if (text_decoration_color = fs[:text_decoration_color])
+          doc.stroke_color = text_decoration_color
+        end
+        underline ? (doc.stroke_line fragment.underline_points) : (doc.stroke_line fragment.strikethrough_points)
+      end
+    end
+  end
+end)

data/lib/asciidoctor/pdf/ext/prawn/formatted_text/fragment.rb

@@ -7,6 +7,12 @@ Prawn::Text::Formatted::Fragment.prepend (Module.new do
     @text = ''
   end
 
+  # Don't strip soft hyphens when repacking unretrieved fragments
+  def include_trailing_white_space!
+    @format_state.delete :normalized_soft_hyphen
+    super
+  end
+
   # Modify the built-in ascender write method to allow an override value to be
   # specified using the format_state hash.
   def ascender= val

data/lib/asciidoctor/pdf/ext/prawn/images.rb

@@ -14,6 +14,7 @@ module Images
     if ::String === file && (((opts = opts.dup).delete :format) == 'svg' || (file.downcase.end_with? '.svg'))
       #opts[:enable_file_requests_with_root] = (::File.dirname file) unless opts.key? :enable_file_requests_with_root
       #opts[:enable_web_requests] = allow_uri_read if !(opts.key? :enable_web_requests) && (respond_to? :allow_uri_read)
+      #opts[:cache_images] = cache_uri if !(opts.key? :cache_images) && (respond_to? :cache_uri)
       #opts[:fallback_font_name] = fallback_svg_font_name if !(opts.key? :fallback_font_name) && (respond_to? :fallback_svg_font_name)
       if (opts.key? :fit) && (fit = opts.delete :fit) && !opts[:width] && !opts[:height]
         svg (::File.read file, mode: 'r:UTF-8'), opts do |svg_doc|

data/lib/asciidoctor/pdf/formatted_text/inline_image_arranger.rb

@@ -66,7 +66,8 @@ module InlineImageArranger
               width: image_w,
               fallback_font_name: doc.fallback_svg_font_name,
               enable_web_requests: doc.allow_uri_read,
-              enable_file_requests_with_root: (::File.dirname image_path)
+              enable_file_requests_with_root: (::File.dirname image_path),
+              cache_images: doc.cache_uri
           svg_size = image_w ? svg_obj.document.sizing :
               # NOTE convert intrinsic dimensions to points; constrain to content width
               (svg_obj.resize width: [(to_pt svg_obj.document.sizing.output_width, :px), available_w].min)
@@ -137,7 +138,7 @@ module InlineImageArranger
         fragment[:image_width] = image_w
         fragment[:image_height] = image_h
       rescue
-        logger.warn %(could not embed image: #{image_path}; #{$!.message})
+        logger.warn %(could not embed image: #{image_path}; #{$!.message}#{::Prawn::Errors::UnsupportedImageType === $! ? '; install prawn-gmagick gem to add support' : ''})
         drop = true # delegate to cleanup logic in ensure block
       ensure
         # NOTE skip rendering image in scratch document or if image can't be loaded

data/lib/asciidoctor/pdf/formatted_text/transform.rb

@@ -24,6 +24,8 @@ class Transform
     'font_color' => :color,
     'font_family' => :font,
     'font_size' => :size,
+    'text_decoration_color' => :text_decoration_color,
+    'text_decoration_width' => :text_decoration_width,
   }
   #DummyText = ?\u0000
 
@@ -73,6 +75,8 @@ class Transform
           font: theme.link_font_family,
           size: theme.link_font_size,
           styles: (to_styles theme.link_font_style, theme.link_text_decoration),
+          text_decoration_color: theme.link_text_decoration_color,
+          text_decoration_width: theme.link_text_decoration_width,
         }.compact,
         mark: {
           color: theme.mark_font_color,
@@ -98,6 +102,8 @@ class Transform
         (accum[role] ||= {})[:styles] = to_styles theme[%(role_#{role}_font_style)], theme[%(role_#{role}_text_decoration)]
         accum
       end
+      @theme_settings['line-through'] = { styles: [:strikethrough].to_set } unless @theme_settings.key? 'line-through'
+      @theme_settings['underline'] = { styles: [:underline].to_set } unless @theme_settings.key? 'underline'
       unless @theme_settings.key? 'big'
         if (base_font_size_large = theme.base_font_size_large)
           @theme_settings['big'] = { size: %(#{(base_font_size_large / theme.base_font_size.to_f).round 4}em) }
@@ -119,6 +125,8 @@ class Transform
         key: { font: 'Courier', styles: [:italic].to_set },
         link: { color: '0000FF' },
         mark: { background_color: 'FFFF00', callback: [TextBackgroundAndBorderRenderer] },
+        'line-through' => { styles: [:strikethrough].to_set },
+        'underline' => { styles: [:underline].to_set },
         'big' => { size: '1.667em' },
         'small' => { size: '0.8333em' },
       }
@@ -323,16 +331,11 @@ class Transform
     end
     # TODO we could limit to select tags, but doesn't seem to really affect performance
     attrs[:class].split.each do |class_name|
-      case class_name
-      when 'underline'
-        styles << :underline
-      when 'line-through'
-        styles << :strikethrough
-      else
-        fragment.update(@theme_settings[class_name]) {|k, oval, nval| k == :styles ? (nval ? oval.merge(nval) : oval.clear) : nval } if @theme_settings.key? class_name
+      if @theme_settings.key? class_name
+        fragment.update(@theme_settings[class_name]) {|k, oval, nval| k == :styles ? (nval ? oval.merge(nval) : oval.clear) : nval }
         if fragment[:background_color] || (fragment[:border_color] && fragment[:border_width])
           fragment[:callback] = ((fragment[:callback] || []) << TextBackgroundAndBorderRenderer).uniq
-        end 
+        end
       end
     end if attrs.key?(:class)
     fragment.delete(:styles) if styles.empty?

data/lib/asciidoctor/pdf/index_catalog.rb

@@ -1,7 +1,8 @@
 # frozen_string_literal: true
 module Asciidoctor; module PDF
   class IndexCatalog
-    include Sanitizer
+    include ::Asciidoctor::PDF::TextTransformer
+
     LeadingAlphaRx = /^\p{Alpha}/
 
     attr_accessor :start_page_number

data/lib/asciidoctor/pdf/optimizer.rb

@@ -1,4 +1,5 @@
 require 'rghost'
+require 'tmpdir'
 
 module Asciidoctor
 module PDF
@@ -17,14 +18,23 @@ class Optimizer
   end
 
   def generate_file target
-    filename_o = (filename = Pathname.new target).sub_ext '-o.pdf'
-    pdfmark = filename.sub_ext '.pdfmark'
-    (::RGhost::Convert.new target).to :pdf,
-      filename: filename_o.to_s,
-      quality: @quality,
-      d: { Printed: false, CannotEmbedFontPolicy: '/Warning', CompatibilityLevel: @compatibility_level },
-      raw: pdfmark.file? ? pdfmark.to_s : nil
-    filename_o.rename target
+    ::Dir::Tmpname.create ['asciidoctor-pdf-', '.pdf'] do |tmpfile|
+      filename = Pathname.new target
+      filename_o = Pathname.new tmpfile
+      pdfmark = filename.sub_ext '.pdfmark'
+      inputs = pdfmark.file? ? [target, pdfmark.to_s] : target
+      (::RGhost::Convert.new inputs).to :pdf,
+        filename: filename_o.to_s,
+        quality: @quality,
+        d: { Printed: false, CannotEmbedFontPolicy: '/Warning', CompatibilityLevel: @compatibility_level }
+      begin
+        filename_o.rename target
+      rescue ::Errno::EXDEV
+        filename.binwrite filename_o.binread
+        filename_o.unlink
+      end
+    end
+    nil
   end
 end
 end

data/lib/asciidoctor/pdf/pdfmark.rb

@@ -11,21 +11,21 @@ class Pdfmark
   def generate
     doc = @doc
     if doc.attr? 'reproducible'
-      mod_date = creation_date = ::Time.at 0
+      mod_date = creation_date = (::Time.at 0).utc
     else
-      mod_date = ::Time.parse doc.attr 'docdatetime' rescue (now ||= ::Time.now)
-      creation_date = ::Time.parse doc.attr 'localdatetime' rescue (now ||= ::Time.now)
+      mod_date = (::Time.parse doc.attr 'docdatetime') rescue (now ||= ::Time.now)
+      creation_date = (::Time.parse doc.attr 'localdatetime') rescue (now ||= ::Time.now)
     end
     # FIXME use sanitize: :plain_text once available
     content = <<~EOS
-    [ /Title #{sanitize(doc.doctitle use_fallback: true).to_pdf}
-      /Author #{(doc.attr 'authors').to_pdf}
-      /Subject #{(doc.attr 'subject').to_pdf}
-      /Keywords #{(doc.attr 'keywords').to_pdf}
-      /ModDate #{mod_date.to_pdf}
-      /CreationDate #{creation_date.to_pdf}
+    [ /Title #{sanitize(doc.doctitle use_fallback: true).to_pdf_object}
+      /Author #{(doc.attr 'authors').to_pdf_object}
+      /Subject #{(doc.attr 'subject').to_pdf_object}
+      /Keywords #{(doc.attr 'keywords').to_pdf_object}
+      /ModDate #{mod_date.to_pdf_object}
+      /CreationDate #{creation_date.to_pdf_object}
       /Creator (Asciidoctor PDF #{::Asciidoctor::PDF::VERSION}, based on Prawn #{::Prawn::VERSION})
-      /Producer #{(doc.attr 'publisher').to_pdf}
+      /Producer #{(doc.attr 'publisher').to_pdf_object}
       /DOCINFO pdfmark
     EOS
     content

data/lib/asciidoctor/pdf/sanitizer.rb

@@ -1,14 +1,4 @@
 # frozen_string_literal: true
-unless RUBY_VERSION >= '2.4'
-  begin
-    require 'unicode' unless defined? Unicode::VERSION
-  rescue LoadError
-    begin
-      require 'active_support/multibyte' unless defined? ActiveSupport::Multibyte
-    rescue LoadError; end
-  end
-end
-
 module Asciidoctor
 module PDF
 module Sanitizer
@@ -29,9 +19,7 @@ module Sanitizer
     'quot' => ?",
   }).default = ??
   SanitizeXMLRx = /<[^>]+>/
-  XMLMarkupRx = /&#?[a-z\d]+;|</
   CharRefRx = /&(?:([a-z][a-z]+\d{0,2})|#(?:(\d\d\d{0,4})|x([a-f\d][a-f\d][a-f\d]{0,3})));/
-  SiftPCDATARx = /(&#?[a-z\d]+;|<[^>]+>)|([^&<]+)/
 
   # Strip leading, trailing and repeating whitespace, remove XML tags and
   # resolve all entities in the specified string.
@@ -51,51 +39,6 @@ module Sanitizer
   def encode_quotes string
     (string.include? ?") ? (string.gsub ?", '&quot;') : string
   end
-
-  def uppercase_pcdata string
-    if XMLMarkupRx.match? string
-      string.gsub(SiftPCDATARx) { $2 ? (uppercase_mb $2) : $1 }
-    else
-      uppercase_mb string
-    end
-  end
-
-  if RUBY_VERSION >= '2.4'
-    def uppercase_mb string
-      string.upcase
-    end
-
-    def lowercase_mb string
-      string.downcase
-    end
-  # NOTE Unicode library is 4x as fast as ActiveSupport::MultiByte::Chars
-  elsif defined? ::Unicode
-    def uppercase_mb string
-      string.ascii_only? ? string.upcase : (::Unicode.upcase string)
-    end
-
-    def lowercase_mb string
-      string.ascii_only? ? string.downcase : (::Unicode.downcase string)
-    end
-  elsif defined? ::ActiveSupport::Multibyte
-    MultibyteChars = ::ActiveSupport::Multibyte::Chars
-
-    def uppercase_mb string
-      string.ascii_only? ? string.upcase : (MultibyteChars.new string).upcase.to_s
-    end
-
-    def lowercase_mb string
-      string.ascii_only? ? string.downcase : (MultibyteChars.new string).downcase.to_s
-    end
-  else
-    def uppercase_mb string
-      string.upcase
-    end
-
-    def lowercase_mb string
-      string.downcase
-    end
-  end
 end
 end
 end

data/lib/asciidoctor/pdf/text_transformer.rb

@@ -0,0 +1,115 @@
+# frozen_string_literal: true
+unless RUBY_VERSION >= '2.4'
+  begin
+    require 'unicode' unless defined? Unicode::VERSION
+  rescue LoadError
+    begin
+      require 'active_support/multibyte' unless defined? ActiveSupport::Multibyte
+    rescue LoadError; end
+  end
+end
+
+module Asciidoctor
+module PDF
+module TextTransformer
+  XMLMarkupRx = /&#?[a-z\d]+;|</
+  PCDATAFilterRx = /(&#?[a-z\d]+;|<[^>]+>)|([^&<]+)/
+  TagFilterRx = /(<[^>]+>)|([^<]+)/
+  WordRx = /\S+/
+  SoftHyphen = ?\u00ad
+
+  def capitalize_words_pcdata string
+    if XMLMarkupRx.match? string
+      string.gsub(PCDATAFilterRx) { $2 ? (capitalize_words_mb $2) : $1 }
+    else
+      captialize_words_mb string
+    end
+  end
+
+  def capitalize_words_mb string
+    string.gsub(WordRx) { capitalize_mb $& }
+  end
+
+  def hyphenate_words_pcdata string, hyphenator
+    if XMLMarkupRx.match? string
+      string.gsub(PCDATAFilterRx) { $2 ? (hyphenate_words $2, hyphenator) : $1 }
+    else
+      hyphenate_words string, hyphenator
+    end
+  end
+
+  def hyphenate_words string, hyphenator
+    string.gsub(WordRx) { hyphenator.visualize $&, SoftHyphen }
+  end
+
+  def lowercase_pcdata string
+    if string.include? '<'
+      string.gsub(TagFilterRx) { $2 ? (lowercase_mb $2) : $1 }
+    else
+      lowercase_mb string
+    end
+  end
+
+  def uppercase_pcdata string
+    if XMLMarkupRx.match? string
+      string.gsub(PCDATAFilterRx) { $2 ? (uppercase_mb $2) : $1 }
+    else
+      uppercase_mb string
+    end
+  end
+
+  if RUBY_VERSION >= '2.4'
+    def capitalize_mb string
+      string.capitalize
+    end
+
+    def lowercase_mb string
+      string.downcase
+    end
+
+    def uppercase_mb string
+      string.upcase
+    end
+  # NOTE Unicode library is 4x as fast as ActiveSupport::MultiByte::Chars
+  elsif defined? ::Unicode
+    def capitalize_mb string
+      string.ascii_only? ? string.capitalize : (::Unicode.capitalize string)
+    end
+
+    def lowercase_mb string
+      string.ascii_only? ? string.downcase : (::Unicode.downcase string)
+    end
+
+    def uppercase_mb string
+      string.ascii_only? ? string.upcase : (::Unicode.upcase string)
+    end
+  elsif defined? ::ActiveSupport::Multibyte
+    MultibyteChars = ::ActiveSupport::Multibyte::Chars
+
+    def capitalize_mb string
+      string.ascii_only? ? string.capitalize : (MultibyteChars.new string).capitalize.to_s
+    end
+
+    def lowercase_mb string
+      string.ascii_only? ? string.downcase : (MultibyteChars.new string).downcase.to_s
+    end
+
+    def uppercase_mb string
+      string.ascii_only? ? string.upcase : (MultibyteChars.new string).upcase.to_s
+    end
+  else
+    def capitalize_mb string
+      string.capitalize
+    end
+
+    def lowercase_mb string
+      string.downcase
+    end
+
+    def uppercase_mb string
+      string.upcase
+    end
+  end
+end
+end
+end