hexapdf 0.43.0 → 0.45.0

data/lib/hexapdf/layout/text_box.rb

@@ -42,7 +42,46 @@ module HexaPDF
     # A TextBox is used for drawing text, either inside a rectangular box or by flowing it around
     # objects of a Frame.
     #
+    # The standard usage is through the helper methods Document::Layout#text and
+    # Document::Layout#formatted_text.
+    #
     # This class uses TextLayouter behind the scenes to do the hard work.
+    #
+    # == Used Box Properties
+    #
+    # The spacing after the last line can be controlled via the style property +last_line_gap+. Also
+    # see TextLayouter#style for other style properties taken into account.
+    #
+    # == Limitations
+    #
+    # When setting the style property 'position' to +:flow+, padding and border to the left and
+    # right as well as a predefined fixed width are not respected and the result will look wrong.
+    #
+    # == Examples
+    #
+    # Showing some text:
+    #
+    #   #>pdf-composer
+    #   composer.box(:text, items: layout.text_fragments("This is some text."))
+    #   # Or easier with the provided convenience method
+    #   composer.text("This is also some text")
+    #
+    # It is possible to flow the text around other objects by using the style property
+    # 'position' with the value +:flow+:
+    #
+    #   #>pdf-composer
+    #   composer.box(:base, width: 30, height: 30,
+    #                style: {margin: 5, position: :float, background_color: "hp-blue-light"})
+    #   composer.text("This is some text. " * 20, position: :flow)
+    #
+    # While top and bottom padding and border can be used with flow positioning, left and right
+    # padding and border are not supported and the result will look wrong:
+    #
+    #   #>pdf-composer
+    #   composer.box(:base, width: 30, height: 30,
+    #                style: {margin: 5, position: :float, background_color: "hp-blue-light"})
+    #   composer.text("This is some text. " * 20, padding: 10, position: :flow,
+    #                 text_align: :justify)
     class TextBox < Box
 
       # Creates a new TextBox object with the given inline items (e.g. TextFragment and InlineBox
@@ -67,94 +106,68 @@ module HexaPDF
         true
       end
 
+      # :nodoc:
+      def draw(canvas, x, y)
+        super(canvas, x + @x_offset, y)
+      end
+
+      # :nodoc:
+      def empty?
+        super && (!@result || @result.lines.empty?)
+      end
+
+      private
+
       # Fits the text box into the Frame.
       #
       # Depending on the 'position' style property, the text is either fit into the current region
       # of the frame using +available_width+ and +available_height+, or fit to the shape of the
       # frame starting from the top (when 'position' is set to :flow).
-      #
-      # The spacing after the last line can be controlled via the style property +last_line_gap+.
-      #
-      # Also see TextLayouter#style for other style properties taken into account.
-      def fit(available_width, available_height, frame)
-        return false if (@initial_width > 0 && @initial_width > available_width) ||
-          (@initial_height > 0 && @initial_height > available_height)
-
+      def fit_content(_available_width, _available_height, frame)
         frame = frame.child_frame(box: self)
-        @width = @x_offset = @height = 0
-        @result = if style.position == :flow
-                    @tl.fit(@items, frame.width_specification, frame.shape.bbox.height,
+        @x_offset = 0
+
+        if style.position == :flow
+          height = (@initial_height > 0 ? @initial_height : frame.shape.bbox.height) - reserved_height
+          @result = @tl.fit(@items, frame.width_specification(reserved_height_top), height,
                             apply_first_text_indent: !split_box?, frame: frame)
-                  else
-                    @width = reserved_width
-                    @height = reserved_height
-                    width = (@initial_width > 0 ? @initial_width : available_width) - @width
-                    height = (@initial_height > 0 ? @initial_height : available_height) - @height
-                    @tl.fit(@items, width, height, apply_first_text_indent: !split_box?, frame: frame)
-                  end
-        @width += if @initial_width > 0 || style.text_align == :center || style.text_align == :right
-                    width
-                  elsif style.position == :flow
-                    min_x = +Float::INFINITY
-                    max_x = -Float::INFINITY
-                    @result.lines.each do |line|
-                      min_x = [min_x, line.x_offset].min
-                      max_x = [max_x, line.x_offset + line.width].max
-                    end
-                    min_x.finite? ? (@x_offset = min_x; max_x - min_x) : 0
-                  else
-                    @result.lines.max_by(&:width)&.width || 0
-                  end
-        @height += if @initial_height > 0 || style.text_valign == :center || style.text_valign == :bottom
-                     height
-                   else
-                     @result.height
-                   end
-        if style.last_line_gap && @result.lines.last
-          @height += style.line_spacing.gap(@result.lines.last, @result.lines.last)
+          min_x = +Float::INFINITY
+          max_x = -Float::INFINITY
+          @result.lines.each do |line|
+            min_x = [min_x, line.x_offset].min
+            max_x = [max_x, line.x_offset + line.width].max
+          end
+          @width = (min_x.finite? ? (@x_offset = min_x; max_x - min_x) : 0) + reserved_width
+          @height = @initial_height > 0 ? @initial_height : @result.height + reserved_height
+        else
+          @result = @tl.fit(@items, @width - reserved_width, @height - reserved_height,
+                            apply_first_text_indent: !split_box?, frame: frame)
+          if style.text_align == :left && @initial_width == 0
+            @width = (@result.lines.max_by(&:width)&.width || 0) + reserved_width
+          end
+          if style.text_valign == :top && @initial_height == 0
+            @height = @result.height + reserved_height
+          end
         end
 
-        @result.status == :success ||
-          (@result.status == :height && @initial_height > 0 && style.overflow == :truncate)
-      end
-
-      # Splits the text box into two boxes if necessary and possible.
-      def split(available_width, available_height, frame)
-        fit(available_width, available_height, frame) unless @result
-
-        if style.position != :flow && (float_compare(@width, available_width) > 0 ||
-                                       float_compare(@height, available_height) > 0)
-          [nil, self]
-        elsif @result.remaining_items.empty?
-          [self]
-        elsif @result.lines.empty?
-          [nil, self]
-        else
-          [self, create_box_for_remaining_items]
+        if style.last_line_gap && @result.lines.last && @initial_height == 0
+          @height += style.line_spacing.gap(@result.lines.last, @result.lines.last)
         end
-      end
 
-      # :nodoc:
-      def draw(canvas, x, y)
-        super(canvas, x + @x_offset, y)
+        if @result.status == :success
+          fit_result.success!
+        elsif @result.status == :height && !@result.lines.empty?
+          fit_result.overflow!
+        end
       end
 
-      # :nodoc:
-      def empty?
-        super && (!@result || @result.lines.empty?)
+      # Splits the text box into two.
+      def split_content
+        [self, create_box_for_remaining_items]
       end
 
-      private
-
       # Draws the text into the box.
       def draw_content(canvas, x, y)
-        return unless @result
-
-        if @result.status == :height && @initial_height > 0 && style.overflow == :error
-          raise HexaPDF::Error, "Text doesn't fit into box with limited height and " \
-            "style property overflow is set to :error"
-        end
-
         return if @result.lines.empty?
         @result.draw(canvas, x - @x_offset, y + content_height)
       end

data/lib/hexapdf/layout/text_fragment.rb

@@ -52,7 +52,7 @@ module HexaPDF
     # The items of a text fragment may be frozen to indicate that the fragment is potentially used
     # multiple times.
     #
-    # The rectangle with the bottom left corner (#x_min, #y_min) and the top right corner (#x_max,
+    # The rectangle with the bottom-left corner (#x_min, #y_min) and the top-right corner (#x_max,
     # #y_max) describes the minimum bounding box of the whole text fragment and is usually *not*
     # equal to the box (0, 0)-(#width, #height).
     class TextFragment

data/lib/hexapdf/layout/text_layouter.rb

@@ -218,7 +218,10 @@ module HexaPDF
 
         # Breaks are detected at: space, tab, zero-width-space, non-breaking space, hyphen,
         # soft-hypen and any valid Unicode newline separator
-        BREAK_RE = /[ \u{A}-\u{D}\u{85}\u{2028}\u{2029}\t\u{200B}\u{00AD}\u{00A0}-]/
+        BREAK_CHARS = {}
+        " \u{A}\u{B}\u{C}\u{D}\u{85}\u{2028}\u{2029}\t\u{200B}\u{00AD}\u{00A0}-".each_char do |c|
+          BREAK_CHARS[c] = true
+        end
 
         # Breaks the items (an array of InlineBox and TextFragment objects) into atomic pieces
         # wrapped by Box, Glue or Penalty items, and returns those as an array.
@@ -235,7 +238,7 @@ module HexaPDF
                 # Collect characters and kerning values until break character is encountered
                 box_items = []
                 while (glyph = item.items[i]) &&
-                    (glyph.kind_of?(Numeric) || !BREAK_RE.match?(glyph.str))
+                    (glyph.kind_of?(Numeric) || !BREAK_CHARS.key?(glyph.str))
                   box_items << glyph
                   i += 1
                 end
@@ -428,9 +431,7 @@ module HexaPDF
           end
 
           line = create_unjustified_line
-          last_line_used = true
-          last_line_used = yield(line, nil) if item.nil? && !line.items.empty?
-
+          last_line_used = (item.nil? && !line.items.empty? ? yield(line, nil) : true)
           item.nil? && last_line_used ? [] : @items[@beginning_of_line_index..-1]
         end
 
@@ -500,9 +501,7 @@ module HexaPDF
           end
 
           line = create_unjustified_line
-          last_line_used = true
-          last_line_used = yield(line, nil) if item.nil? && !line.items.empty?
-
+          last_line_used = (item.nil? && !line.items.empty? ? yield(line, nil) : true)
           item.nil? && last_line_used ? [] : @items[@beginning_of_line_index..-1]
         end
 

data/lib/hexapdf/parser.rb

@@ -365,11 +365,14 @@ module HexaPDF
         # Need to iterate through the whole lines array in case there are multiple %%EOF to try
         eof_index = 0
         while (eof_index = lines[0..(eof_index - 1)].rindex {|l| l.strip == '%%EOF' })
-          if lines[eof_index - 1].strip =~ /\Astartxref\s(\d+)\z/
+          if eof_index > 0 && lines[eof_index - 1].strip =~ /\Astartxref\s(\d+)\z/
             startxref_offset = $1.to_i
             startxref_mangled = true
             break # we found it even if it the syntax is not entirely correct
-          elsif eof_index < 2 || lines[eof_index - 2].strip != "startxref"
+          elsif eof_index < 2
+            startxref_missing = true
+            break
+          elsif lines[eof_index - 2].strip != "startxref"
             startxref_missing = true
           else
             startxref_offset = lines[eof_index - 1].to_i

data/lib/hexapdf/rectangle.rb

@@ -48,8 +48,8 @@ module HexaPDF
   #
   #   [left, bottom, right, top]
   #
-  # where +left+ is the bottom left x-coordinate, +bottom+ is the bottom left y-coordinate, +right+
-  # is the top right x-coordinate and +top+ is the top right y-coordinate.
+  # where +left+ is the bottom-left x-coordinate, +bottom+ is the bottom-left y-coordinate, +right+
+  # is the top-right x-coordinate and +top+ is the top-right y-coordinate.
   #
   # See: PDF2.0 s7.9.5
   class Rectangle < HexaPDF::PDFArray
@@ -119,8 +119,8 @@ module HexaPDF
     #:nodoc:
     RECTANGLE_ERROR_MSG = "A PDF rectangle structure must contain an array of four numbers"
 
-    # Ensures that the value is an array containing four numbers that specify the bottom left and
-    # top right corner.
+    # Ensures that the value is an array containing four numbers that specify the bottom-left and
+    # top-right corners.
     def after_data_change
       super
       unless value.size == 4 && all? {|v| v.kind_of?(Numeric) }

data/lib/hexapdf/type/file_specification.rb

@@ -158,11 +158,11 @@ module HexaPDF
       end
 
       # :call-seq:
-      #   file_spec.embed(filename, name: File.basename(filename), register: true)   -> ef_stream
-      #   file_spec.embed(io, name:, register: true)                                 -> ef_stream
+      #   file_spec.embed(filename, name: File.basename(filename), mime_type: nil, register: true)   -> ef_stream
+      #   file_spec.embed(io, name:, mime_type: nil, register: true)                                 -> ef_stream
       #
-      # Embeds the given file or IO stream into the PDF file, sets the path accordingly and returns
-      # the created stream object.
+      # Embeds the given file or IO stream into the PDF file, sets the path and MIME type
+      # accordingly and returns the created stream object.
       #
       # If a file is given, the +name+ option defaults to the basename of the file. However, if an
       # IO object is given, the +name+ argument is mandatory.
@@ -177,13 +177,16 @@ module HexaPDF
       # name::
       #     The name that should be used as path value and when registering.
       #
+      # mime_type::
+      #     Optionally specifies the MIME type of the file.
+      #
       # register::
       #     Specifies whether the embedded file will be added to the EmbeddedFiles name tree under
       #     the +name+. If the name is already taken, it's value is overwritten.
       #
       # The file has to be available until the PDF document gets written because reading and
       # writing is done lazily.
-      def embed(file_or_io, name: nil, register: true)
+      def embed(file_or_io, name: nil, mime_type: nil, register: true)
         name ||= File.basename(file_or_io) if file_or_io.kind_of?(String)
         if name.nil?
           raise ArgumentError, "The name argument is mandatory when given an IO object"
@@ -194,6 +197,7 @@ module HexaPDF
 
         self[:EF] ||= {}
         ef_stream = self[:EF][:UF] = self[:EF][:F] = document.add({Type: :EmbeddedFile})
+        ef_stream[:Subtype] = mime_type.to_sym if mime_type
         stat = if file_or_io.kind_of?(String)
                  File.stat(file_or_io)
                elsif file_or_io.respond_to?(:stat)

data/lib/hexapdf/type/form.rb

@@ -159,8 +159,8 @@ module HexaPDF
       # retained without the need for parsing its contents.
       #
       # If the bounding box of the form XObject doesn't have its origin at (0, 0), the canvas origin
-      # is translated into the bottom left corner so that this detail doesn't matter when using the
-      # canvas. This means that the canvas' origin is always at the bottom left corner of the
+      # is translated into the bottom-left corner so that this detail doesn't matter when using the
+      # canvas. This means that the canvas' origin is always at the bottom-left corner of the
       # bounding box.
       #
       # *Note* that a canvas can only be retrieved for initially empty form XObjects!

data/lib/hexapdf/type/graphics_state_parameter.rb

@@ -51,7 +51,7 @@ module HexaPDF
 
       define_type :ExtGState
 
-      define_field :Type,          type: Symbol, required: true, default: type
+      define_field :Type,          type: Symbol, default: type
       define_field :LW,            type: Numeric, version: "1.3"
       define_field :LC,            type: Integer, version: "1.3"
       define_field :LJ,            type: Integer, version: "1.3"

data/lib/hexapdf/version.rb

@@ -37,6 +37,6 @@
 module HexaPDF
 
   # The version of HexaPDF.
-  VERSION = '0.43.0'
+  VERSION = '0.45.0'
 
 end

data/test/hexapdf/content/test_canvas_composer.rb

@@ -47,21 +47,20 @@ describe HexaPDF::Content::CanvasComposer do
     end
 
     it "splits the box if possible" do
-      @composer.draw_box(create_box(width: 400, style: {position: :float}))
-      box = create_box(width: 400, height: 100)
-      box.define_singleton_method(:split) do |*|
-        [box, HexaPDF::Layout::Box.new(height: 100) {}]
-      end
+      @composer.draw_box(create_box(width: 300, height: 300, style: {position: :float}))
+      box = create_box(style: {mask_mode: :box})
+      box.define_singleton_method(:fit_content) {|*| fit_result.overflow! }
+      box.define_singleton_method(:split_content) { [box, HexaPDF::Layout::Box.new(height: 100) {}] }
       @composer.draw_box(box)
       assert_operators(@composer.canvas.contents,
                        [[:save_graphics_state],
-                        [:concatenate_matrix, [1, 0, 0, 1, 0, 0]],
+                        [:concatenate_matrix, [1, 0, 0, 1, 0, 541.889764]],
                         [:restore_graphics_state],
                         [:save_graphics_state],
-                        [:concatenate_matrix, [1, 0, 0, 1, 400, 741.889764]],
+                        [:concatenate_matrix, [1, 0, 0, 1, 300, 0]],
                         [:restore_graphics_state],
                         [:save_graphics_state],
-                        [:concatenate_matrix, [1, 0, 0, 1, 400, 641.889764]],
+                        [:concatenate_matrix, [1, 0, 0, 1, 0, 441.889764]],
                         [:restore_graphics_state]])
     end
 
@@ -77,6 +76,12 @@ describe HexaPDF::Content::CanvasComposer do
                         [:restore_graphics_state]])
     end
 
+    it "handles truncated boxes correctly" do
+      box = create_box(height: 400, style: {overflow: :truncate})
+      box.define_singleton_method(:fit_content) {|*| fit_result.overflow! }
+      assert_same(box, @composer.draw_box(box))
+    end
+
     it "returns the last drawn box" do
       box = create_box(height: 400)
       assert_same(box, @composer.draw_box(box))

data/test/hexapdf/document/test_files.rb

@@ -43,6 +43,11 @@ describe HexaPDF::Document::Files do
       assert_equal('Some file', spec[:Desc])
     end
 
+    it "optionally sets the MIME type of an embedded file" do
+      spec = @doc.files.add(@file.path, mime_type: 'application/pdf')
+      assert_equal(:'application/pdf', spec.embedded_file_stream[:Subtype])
+    end
+
     it "requires the name argument when given an IO object" do
       assert_raises(ArgumentError) { @doc.files.add(StringIO.new) }
     end

data/test/hexapdf/document/test_layout.rb

@@ -141,6 +141,22 @@ describe HexaPDF::Document::Layout do
     end
   end
 
+  describe "styles" do
+    it "returns the existing styles" do
+      @layout.style(:test, font_size: 20)
+      assert_equal([:base, :test], @layout.styles.keys)
+    end
+
+    it "sets multiple styles at once" do
+      styles = @layout.styles(
+        test: {font_size: 20},
+        test2: {font_size: 30},
+      )
+      assert_same(styles, @layout.styles)
+      assert_equal([:base, :test, :test2], @layout.styles.keys)
+    end
+  end
+
   describe "inline_box" do
     it "takes a box as argument" do
       box = HexaPDF::Layout::Box.create(width: 10, height: 10)

data/test/hexapdf/document/test_metadata.rb

@@ -187,6 +187,27 @@ describe HexaPDF::Document::Metadata do
       assert_equal(metadata, @doc.catalog[:Metadata].stream.sub(/(?<=id=")\w+/, ''))
     end
 
+    it "writes the custom metadata" do
+      @metadata.delete
+      @metadata.custom_metadata("<rdf:Description>Test</rdf:Description>")
+      @metadata.custom_metadata("<rdf:Description>Test2</rdf:Description>")
+      @doc.write(StringIO.new, update_fields: false)
+      metadata = <<~XMP
+        <?xpacket begin="" id=""?>
+        <x:xmpmeta xmlns:x="adobe:ns:meta/">
+        <rdf:RDF xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#">
+        <rdf:Description rdf:about="" xmlns:pdf="http://ns.adobe.com/pdf/1.3/">
+        <pdf:Producer>HexaPDF version #{HexaPDF::VERSION}</pdf:Producer>
+        </rdf:Description>
+        <rdf:Description>Test</rdf:Description>
+        <rdf:Description>Test2</rdf:Description>
+        </rdf:RDF>
+        </x:xmpmeta>
+        <?xpacket end="r"?>
+      XMP
+      assert_equal(metadata, @doc.catalog[:Metadata].stream.sub(/(?<=id=")\w+/, ''))
+    end
+
     it "writes the XMP metadata" do
       title = HexaPDF::Document::Metadata::LocalizedString.new('Der Titel')
       title.language = 'de'