hexapdf 0.32.2 → 0.34.0

data/lib/hexapdf/layout/text_box.rb

@@ -54,6 +54,13 @@ module HexaPDF
         @result = nil
       end
 
+      # Returns the text that will be drawn.
+      #
+      # This will ignore any inline boxes or kerning values.
+      def text
+        @items.map {|item| item.kind_of?(TextFragment) ? item.text : '' }.join
+      end
+
       # Returns +true+ as the 'position' style property value :flow is supported.
       def supports_position_flow?
         true
@@ -74,13 +81,14 @@ module HexaPDF
 
         @width = @height = 0
         @result = if style.position == :flow
-                    @tl.fit(@items, frame.width_specification, frame.shape.bbox.height)
+                    @tl.fit(@items, frame.width_specification, frame.shape.bbox.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)
+                    @tl.fit(@items, width, height, apply_first_text_indent: !split_box?, frame: frame)
                   end
         @width += if @initial_width > 0 || style.align == :center || style.align == :right
                     width
@@ -103,7 +111,8 @@ module HexaPDF
       def split(available_width, available_height, frame)
         fit(available_width, available_height, frame) unless @result
 
-        if style.position != :flow && (@width > available_width || @height > available_height)
+        if style.position != :flow && (float_compare(@width, available_width) > 0 ||
+                                       float_compare(@height, available_height) > 0)
           [nil, self]
         elsif @result.remaining_items.empty?
           [self]

data/lib/hexapdf/layout/text_fragment.rb

@@ -111,6 +111,11 @@ module HexaPDF
         @properties = properties
       end
 
+      # Returns the text of the fragment.
+      def text
+        items.reject {|i| i.kind_of?(Numeric) }.map(&:str).join
+      end
+
       # Creates a new TextFragment with the same style and custom properties as this one but with
       # the given +items+.
       def dup_attributes(items)
@@ -259,7 +264,7 @@ module HexaPDF
       # The width of the text fragment.
       #
       # It is the sum of the widths of its items and is calculated by using the algorithm presented
-      # in PDF1.7 s9.4.4. By using kerning values as the first and/or last items, the text contained
+      # in PDF2.0 s9.4.4. By using kerning values as the first and/or last items, the text contained
       # in the fragment may spill over the left and/or right boundary.
       def width
         @width ||= @items.sum {|item| style.scaled_item_width(item) }
@@ -283,6 +288,28 @@ module HexaPDF
         :text
       end
 
+      # Creates a new text fragment that repeats this fragment's items and applies the necessary
+      # spacing so that the returned text fragment fills the given +width+ completely.
+      #
+      # If the given +width+ is less than the fragment's width, +self+ is returned.
+      def fill_horizontal!(width)
+        return self if width < self.width
+
+        factor, rest = width.divmod(self.width)
+        items = @items * factor
+        rest = @items.inject(rest) do |available_width, item|
+          new_available_width = available_width - style.scaled_item_width(item)
+          break available_width if new_available_width < 0
+          items << item
+          new_available_width
+        end
+
+        spacing = rest / (items.size - 1)
+        new_style = @style.dup.update(character_spacing: spacing)
+        items << spacing / new_style.scaled_font_size # correct spacing after last item
+        self.class.new(items, new_style, properties: @properties.dup)
+      end
+
       # Clears all cached values.
       #
       # This method needs to be called if the fragment's items or attributes are changed!
@@ -293,8 +320,7 @@ module HexaPDF
 
       # :nodoc:
       def inspect
-        "#<#{self.class.name} #{items.reject {|i| i.kind_of?(Numeric) }.map(&:str).join.inspect} " \
-          "#{items.inspect}>"
+        "#<#{self.class.name} #{text.inspect} #{items.inspect}>"
       end
 
       private

data/lib/hexapdf/layout/text_layouter.rb

@@ -51,7 +51,8 @@ module HexaPDF
     # * Existing line breaking characters inside of TextFragment objects are respected when fitting
     #   text. If this is not wanted, they have to be removed beforehand.
     #
-    # * The first line may be indented by setting Style#text_indent which may also be negative.
+    # * The first line of each paragraph may be indented by setting Style#text_indent which may also
+    #   be negative.
     #
     # * Text can be fitted into arbitrarily shaped areas, even containing holes.
     #
@@ -339,8 +340,8 @@ module HexaPDF
         # current start of the line index should be stored for later use.
         #
         # After the algorithm is finished, it returns the unused items.
-        def self.call(items, width_block, &block)
-          obj = new(items, width_block)
+        def self.call(items, width_block, frame, &block)
+          obj = new(items, width_block, frame)
           if width_block.arity == 1
             obj.variable_width_wrapping(&block)
           else
@@ -352,9 +353,10 @@ module HexaPDF
 
         # Creates a new line wrapping object that arranges the +items+ on lines with the given
         # width.
-        def initialize(items, width_block)
+        def initialize(items, width_block, frame)
           @items = items
           @width_block = width_block
+          @frame = frame
           @line_items = []
           @width = 0
           @glue_items = []
@@ -362,6 +364,7 @@ module HexaPDF
           @last_breakpoint_index = 0
           @last_breakpoint_line_items_index = 0
           @break_prohibited_state = false
+          @fill_horizontal = false
 
           @height_calc = Line::HeightCalculator.new
           @line = DummyLine.new(0, 0)
@@ -504,9 +507,11 @@ module HexaPDF
         #
         # Returns +true+ if the item could be added and +false+ otherwise.
         def add_box_item(item)
+          item.fit_wrapped_box(@frame&.context) if item.kind_of?(InlineBox)
           return false unless @width + item.width <= @available_width
           @line_items.concat(@glue_items).push(item)
           @width += item.width
+          @fill_horizontal ||= item.style.fill_horizontal
           @glue_items.clear
           true
         end
@@ -546,6 +551,17 @@ module HexaPDF
 
         # Creates a Line object from the current line items.
         def create_line
+          if @fill_horizontal
+            rest_width = @available_width - @width
+            indices = []
+            @line_items.each_with_index do |item, index|
+              next unless item.style.fill_horizontal
+              indices << [index, item.style.fill_horizontal]
+              rest_width += item.width
+            end
+            unit_width = rest_width / indices.sum(&:last)
+            indices.each {|i, count| @line_items[i] = @line_items[i].fill_horizontal!(unit_width * count) }
+          end
           Line.new(@line_items)
         end
 
@@ -565,6 +581,7 @@ module HexaPDF
           @last_breakpoint_index = index
           @last_breakpoint_line_items_index = 0
           @break_prohibited_state = false
+          @fill_horizontal = false
           @available_width = @width_block.call(@line)
         end
 
@@ -658,7 +675,7 @@ module HexaPDF
       end
 
       # :call-seq:
-      #   text_layouter.fit(items, width, height) -> result
+      #   text_layouter.fit(items, width, height, apply_first_text_indent: true) -> result
       #
       # Fits the items into the given area and returns a Result object with all the information.
       #
@@ -693,7 +710,14 @@ module HexaPDF
       # The text segmentation algorithm specified via #style is applied to the items in case they
       # are not already in segmented form. This also means that Result#remaining_items always
       # contains segmented items.
-      def fit(items, width, height)
+      #
+      # Optional arguments:
+      #
+      # +apply_first_text_indent+::
+      #     Specifies whether style.text_indent should be applied to the first line. This should be
+      #     set to +false+ if the items start with a continuation of a paragraph instead of starting
+      #     a new paragraph (e.g. after a page break).
+      def fit(items, width, height, apply_first_text_indent: true, frame: nil)
         unless items.empty? || items[0].respond_to?(:type)
           items = style.text_segmentation_algorithm.call(items)
         end
@@ -704,7 +728,7 @@ module HexaPDF
         rest = items
 
         # processing state variables
-        indent = style.text_indent
+        indent = apply_first_text_indent ? style.text_indent : 0
         line_fragments = []
         line_height = 0
         previous_line = nil
@@ -757,7 +781,7 @@ module HexaPDF
           too_wide_box = nil
           line_height = 0
 
-          rest = style.text_line_wrapping_algorithm.call(rest, width_block) do |line, item|
+          rest = style.text_line_wrapping_algorithm.call(rest, width_block, frame) do |line, item|
             # make sure empty lines broken by mandatory paragraph breaks are not empty
             line << TextFragment.new([], style) if item&.type != :box && line.items.empty?
 

data/lib/hexapdf/layout.rb

@@ -57,6 +57,7 @@ module HexaPDF
     autoload(:ColumnBox, 'hexapdf/layout/column_box')
     autoload(:ListBox, 'hexapdf/layout/list_box')
     autoload(:PageStyle, 'hexapdf/layout/page_style')
+    autoload(:TableBox, 'hexapdf/layout/table_box')
 
   end
 

data/lib/hexapdf/name_tree_node.rb

@@ -55,7 +55,7 @@ module HexaPDF
   # HexaPDF::Utils::SortedTreeNode) to add or retrieve entries. They ensure that the name tree stays
   # valid.
   #
-  # See: PDF1.7 s7.9.6
+  # See: PDF2.0 s7.9.6
   class NameTreeNode < Dictionary
 
     include Utils::SortedTreeNode

data/lib/hexapdf/number_tree_node.rb

@@ -44,7 +44,7 @@ module HexaPDF
   # Number trees are similar to name trees but use integers as keys instead of strings. See
   # HexaPDF::NameTreeNode for a more detailed explanation.
   #
-  # See: PDF1.7 s7.9.7, HexaPDF::NameTreeNode
+  # See: PDF2.0 s7.9.7, HexaPDF::NameTreeNode
   class NumberTreeNode < Dictionary
 
     include Utils::SortedTreeNode

data/lib/hexapdf/object.rb

@@ -117,7 +117,7 @@ module HexaPDF
   #
   # See: HexaPDF::Dictionary, HexaPDF::Stream, HexaPDF::Reference, HexaPDF::Document
   #
-  # See: PDF1.7 s7.3.10, s7.3.8
+  # See: PDF2.0 s7.3.10, s7.3.8
   class Object
 
     include Comparable
@@ -143,18 +143,27 @@ module HexaPDF
 
     # Makes sure that the object itself as well as all nested values are direct objects.
     #
+    # The +document+ argument needs to contain the Document instance to which +object+ belongs so
+    # that references can be correctly resolved.
+    #
     # If an indirect object is found, it is turned into a direct object and the indirect object is
     # deleted from the document.
-    def self.make_direct(object)
+    def self.make_direct(object, document)
       if object.kind_of?(HexaPDF::Object) && object.indirect?
+        raise HexaPDF::Error, "Can't make a stream object a direct object" if object.data.stream
         object_to_delete = object
         object = object.value
         object_to_delete.document.delete(object_to_delete)
       end
-      if object.kind_of?(Hash)
-        object.transform_values! {|val| make_direct(val) }
-      elsif object.kind_of?(Array)
-        object.map! {|val| make_direct(val) }
+      case object
+      when HexaPDF::Object
+        object.data.value = make_direct(object.data.value, document)
+      when Hash
+        object.transform_values! {|val| make_direct(val, document) }
+      when Array
+        object.map! {|val| make_direct(val, document) }
+      when Reference
+        object = make_direct(document.object(object), document)
       end
       object
     end
@@ -255,7 +264,7 @@ module HexaPDF
     # type.
     #
     # However, the Type and Subtype fields can easily be used for this. Subclasses for PDF objects
-    # that don't have such fields may use a unique name that has to begin with XX (see PDF1.7 sE.2)
+    # that don't have such fields may use a unique name that has to begin with XX (see PDF2.0 sE.2)
     # and therefore doesn't clash with names defined by the PDF specification.
     #
     # For basic objects this always returns +:Unknown+.
@@ -297,6 +306,8 @@ module HexaPDF
     end
 
     # Makes a deep copy of the source PDF object and resets the object identifier.
+    #
+    # Note that indirect references are *not* copied! If that is also needed, use Importer::copy.
     def deep_copy
       obj = dup
       obj.instance_variable_set(:@data, @data.dup)

data/lib/hexapdf/parser.rb

@@ -41,11 +41,11 @@ require 'hexapdf/xref_section'
 
 module HexaPDF
 
-  # Parses an IO stream according to PDF1.7 to get at the contained objects.
+  # Parses an IO stream according to PDF2.0 to get at the contained objects.
   #
   # This class also contains higher-level methods for getting indirect objects and revisions.
   #
-  # See: PDF1.7 s7
+  # See: PDF2.0 s7
   class Parser
 
     # The IO stream which is parsed.
@@ -125,7 +125,7 @@ module HexaPDF
     #
     # Returns an array containing [object, oid, gen, stream].
     #
-    # See: PDF1.7 s7.3.10, s7.3.8
+    # See: PDF2.0 s7.3.10, s7.3.8
     def parse_indirect_object(offset = nil)
       @tokenizer.pos = offset + @header_offset if offset
       oid = @tokenizer.next_token
@@ -267,7 +267,7 @@ module HexaPDF
     #
     # This method can only parse cross-reference sections, not cross-reference streams!
     #
-    # See: PDF1.7 s7.5.4, s7.5.5; ADB1.7 sH.3-3.4.3
+    # See: PDF2.0 s7.5.4, s7.5.5; ADB1.7 sH.3-3.4.3
     def parse_xref_section_and_trailer(offset)
       @tokenizer.pos = offset + @header_offset
       token = @tokenizer.next_token
@@ -346,7 +346,7 @@ module HexaPDF
     #
     # If strict parsing is disabled, the whole file is searched for the offset.
     #
-    # See: PDF1.7 s7.5.5, ADB1.7 sH.3-3.4.4
+    # See: PDF2.0 s7.5.5, ADB1.7 sH.3-3.4.4
     def startxref_offset
       return @startxref_offset if defined?(@startxref_offset)
 
@@ -397,7 +397,7 @@ module HexaPDF
 
     # Returns the PDF version number that is stored in the file header.
     #
-    # See: PDF1.7 s7.5.2
+    # See: PDF2.0 s7.5.2
     def file_header_version
       unless @header_version
         raise_malformed("PDF file header is missing or corrupt", pos: 0)
@@ -413,7 +413,7 @@ module HexaPDF
     # restriction so that the header may appear in the first 1024 bytes. We follow the Adobe
     # convention.
     #
-    # See: PDF1.7 s7.5.2, ADB1.7 sH.3-3.4.1
+    # See: PDF2.0 s7.5.2, ADB1.7 sH.3-3.4.1
     def retrieve_pdf_header_offset_and_version
       @io.seek(0)
       @header_offset = (@io.read(1024) || '').index(/%PDF-(\d\.\d)/) || 0

data/lib/hexapdf/pdf_array.rb

@@ -44,7 +44,7 @@ module HexaPDF
   # #[] method. Therefore not all Array methods are implemented - use the #value directly if other
   # methods are needed.
   #
-  # See: PDF1.7 s7.3.6
+  # See: PDF2.0 s7.3.6
   class PDFArray < HexaPDF::Object
 
     include Enumerable

data/lib/hexapdf/rectangle.rb

@@ -51,7 +51,7 @@ module HexaPDF
   # 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: PDF1.7 s7.9.5
+  # See: PDF2.0 s7.9.5
   class Rectangle < HexaPDF::PDFArray
 
     # Returns the x-coordinate of the bottom-left corner.

data/lib/hexapdf/reference.rb

@@ -50,7 +50,7 @@ module HexaPDF
   # keys. Furthermore the implementation is compatible to the one of Object, i.e. the hash of a
   # Reference object is the same as the hash of an indirect Object.
   #
-  # See: PDF1.7 s7.3.10, Object
+  # See: PDF2.0 s7.3.10, Object
   class Reference
 
     include Comparable

data/lib/hexapdf/revision.rb

@@ -48,7 +48,7 @@ module HexaPDF
   # If a revision doesn't have an associated cross-reference section, it wasn't created from a PDF
   # file.
   #
-  # See: PDF1.7 s7.5.6, Revisions
+  # See: PDF2.0 s7.5.6, Revisions
   class Revision
 
     include Enumerable

data/lib/hexapdf/revisions.rb

@@ -55,7 +55,7 @@ module HexaPDF
   # this should only be done if one is familiar with the inner workings of HexaPDF. Otherwise it is
   # best to use the convenience methods of this class to create, access or delete indirect objects.
   #
-  # See: PDF1.7 s7.5.6, HexaPDF::Revision
+  # See: PDF2.0 s7.5.6, HexaPDF::Revision
   class Revisions
 
     class << self
@@ -76,7 +76,7 @@ module HexaPDF
           seen_xref_offsets = {}
 
           while offset && !seen_xref_offsets.key?(offset)
-            # PDF1.7 s7.5.5 states that :Prev needs to be indirect, Adobe's reference 3.4.4 says it
+            # PDF2.0 s7.5.5 states that :Prev needs to be indirect, Adobe's reference 3.4.4 says it
             # should be direct. Adobe's POV is followed here. Same with :XRefStm.
             xref_section, trailer = parser.load_revision(offset)
             seen_xref_offsets[offset] = true
@@ -167,7 +167,7 @@ module HexaPDF
     # For references to unknown objects, +nil+ is returned but free objects are represented by a
     # PDF Null object, not by +nil+!
     #
-    # See: PDF1.7 s7.3.9
+    # See: PDF2.0 s7.3.9
     def object(ref)
       i = @revisions.size - 1
       while i >= 0

data/lib/hexapdf/serializer.rb

@@ -79,7 +79,7 @@ module HexaPDF
   #
   # If no serialization method for a specific class is found, the ancestors classes are tried.
   #
-  # See: PDF1.7 s7.3
+  # See: PDF2.0 s7.3
   class Serializer
 
     # The encrypter to use for encrypting strings and streams. If +nil+, strings and streams are not
@@ -163,21 +163,21 @@ module HexaPDF
 
     # Serializes the +nil+ value.
     #
-    # See: PDF1.7 s7.3.9
+    # See: PDF2.0 s7.3.9
     def serialize_nilclass(_obj)
       "null"
     end
 
     # Serializes the +true+ value.
     #
-    # See: PDF1.7 s7.3.2
+    # See: PDF2.0 s7.3.2
     def serialize_trueclass(_obj)
       "true"
     end
 
     # Serializes the +false+ value.
     #
-    # See: PDF1.7 s7.3.2
+    # See: PDF2.0 s7.3.2
     def serialize_falseclass(_obj)
       "false"
     end
@@ -187,21 +187,21 @@ module HexaPDF
     # This method should be used for cases where it is known that the object is either an Integer
     # or a Float.
     #
-    # See: PDF1.7 s7.3.3
+    # See: PDF2.0 s7.3.3
     def serialize_numeric(obj)
       obj.kind_of?(Integer) ? obj.to_s : serialize_float(obj)
     end
 
     # Serializes an Integer object.
     #
-    # See: PDF1.7 s7.3.3
+    # See: PDF2.0 s7.3.3
     def serialize_integer(obj)
       obj.to_s
     end
 
     # Serializes a Float object.
     #
-    # See: PDF1.7 s7.3.3
+    # See: PDF2.0 s7.3.3
     def serialize_float(obj)
       if -0.0001 < obj && obj < 0.0001 && obj != 0
         sprintf("%.6f", obj)
@@ -215,7 +215,7 @@ module HexaPDF
     # The regexp matches all characters that need to be escaped and the substs hash contains the
     # mapping from these characters to their escaped form.
     #
-    # See PDF1.7 s7.3.5
+    # See PDF2.0 s7.3.5
     NAME_SUBSTS = {} # :nodoc:
     [0..32, 127..255, Tokenizer::DELIMITER.bytes, Tokenizer::WHITESPACE.bytes, [35]].each do |a|
       a.each {|c| NAME_SUBSTS[c.chr] = "##{c.to_s(16).rjust(2, '0')}" }
@@ -225,7 +225,7 @@ module HexaPDF
 
     # Serializes a Symbol object (i.e. a PDF name object).
     #
-    # See: PDF1.7 s7.3.5
+    # See: PDF2.0 s7.3.5
     def serialize_symbol(obj)
       NAME_CACHE[obj] ||=
         begin
@@ -240,7 +240,7 @@ module HexaPDF
 
     # Serializes an Array object.
     #
-    # See: PDF1.7 s7.3.6
+    # See: PDF2.0 s7.3.6
     def serialize_array(obj)
       str = +"["
       index = 0
@@ -256,7 +256,7 @@ module HexaPDF
 
     # Serializes a Hash object (i.e. a PDF dictionary object).
     #
-    # See: PDF1.7 s7.3.7
+    # See: PDF2.0 s7.3.7
     def serialize_hash(obj)
       str = +"<<"
       obj.each do |k, v|
@@ -274,7 +274,7 @@ module HexaPDF
 
     # Serializes a String object.
     #
-    # See: PDF1.7 s7.3.4
+    # See: PDF2.0 s7.3.4
     def serialize_string(obj)
       obj = if @encrypter && @object.kind_of?(HexaPDF::Object) && @object.indirect?
               encrypter.encrypt_string(obj, @object)
@@ -294,7 +294,7 @@ module HexaPDF
     # The ISO PDF specification differs in respect to the supported date format. When converting
     # to a date string, a format suitable for both is output.
     #
-    # See: PDF1.7 s7.9.4, ADB1.7 3.8.3
+    # See: PDF2.0 s7.9.4, ADB1.7 3.8.3
     def serialize_time(obj)
       zone = obj.strftime("%z'")
       if zone == "+0000'"
@@ -330,14 +330,14 @@ module HexaPDF
       end
     end
 
-    # See: PDF1.7 s7.3.10
+    # See: PDF2.0 s7.3.10
     def serialize_hexapdf_reference(obj)
       "#{obj.oid} #{obj.gen} R"
     end
 
     # Serializes the streams dictionary and its stream.
     #
-    # See: PDF1.7 s7.3.8
+    # See: PDF2.0 s7.3.8
     def serialize_hexapdf_stream(obj)
       if !obj.indirect?
         raise HexaPDF::Error, "Can't serialize PDF stream without object identifier"

data/lib/hexapdf/stream.rb

@@ -88,7 +88,9 @@ module HexaPDF
 
     # Returns a Fiber for getting at the data of the stream represented by this object.
     def fiber(chunk_size = 0)
-      if @source.kind_of?(Proc)
+      if @source.kind_of?(FiberDoubleForString)
+        @source.dup
+      elsif @source.kind_of?(Proc)
         FiberWithLength.new(@length, &@source)
       elsif @source.kind_of?(String)
         HexaPDF::Filter.source_from_file(@source, pos: @offset || 0, length: @length || -1,
@@ -134,7 +136,7 @@ module HexaPDF
   #
   # Note that support for external streams (/F, /FFilter, /FDecodeParms) is not yet implemented!
   #
-  # See: PDF1.7 s7.3.8, Dictionary
+  # See: PDF2.0 s7.3.8, Dictionary
   class Stream < Dictionary
 
     define_field :Length,       type: Integer # not required, will be auto-filled when writing
@@ -276,9 +278,8 @@ module HexaPDF
       end
     end
 
-    # :nodoc:
     # A mapping from short name to long name for filters.
-    FILTER_MAP = {AHx: :ASCIIHexDecode, A85: :ASCII85Decode, LZW: :LZWDecode,
+    FILTER_MAP = {AHx: :ASCIIHexDecode, A85: :ASCII85Decode, LZW: :LZWDecode, # :nodoc:
                   Fl: :FlateDecode, RL: :RunLengthDecode, CCF: :CCITTFaxDecode,
                   DCT: :DCTDecode}.freeze