hexapdf 1.2.0 → 1.3.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 041330c9846091186ddeecd7900a7919bb96da26b3513b047f03c87888efedcd
-  data.tar.gz: ff89d47f1eeac1d2dda4982224403318528401918ccfe7f1bb775d05d8f0e6f3
+  metadata.gz: e0e469ae650b98b48c88b662dab27cb405b676c706639c8c1cf358d6aefc8f53
+  data.tar.gz: d170526233e1e9aa37403c1bd8d2aa38697641c1b4fb4d4a5f6d8f02f299585b
 SHA512:
-  metadata.gz: 508fa118a26d5825f9ae2a105cdf4717b02901fd3095c49cf5855caae7648bf60b9cea414bb6670dd3bd0c82f99b076a8df7ffb79efa60990003468d7f3a9db9
-  data.tar.gz: 613323b8b2a93a01e31ec9e35664e6dc2939b210bbe7d7c055cb861eac6cd530978b14256a55a8e51f2e3804be06808440d5816c73e808df4af3f111f325b9c9
+  metadata.gz: fd18408ed2c2474e3395bf59b3a0281cb23005d16bf7d5b9d93f673dfa131fc215f342018d0cdce2de053db85c000a68c28a76ad980b0432b1a159bd51b1ed0c
+  data.tar.gz: c58b13ed27c980bb9670b9521a12b1640fc3960aea1188f6135951cfd1503a28e7633cb3f6be8c63d478d64a88e0bbdc9b84bdbc982aa3ec000aeac1a8627597

data/CHANGELOG.md

@@ -1,3 +1,48 @@
+## 1.3.0 - 2025-04-23
+
+### Added
+
+* [HexaPDF::Type::Annotations::Square] for rectangle annotations as well as
+  [HexaPDF::Document::Annotations#create_rectangle]
+* [HexaPDF::Type::Annotations::Circle] for ellipse annotations as well as
+  [HexaPDF::Document::Annotations#create_ellipse]
+* Basic appearance generation for push button fields
+* [HexaPDF::Type::Annotation::BorderEffect] type class
+* [HexaPDF::Type::Annotations::BorderEffect] module that provides convenience
+  access to the border effect dictionary
+* [HexaPDF::Document::Layout#style?] and [HexaPDF::Composer#style?] for checking
+  whether a given style (name) exists
+* [HexaPDF::Layout::Style#each_property] for iterating over all set properties
+* [HexaPDF::Layout::Style#merge] for merging another style instance
+* [HexaPDF::Layout::Style#box_options] for specifying box initialization options
+* [HexaPDF::Layout::Style#font_bold] and [HexaPDF::Layout::Style#font_italic]
+  for setting bold and/or italic variants independently of the font name
+* [HexaPDF::PDFArray#map!] for mapping elements in-place
+* [HexaPDF::PDFArray#compact!] for removing `nil` elements
+
+### Changed
+
+* **Breaking change**: [HexaPDF::Type::Annotations::Widget::MarkerStyle::new]
+  got a new positional argument
+* [HexaPDF::Type::Annotations::Widget#marker_style] to allow setting and
+  retrieving the font for push buttons
+* Extracted `#interior_color` from [HexaPDF::Type::Annotations::Line] into
+  [HexaPDF::Type::Annotations::InteriorColor]
+* CLI command `hexapdf inspect` to support decoding Form XObject streams
+* [HexaPDF::Layout::Style#line_spacing] to accept a `LineSpacing` object when
+  setting the value
+
+### Fixed
+
+* Text extraction with macOS Preview due a bug in Preview
+* [HexaPDF::PDFArray#reject!] to work according to documented method signature
+* [HexaPDF::Type::AcroForm::Field#create_widget] to ensure the proper type
+  class is stored in the document in case an embedded widget is extracted
+* [HexaPDF::Type::AcroForm::Form] validation to ensure that all field objects in
+  the field hierarchy are using a field type class
+* [HexaPDF::Type::AcroForm::Form] validation to delete merged fields
+
+
 ## 1.2.0 - 2025-02-10
 
 ### Added

data/README.md

@@ -82,7 +82,7 @@ canvas.text("Hello World!", at: [20, 400])
 doc.write("hello-world.pdf")
 ~~~
 
-For detailed information have a look at the [HexaPDF website][website] where you will the API
+For detailed information have a look at the [HexaPDF website][website] where you will find the API
 documentation, example code and more.
 
 It is recommend to use the HTML API documentation provided by the HexaPDF website as it is enhanced

data/lib/hexapdf/cli/inspect.rb

@@ -188,12 +188,20 @@ module HexaPDF
                   end
             serialize(obj.value, recursive: true) if obj
 
-          when 's', 'stream', 'raw', 'raw-stream'
+          when 's', 'stream', 'raw', 'raw-stream', 'sd'
             if (obj = pdf_object_from_string_reference(data.shift) rescue $stderr.puts($!.message)) &&
                 obj.kind_of?(HexaPDF::Stream)
-              source = (command.start_with?('raw') ? obj.stream_source : obj.stream_decoder)
-              while source.alive? && (stream_data = source.resume)
-                $stdout.write(stream_data)
+              if command == 'sd'
+                if obj.respond_to?(:process_contents)
+                  obj.process_contents(ContentProcessor.new)
+                else
+                  $stderr.puts("Error: The object is not a Form XObject or page")
+                end
+              else
+                source = (command.start_with?('raw') ? obj.stream_source : obj.stream_decoder)
+                while source.alive? && (stream_data = source.resume)
+                  $stdout.write(stream_data)
+                end
               end
             elsif command_parser.verbosity_info?
               $stderr.puts("Note: Object has no stream data")
@@ -427,6 +435,7 @@ module HexaPDF
         ["OID[,GEN] | o[bject] OID[,GEN]", "Print object"],
         ["r[ecursive] OID[,GEN]", "Print object recursively"],
         ["s[tream] OID[,GEN]", "Print filtered stream"],
+        ["sd OID[,GEN]", "Print the decoded stream of a Form XObject or page"],
         ["raw[-stream] OID[,GEN]", "Print raw stream"],
         ["rev[ision] [NUMBER]", "Print or extract revision"],
         ["x[ref] OID[,GEN]", "Print the cross-reference entry"],

data/lib/hexapdf/composer.rb

@@ -261,6 +261,20 @@ module HexaPDF
       @document.layout.style(name, base: base, **properties)
     end
 
+    # Returns +true+ if a style with the given +name+ exists, else +false+.
+    #
+    # See HexaPDF::Document::Layout#style for details; this method is just a thin wrapper around
+    # that method.
+    #
+    # Example:
+    #
+    #   composer.style(:header, font: 'Helvetica')
+    #   composer.style?(:header)     # => true
+    #   composer.style?(:paragraph)  # => false
+    def style?(name)
+      @document.layout.style?(name)
+    end
+
     # :call-seq:
     #    composer.styles              -> styles
     #    composer.styles(**mapping)   -> styles

data/lib/hexapdf/configuration.rb

@@ -709,6 +709,7 @@ module HexaPDF
                         XXAcroFormField: 'HexaPDF::Type::AcroForm::Field',
                         XXAppearanceDictionary: 'HexaPDF::Type::Annotation::AppearanceDictionary',
                         Border: 'HexaPDF::Type::Annotation::Border',
+                        XXBorderEffect: 'HexaPDF::Type::Annotation::BorderEffect',
                         SigFieldLock: 'HexaPDF::Type::AcroForm::SignatureField::LockDictionary',
                         SV: 'HexaPDF::Type::AcroForm::SignatureField::SeedValueDictionary',
                         SVCert: 'HexaPDF::Type::AcroForm::SignatureField::CertificateSeedValueDictionary',
@@ -766,6 +767,8 @@ module HexaPDF
                           Link: 'HexaPDF::Type::Annotations::Link',
                           Widget: 'HexaPDF::Type::Annotations::Widget',
                           Line: 'HexaPDF::Type::Annotations::Line',
+                          Square: 'HexaPDF::Type::Annotations::Square',
+                          Circle: 'HexaPDF::Type::Annotations::Circle',
                           XML: 'HexaPDF::Type::Metadata',
                           GTS_PDFX: 'HexaPDF::Type::OutputIntent',
                           GTS_PDFA1: 'HexaPDF::Type::OutputIntent',
@@ -795,6 +798,8 @@ module HexaPDF
                           Link: 'HexaPDF::Type::Annotations::Link',
                           Widget: 'HexaPDF::Type::Annotations::Widget',
                           Line: 'HexaPDF::Type::Annotations::Line',
+                          Square: 'HexaPDF::Type::Annotations::Square',
+                          Circle: 'HexaPDF::Type::Annotations::Circle',
                         },
                         XXAcroFormField: {
                           Tx: 'HexaPDF::Type::AcroForm::TextField',

data/lib/hexapdf/document/annotations.rb

@@ -69,12 +69,12 @@ module HexaPDF
       # +create_type+ method.
       #
       # The +options+ are passed on the specific annotation creation method.
-      def create(type, page, **options)
+      def create(type, page, *args, **options)
         method_name = "create_#{type}"
         unless respond_to?(method_name)
           raise ArgumentError, "Invalid type specified"
         end
-        send("create_#{type}", page, **options)
+        send("create_#{type}", page, *args, **options)
       end
 
       # :call-seq:
@@ -100,6 +100,64 @@ module HexaPDF
           border_style(color: 0, width: 1)
       end
 
+      # :call-seq:
+      #   annotations.create_rectangle(page, x, y, width, height)  -> annotation
+      #
+      # Creates a rectangle (called "square" in the PDF specification) annotation with the
+      # lower-left corner at (+x+, +y+) and the given +width+ and +height+.
+      #
+      # The rectangle uses a black stroke color, no interior color and a line width of 1pt by
+      # default. It can be further styled using the convenience methods on the returned annotation
+      # object.
+      #
+      # Example:
+      #
+      #   #>pdf-small
+      #   doc.annotations.create_rectangle(doc.pages[0], 20, 20, 20, 60).
+      #     regenerate_appearance
+      #
+      #   doc.annotations.create_rectangle(doc.pages[0], 60, 20, 20, 60).
+      #     border_style(color: "hp-blue", width: 2).
+      #     interior_color("hp-orange").
+      #     regenerate_appearance
+      #
+      # See: Type::Annotations::Square
+      def create_rectangle(page, x, y, w, h)
+        annot = create_and_add_to_page(:Square, page)
+        annot[:Rect] = [x, y, x + w, y + h]
+        annot.border_style(color: 0, width: 1)
+        annot
+      end
+
+      # :call-seq:
+      #   annotations.create_ellipse(page, cx, cy, a:, b:)  -> annotation
+      #
+      # Creates an ellipse (called "circle" in the PDF specification) annotation with the center
+      # point at (+cx+, +cy+), the semi-major axis +a+ and the semi-minor axis +b+.
+      #
+      # The ellipse uses a black stroke color, no interior color and a line width of 1pt by
+      # default. It can be further styled using the convenience methods on the returned annotation
+      # object.
+      #
+      # Example:
+      #
+      #   #>pdf-small
+      #   doc.annotations.create_ellipse(doc.pages[0], 30, 50, a: 15, b: 20).
+      #     regenerate_appearance
+      #
+      #   doc.annotations.create_ellipse(doc.pages[0], 70, 50, a: 15, b: 20).
+      #     border_style(color: "hp-blue", width: 2).
+      #     interior_color("hp-orange").
+      #     regenerate_appearance
+      #
+      # See: Type::Annotations::Circle
+      def create_ellipse(page, x, y, a:, b:)
+        annot = create_and_add_to_page(:Circle, page)
+        annot[:Rect] = [x - a, y - b, x + a, y + b]
+        annot.border_style(color: 0, width: 1)
+        annot
+      end
+
       private
 
       # Returns the root of the destinations name tree.

data/lib/hexapdf/document/layout.rb

@@ -218,6 +218,19 @@ module HexaPDF
         style
       end
 
+      # Returns +true+ if a style with the given +name+ exists, else +false+.
+      #
+      # Example:
+      #
+      #   layout.style(:header, font: 'Helvetica')
+      #   layout.style?(:header)     # => true
+      #   layout.style?(:paragraph)  # => false
+      #
+      # See: #style
+      def style?(name)
+        @styles.key?(name)
+      end
+
       # :call-seq:
       #    layout.styles            -> styles
       #    layout.styles(**mapping)   -> styles
@@ -286,8 +299,9 @@ module HexaPDF
             box_options[:children] = ChildrenCollector.collect(self, &block)
           end
         end
+        style = retrieve_style(style)
         box_class_for_name(name).new(width: width, height: height,
-                                     style: retrieve_style(style), **box_options, &box_block)
+                                     style: style, **style.box_options, **box_options, &box_block)
       end
 
       # Creates an array of HexaPDF::Layout::TextFragment objects for the given +text+.
@@ -354,7 +368,7 @@ module HexaPDF
         box_style = (box_style ? retrieve_style(box_style) : style)
         box_class_for_name(:text).new(items: text_fragments(text, style: style),
                                       width: width, height: height, properties: properties,
-                                      style: box_style)
+                                      style: box_style, **box_style.box_options)
       end
       alias text text_box
 
@@ -457,7 +471,8 @@ module HexaPDF
           end
         end
         box_class_for_name(:text).new(items: data, width: width, height: height,
-                                      properties: properties, style: box_style)
+                                      properties: properties, style: box_style,
+                                      **box_style.box_options)
       end
       alias formatted_text formatted_text_box
 
@@ -479,7 +494,7 @@ module HexaPDF
         style = retrieve_style(style, style_properties)
         image = file.kind_of?(HexaPDF::Stream) ? file : @document.images.add(file)
         box_class_for_name(:image).new(image: image, width: width, height: height,
-                                       properties: properties, style: style)
+                                       properties: properties, style: style, **style.box_options)
       end
       alias image image_box
 
@@ -608,7 +623,8 @@ module HexaPDF
         end
         box_class_for_name(:table).new(cells: cells, column_widths: column_widths, header: header,
                                        footer: footer, cell_style: cell_style, width: width,
-                                       height: height, properties: properties, style: style)
+                                       height: height, properties: properties, style: style,
+                                       **style.box_options)
       end
       alias table table_box
 
@@ -666,6 +682,22 @@ module HexaPDF
         end
       end
 
+      FONT_BOLD_VARIANT_MAPPER = { #:nodoc:
+        nil => {true => :bold, false: :none},
+        none: {true => :bold, false: :none},
+        bold: {true => :bold, false: :none},
+        italic: {true => :bold_italic, false: :italic},
+        bold_italic: {true => :bold_italic, false: :italic},
+      }
+
+      FONT_ITALIC_VARIANT_MAPPER = { #:nodoc:
+        nil => {true => :italic, false: :none},
+        none: {true => :italic, false: :none},
+        italic: {true => :italic, false: :none},
+        bold: {true => :bold_italic, false: :bold},
+        bold_italic: {true => :bold_italic, false: :bold},
+      }
+
       # Retrieves the appropriate HexaPDF::Layout::Style object based on the +style+ and
       # +properties+ arguments.
       #
@@ -690,7 +722,14 @@ module HexaPDF
         end
         unless style.font.respond_to?(:pdf_object)
           name, options = *style.font
-          style.font(@document.fonts.add(name, **(options || {})))
+          options ||= {}
+          if style.font_bold?
+            options[:variant] = FONT_BOLD_VARIANT_MAPPER.dig(options[:variant], style.font_bold)
+          end
+          if style.font_italic?
+            options[:variant] = FONT_ITALIC_VARIANT_MAPPER.dig(options[:variant], style.font_italic)
+          end
+          style.font(@document.fonts.add(name, **options))
         end
         style
       end

data/lib/hexapdf/error.rb

@@ -94,9 +94,17 @@ module HexaPDF
     end
 
     def message # :nodoc:
-      "No glyph for #{glyph.str.inspect} in font '#{glyph.font_wrapper.wrapped_font.full_name}' " \
-        "found. \n\n" \
-        "Use the configuration option 'font.on_missing_glyph' to customize missing glyph handling."
+      str = "No glyph for #{glyph.str.inspect} in font '#{glyph.font_wrapper.wrapped_font.full_name}' " \
+            "found. \n\n"
+      str << if glyph.font_wrapper.font_type == :Type1
+               "The used Type1 font only contains a very limited number of glyphs. TrueType " \
+               "fonts usually provide a much wider array of glyphs. Use the configuration option " \
+               "'font.map' to register appropriate font files. Also have a look at the " \
+               "'font.default' and 'font.fallback' options. "
+             else
+               "Maybe register another #{glyph.font_wrapper.font_type} font that contains the " \
+               "needed glyph and use it as fallback via the configuration option 'font.fallback'."
+             end
     end
 
   end

data/lib/hexapdf/font/true_type/subsetter.rb

@@ -63,6 +63,16 @@ module HexaPDF
         def use_glyph(glyph_id)
           return @glyph_map[glyph_id] if @glyph_map.key?(glyph_id)
           @last_id += 1
+          # Handle codes for ASCII characters \r (13), (, ) (40, 41) and \ (92) specially so that
+          # they never appear in the output (PDF serialization would need to escape them)
+          if @last_id == 13 || @last_id == 40 || @last_id == 92
+            @glyph_map[:"s#{@last_id}"] = @last_id
+            if @last_id == 40
+              @last_id += 1
+              @glyph_map[:"s#{@last_id}"] = @last_id
+            end
+            @last_id += 1
+          end
           @glyph_map[glyph_id] = @last_id
         end
 
@@ -107,7 +117,7 @@ module HexaPDF
           locations = []
 
           @glyph_map.each_key do |old_gid|
-            glyph = orig_glyf[old_gid]
+            glyph = orig_glyf[old_gid.kind_of?(Symbol) ? 0 : old_gid]
             locations << table.size
             data = glyph.raw_data
             if glyph.compound?
@@ -166,7 +176,10 @@ module HexaPDF
         # Adds the components of compound glyphs to the subset.
         def add_glyph_components
           glyf = @font[:glyf]
-          @glyph_map.keys.each {|gid| glyf[gid].components&.each {|cgid| use_glyph(cgid) } }
+          @glyph_map.keys.each do |gid|
+            next if gid.kind_of?(Symbol)
+            glyf[gid].components&.each {|cgid| use_glyph(cgid) }
+          end
         end
 
       end

data/lib/hexapdf/layout/style.rb

@@ -603,11 +603,36 @@ module HexaPDF
       #   style.update(**properties)    -> style
       #
       # Updates the style's properties using the key-value pairs specified by the +properties+ hash.
+      #
+      # Also see: #merge
       def update(**properties)
         properties.each {|key, value| send(key, value) }
         self
       end
 
+      # Yields all set properties.
+      def each_property # :yield: property, value
+        return to_enum(__method__) unless block_given?
+        instance_variables.each do |iv|
+          (val = PROPERTIES[iv]) && yield(val, instance_variable_get(iv))
+        end
+      end
+
+      # :call-seq:
+      #   style.merge(other_style)   -> style
+      #
+      # Merges the set properties of the +other_style+ object into this one.
+      #
+      # Note that merging is done on a per-property basis. So if a complex property is set on
+      # +other_style+ and also on +self+, the +other_style+ value completely overwrites the one from
+      # +self+.
+      #
+      # Also see: #update
+      def merge(other)
+        other.each_property {|property, value| send(property, value) }
+        self
+      end
+
       ##
       # :method: font
       # :call-seq:
@@ -615,8 +640,9 @@ module HexaPDF
       #
       # The font to be used, must be set to a valid font wrapper object before it can be used.
       #
-      # HexaPDF::Composer handles this property specially in that it resolves a set string or array
-      # to a font wrapper object before doing anything else with the style object.
+      # HexaPDF::Document::Layout handles this property - together with #font_bold and #font_italic
+      # - specially in that it resolves a set string or array to a font wrapper object before doing
+      # anything else with the style object.
       #
       # This is the only style property without a default value!
       #
@@ -633,6 +659,48 @@ module HexaPDF
       #   composer.text("Courier Bold", font: "Courier bold")
       #   composer.text("Courier Bold also", font: ["Courier", variant: :bold])
 
+      ##
+      # :method: font_bold
+      # :call-seq:
+      #   font_bold(bold = false)
+      #
+      # Specifies whether the bold variant of the font is used.
+      #
+      # Note that this property only has affect if #font is not already set to a font wrapper
+      # object and if it is set explicitly (i.e. #font_bold? returns +true+).
+      #
+      # See #font, #font_italic
+      #
+      # Examples:
+      #
+      #   #>pdf-composer100
+      #   composer.text("Helvetica bold", font: "Helvetica", font_bold: true)
+      #
+      #   helvetica_bold = composer.document.fonts.add("Helvetica", variant: :bold)
+      #   composer.text("Helvetica bold", font: helvetica_bold, font_bold: false)
+      #   composer.text("Helvetica", font: ["Helvetica", {variant: :bold}], font_bold: false)
+
+      ##
+      # :method: font_italic
+      # :call-seq:
+      #   font_italic(bold = false)
+      #
+      # Specifies whether the italic variant of the font is used.
+      #
+      # Note that this property only has affect if #font is not already set to a font wrapper
+      # object and if it is set explicitly (i.e. #font_italic? returns +true+).
+      #
+      # See #font, #font_bold.
+      #
+      # Examples:
+      #
+      #   #>pdf-composer100
+      #   composer.text("Helvetica italic", font: "Helvetica", font_italic: true)
+      #
+      #   helvetica_bold = composer.document.fonts.add("Helvetica", variant: :italic)
+      #   composer.text("Helvetica italic", font: helvetica_bold, font_italic: false)
+      #   composer.text("Helvetica", font: ["Helvetica", {variant: :italic}], font_italic: false)
+
       ##
       # :method: font_size
       # :call-seq:
@@ -1021,7 +1089,7 @@ module HexaPDF
       #
       # This method can set the line spacing in two ways:
       #
-      # * Using two positional arguments +type+ and +value+.
+      # * Using the positional, mandatory argument +type+ and the optional +value+.
       # * Or a hash with the keys +type+ and +value+.
       #
       # Note that the last line has no additional spacing after it by default. Set #last_line_gap
@@ -1422,8 +1490,33 @@ module HexaPDF
       #   composer.text("This is some longer text that does not appear in two lines.",
       #                 height: 15, overflow: :truncate)
 
-      [
+      ##
+      # :method: box_options
+      # :call-seq:
+      #   box_options(**options)
+      #
+      # Contains initialization arguments for the box instance that is created with this
+      # style. Together with the other style properties this allows the complete specification of a
+      # box instance just via a Style instance.
+      #
+      # Note that this property is only used by the HexaPDF::Document::Layout methods when a box
+      # instance is created. If a box instance is created directly, this property has no effect.
+      #
+      # Examples:
+      #
+      #   #>pdf-composer100
+      #   composer.style(:my_list, box_options: {marker_type: :decimal, item_spacing: 15})
+      #   composer.list(style: :my_list) do |list|
+      #     list.text("This is some text.")
+      #     list.text("This is some other text.")
+      #   end
+
+
+      # :nodoc:
+      PROPERTIES = [
         [:font, "raise HexaPDF::Error, 'No font set'"],
+        [:font_bold, false],
+        [:font_italic, false],
         [:font_size, 10],
         [:line_height, nil],
         [:character_spacing, 0],
@@ -1457,8 +1550,8 @@ module HexaPDF
         [:text_valign, :top, {valid_values: [:top, :center, :bottom]}],
         [:text_indent, 0],
         [:line_spacing, "LineSpacing.new(type: :single)",
-         {setter: "LineSpacing.new(**(value.kind_of?(Symbol) || value.kind_of?(Numeric) ? " \
-           "{type: value, value: extra_arg} : value))",
+         {setter: "LineSpacing.new(**(value.kind_of?(Symbol) || value.kind_of?(Numeric) || " \
+           "value.kind_of?(LineSpacing) ? {type: value, value: extra_arg} : value))",
           extra_args: ", extra_arg = nil"}],
         [:last_line_gap, false, {valid_values: [true, false]}],
         [:fill_horizontal, nil],
@@ -1475,6 +1568,7 @@ module HexaPDF
         [:mask_mode, :default, {valid_values: [:default, :none, :box, :fill_horizontal,
                                                :fill_frame_horizontal, :fill_vertical, :fill]}],
         [:overflow, :error],
+        [:box_options, {}],
       ].each do |name, default, options = {}|
         default = default.inspect unless default.kind_of?(String)
         setter = options.delete(:setter) || "value"
@@ -1500,7 +1594,7 @@ module HexaPDF
           end
         EOF
         alias_method("#{name}=", name)
-      end
+      end.each_with_object({}) {|arr, hash| hash[:"@#{arr.first}"] = arr.first }
 
       ##
       # :method: text_segmentation_algorithm

data/lib/hexapdf/object.rb

@@ -305,8 +305,8 @@ module HexaPDF
       result
     rescue HexaPDF::Error
       raise
-    rescue StandardError
-      yield("Error: Unexpected value encountered", false, self) if block_given?
+    rescue StandardError => e
+      yield("Unexpected error encountered: #{e.message}", false, self) if block_given?
       false
     end
 

data/lib/hexapdf/pdf_array.rb

@@ -143,10 +143,32 @@ module HexaPDF
     #   array.reject! {|item| block }   -> array or nil
     #   array.reject!                   -> Enumerator
     #
-    # Deletes all elements from the array for which the block returns +true+. If no changes were
-    # done, returns +nil+.
+    # Deletes all elements from the array for which the block returns +true+ and returns +self+. If
+    # no changes were done, returns +nil+.
     def reject!
-      value.reject! {|item| yield(process_entry(item)) }
+      return to_enum(__method__) unless block_given?
+      value.reject! {|item| yield(process_entry(item)) } && self
+    end
+
+    # :call-seq:
+    #   array.map! {|item| block }   -> array
+    #   array.map!                   -> Enumerator
+    #
+    # Maps all elements from the array in-place to the respective return value of the block+ and
+    # returns +self+.
+    def map!
+      return to_enum(__method__) unless block_given?
+      value.map! {|item| yield(process_entry(item)) }
+      self
+    end
+
+    # :call-seq:
+    #   array.compact!   -> array or nil
+    #
+    # Removes all +nil+ elements from the array. Returns +self+ if any elements were removed, +nil+
+    # otherwise.
+    def compact!
+      value.compact! && self
     end
 
     # :call-seq:

data/lib/hexapdf/tokenizer.rb

@@ -278,6 +278,9 @@ module HexaPDF
 
     REFERENCE_RE = /[#{WHITESPACE}]+([+]?\d+)[#{WHITESPACE}]+R#{WHITESPACE_OR_DELIMITER_RE}/ # :nodoc:
 
+    WHITESPACE_OR_DELIMITER_LUT = [] # :nodoc:
+    (WHITESPACE + DELIMITER).each_byte {|x| WHITESPACE_OR_DELIMITER_LUT[x] = true }
+
     # Parses the number (integer or real) at the current position.
     #
     # See: PDF2.0 s7.3.3
@@ -285,7 +288,7 @@ module HexaPDF
       prepare_string_scanner(40)
       pos = self.pos
       if (tmp = @ss.scan_integer)
-        if @ss.eos? || @ss.match?(WHITESPACE_OR_DELIMITER_RE)
+        if @ss.eos? || WHITESPACE_OR_DELIMITER_LUT[@ss.peek_byte]
           # Handle object references, see PDF2.0 s7.3.10
           prepare_string_scanner(10)
           if @ss.scan(REFERENCE_RE)