hexapdf 0.46.0 → 1.6.0

data/lib/hexapdf/parser.rb

@@ -4,7 +4,7 @@
 # This file is part of HexaPDF.
 #
 # HexaPDF - A Versatile PDF Creation and Manipulation Library For Ruby
-# Copyright (C) 2014-2024 Thomas Leitner
+# Copyright (C) 2014-2025 Thomas Leitner
 #
 # HexaPDF is free software: you can redistribute it and/or modify it
 # under the terms of the GNU Affero General Public License version 3 as
@@ -112,11 +112,25 @@ module HexaPDF
         end
 
       if xref_entry.oid != 0 && (oid != xref_entry.oid || gen != xref_entry.gen)
-        raise_malformed("The oid,gen (#{oid},#{gen}) values of the indirect object don't match " \
-                        "the values (#{xref_entry.oid},#{xref_entry.gen}) from the xref")
+        msg = "The oid,gen (#{oid},#{gen}) values of the indirect object don't match " \
+              "the values (#{xref_entry.oid},#{xref_entry.gen}) from the xref"
+        # Some invalid PDFs contain entries where the generation number in the xref is different
+        # from the one found in the indirect object. If the file were reconstructed the generation
+        # number from the indirect object itself would be used.
+        # To gracefully handle such invalid PDFs they need to have a single revision.
+        # The other code part that handles this is in Revision#object.
+        if oid == xref_entry.oid && @document.revisions.count == 1
+          maybe_raise(msg, pos: xref_entry.pos)
+        else
+          raise_malformed(msg)
+        end
       end
 
-      @document.wrap(obj, oid: oid, gen: gen, stream: stream)
+      if obj.kind_of?(Reference)
+        @document.deref(obj)
+      else
+        @document.wrap(obj, oid: oid, gen: gen, stream: stream)
+      end
     rescue HexaPDF::MalformedPDFError
       reconstructed_revision.object(xref_entry) ||
         @document.wrap(nil, oid: xref_entry.oid, gen: xref_entry.gen)
@@ -184,7 +198,7 @@ module HexaPDF
         length = if object[:Length].kind_of?(Integer)
                    object[:Length]
                  elsif object[:Length].kind_of?(Reference)
-                   @document.deref(object[:Length]).value
+                   @document.deref(object[:Length])&.value || 0
                  else
                    0
                  end
@@ -205,9 +219,24 @@ module HexaPDF
         tok = @tokenizer.next_token
 
         object[:Length] = length
+        if object.key?(:Filter)
+          begin
+            object[:Filter] = @document.unwrap(object[:Filter])
+          rescue HexaPDF::Error
+            maybe_raise("Invalid /Filter entry for stream", pos: @tokenizer.pos)
+            object.delete(:Filter)
+          end
+        end
+        if object.key?(:DecodeParms)
+          begin
+            object[:DecodeParms] = @document.unwrap(object[:DecodeParms])
+          rescue HexaPDF::Error
+            maybe_raise("Invalid /DecodeParms entry for stream", pos: @tokenizer.pos)
+            object.delete(:DecodeParms)
+          end
+        end
         stream = StreamData.new(@tokenizer.io, offset: pos, length: length,
-                                filter: @document.unwrap(object[:Filter]),
-                                decode_parms: @document.unwrap(object[:DecodeParms]))
+                                filter: object[:Filter], decode_parms: object[:DecodeParms])
       end
 
       unless tok.kind_of?(Tokenizer::Token) && tok == 'endobj'

data/lib/hexapdf/pdf_array.rb

@@ -4,7 +4,7 @@
 # This file is part of HexaPDF.
 #
 # HexaPDF - A Versatile PDF Creation and Manipulation Library For Ruby
-# Copyright (C) 2014-2024 Thomas Leitner
+# Copyright (C) 2014-2025 Thomas Leitner
 #
 # HexaPDF is free software: you can redistribute it and/or modify it
 # under the terms of the GNU Affero General Public License version 3 as
@@ -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/rectangle.rb

@@ -4,7 +4,7 @@
 # This file is part of HexaPDF.
 #
 # HexaPDF - A Versatile PDF Creation and Manipulation Library For Ruby
-# Copyright (C) 2014-2024 Thomas Leitner
+# Copyright (C) 2014-2025 Thomas Leitner
 #
 # HexaPDF is free software: you can redistribute it and/or modify it
 # under the terms of the GNU Affero General Public License version 3 as

data/lib/hexapdf/reference.rb

@@ -4,7 +4,7 @@
 # This file is part of HexaPDF.
 #
 # HexaPDF - A Versatile PDF Creation and Manipulation Library For Ruby
-# Copyright (C) 2014-2024 Thomas Leitner
+# Copyright (C) 2014-2025 Thomas Leitner
 #
 # HexaPDF is free software: you can redistribute it and/or modify it
 # under the terms of the GNU Affero General Public License version 3 as
@@ -87,7 +87,7 @@ module HexaPDF
 
     # Computes the hash value based on the object and generation numbers.
     def hash
-      oid.hash ^ gen.hash
+      [oid, gen].hash
     end
 
     # Returns the object identifier as "oid,gen".

data/lib/hexapdf/revision.rb

@@ -4,7 +4,7 @@
 # This file is part of HexaPDF.
 #
 # HexaPDF - A Versatile PDF Creation and Manipulation Library For Ruby
-# Copyright (C) 2014-2024 Thomas Leitner
+# Copyright (C) 2014-2025 Thomas Leitner
 #
 # HexaPDF is free software: you can redistribute it and/or modify it
 # under the terms of the GNU Affero General Public License version 3 as
@@ -128,6 +128,11 @@ module HexaPDF
         @objects[oid, gen]
       elsif (xref_entry = @xref_section[oid, gen])
         load_object(xref_entry)
+      elsif (xref_entry = @xref_section[oid]) && (obj = load_object(xref_entry))&.gen == gen
+        # This branch handles invalid PDFs with a single revision containing xref entries where the
+        # gen doesn't match the gen of the indirect object. Also see the special handling in
+        # Parser#load_object.
+        obj
       else
         nil
       end
@@ -219,8 +224,7 @@ module HexaPDF
         seen = {}
         @objects.each {|oid, _gen, data| seen[oid] = true; yield(data) }
         @xref_section.each do |oid, _gen, data|
-          next if seen.key?(oid)
-          yield(@objects[oid] || load_object(data))
+          yield(@objects[oid] || load_object(data)) unless seen.key?(oid)
         end
         @all_objects_loaded = true
       end

data/lib/hexapdf/revisions.rb

@@ -4,7 +4,7 @@
 # This file is part of HexaPDF.
 #
 # HexaPDF - A Versatile PDF Creation and Manipulation Library For Ruby
-# Copyright (C) 2014-2024 Thomas Leitner
+# Copyright (C) 2014-2025 Thomas Leitner
 #
 # HexaPDF is free software: you can redistribute it and/or modify it
 # under the terms of the GNU Affero General Public License version 3 as

data/lib/hexapdf/serializer.rb

@@ -4,7 +4,7 @@
 # This file is part of HexaPDF.
 #
 # HexaPDF - A Versatile PDF Creation and Manipulation Library For Ruby
-# Copyright (C) 2014-2024 Thomas Leitner
+# Copyright (C) 2014-2025 Thomas Leitner
 #
 # HexaPDF is free software: you can redistribute it and/or modify it
 # under the terms of the GNU Affero General Public License version 3 as
@@ -276,16 +276,16 @@ module HexaPDF
     #
     # See: PDF2.0 s7.3.4
     def serialize_string(obj)
+      if obj.encoding != Encoding::BINARY && obj.match?(/[^ -~\t\r\n]/)
+        utf16_encoded = true
+        obj = "\xFE\xFF".b << obj.encode(Encoding::UTF_16BE).force_encoding(Encoding::BINARY)
+      end
       obj = if @encrypter && @object.kind_of?(HexaPDF::Object) && @object.indirect?
               encrypter.encrypt_string(obj, @object)
-            elsif obj.encoding != Encoding::BINARY
-              if obj.match?(/[^ -~\t\r\n]/)
-                "\xFE\xFF".b << obj.encode(Encoding::UTF_16BE).force_encoding(Encoding::BINARY)
-              else
-                obj.b
-              end
+            elsif utf16_encoded
+              obj
             else
-              obj.dup
+              obj.b
             end
       obj.gsub!(/[()\\\r]/n, STRING_ESCAPE_MAP)
       "(#{obj})"

data/lib/hexapdf/stream.rb

@@ -4,7 +4,7 @@
 # This file is part of HexaPDF.
 #
 # HexaPDF - A Versatile PDF Creation and Manipulation Library For Ruby
-# Copyright (C) 2014-2024 Thomas Leitner
+# Copyright (C) 2014-2025 Thomas Leitner
 #
 # HexaPDF is free software: you can redistribute it and/or modify it
 # under the terms of the GNU Affero General Public License version 3 as

data/lib/hexapdf/task/dereference.rb

@@ -4,7 +4,7 @@
 # This file is part of HexaPDF.
 #
 # HexaPDF - A Versatile PDF Creation and Manipulation Library For Ruby
-# Copyright (C) 2014-2024 Thomas Leitner
+# Copyright (C) 2014-2025 Thomas Leitner
 #
 # HexaPDF is free software: you can redistribute it and/or modify it
 # under the terms of the GNU Affero General Public License version 3 as

data/lib/hexapdf/task/merge_acro_form.rb

@@ -0,0 +1,164 @@
+# -*- encoding: utf-8; frozen_string_literal: true -*-
+#
+#--
+# This file is part of HexaPDF.
+#
+# HexaPDF - A Versatile PDF Creation and Manipulation Library For Ruby
+# Copyright (C) 2014-2025 Thomas Leitner
+#
+# HexaPDF is free software: you can redistribute it and/or modify it
+# under the terms of the GNU Affero General Public License version 3 as
+# published by the Free Software Foundation with the addition of the
+# following permission added to Section 15 as permitted in Section 7(a):
+# FOR ANY PART OF THE COVERED WORK IN WHICH THE COPYRIGHT IS OWNED BY
+# THOMAS LEITNER, THOMAS LEITNER DISCLAIMS THE WARRANTY OF NON
+# INFRINGEMENT OF THIRD PARTY RIGHTS.
+#
+# HexaPDF is distributed in the hope that it will be useful, but WITHOUT
+# ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+# FITNESS FOR A PARTICULAR PURPOSE. See the GNU Affero General Public
+# License for more details.
+#
+# You should have received a copy of the GNU Affero General Public License
+# along with HexaPDF. If not, see <http://www.gnu.org/licenses/>.
+#
+# The interactive user interfaces in modified source and object code
+# versions of HexaPDF must display Appropriate Legal Notices, as required
+# under Section 5 of the GNU Affero General Public License version 3.
+#
+# In accordance with Section 7(b) of the GNU Affero General Public
+# License, a covered work must retain the producer line in every PDF that
+# is created or manipulated using HexaPDF.
+#
+# If the GNU Affero General Public License doesn't fit your need,
+# commercial licenses are available at <https://gettalong.at/hexapdf/>.
+#++
+
+require 'hexapdf/serializer'
+
+module HexaPDF
+  module Task
+
+    # Task for merging an AcroForm from one PDF into another.
+    #
+    # It takes care of
+    #
+    # * adding the fields to the main Type::AcroForm::Form dictionary,
+    # * adjusting the field names so that they are unique,
+    # * and merging the properties of the main AcroForm dictionary itself and adjusting field
+    #   information appropriately.
+    #
+    # Note that the pages with the fields need to be imported already.
+    #
+    # The steps for using this task are:
+    #
+    # 1. Import the pages into the target document and add all imported pages to an array
+    # 2. Call this task using the created array of pages.
+    #
+    # Example:
+    #
+    #   pages = doc.pages.map {|page| target.pages.add(target.import(page)) }
+    #   target.task(:merge_acro_form, source: doc, pages: pages)
+    module MergeAcroForm
+
+      # Performs the necessary steps to merge the AcroForm fields from the +source+ into the target
+      # document +doc+.
+      #
+      # +source+::
+      #     Specifies the source PDF document the information from which should be merged into the
+      #     target document.
+      #
+      # +pages+::
+      #     An array of pages that were imported from +source+ and contain the widgets of the fields
+      #     that should be merged.
+      def self.call(doc, source:, pages:)
+        return unless source.acro_form
+
+        acro_form = doc.acro_form(create: true)
+
+        # Determine a unique name for root field and create root field
+        import_name = 'merged_' +
+                      (acro_form.root_fields.select {|field| field[:T] =~ /\Amerged_\d+\z/ }.
+                        map {|field| field[:T][/\d+/].to_i }.sort.last || 0).succ.to_s
+        root_field = doc.add({T: import_name, Kids: []})
+        acro_form.root_fields << root_field
+
+        # Merge the main AcroForm dictionary
+        font_name_mapping = merge_form_dictionary(acro_form, source.acro_form, root_field)
+        font_name_re = font_name_mapping.keys.map {|name| Regexp.escape(name) }.join('|')
+        root_field[:DA] && root_field[:DA].sub!(font_name_re, font_name_mapping)
+
+        # Process all field widgets of the given pages
+        process_calculate_actions = false
+        signature_field_seen = false
+        pages.each do |page|
+          page.each_annotation do |widget|
+            next unless widget[:Subtype] == :Widget
+            field = widget.form_field
+
+            # Correct the font name in the default appearance string
+            widget[:DA] && widget[:DA].sub!(font_name_re, font_name_mapping)
+            field[:DA] && field[:DA].sub!(font_name_re, font_name_mapping)
+
+            process_calculate_actions = true if field[:AA]&.[](:C)
+            signature_field_seen = true if field.field_type == :Sig
+
+            # Add to the root field
+            field = field[:Parent] while field[:Parent]
+            if field != root_field
+              field[:Parent] = root_field
+              root_field[:Kids] << field
+            end
+          end
+        end
+
+        # Update calculation JavaScript actions with changed field names
+        fix_calculate_actions(acro_form, source.acro_form, import_name) if process_calculate_actions
+
+        # Update signature flags if necessary
+        if signature_field_seen && source.acro_form.signature_flag?(:signatures_exist)
+          acro_form.signature_flag(:signatures_exist)
+        end
+      end
+
+      # Merges the AcroForm +source_form+ into the +target_form+ and returns a mapping of old font
+      # names to new ones.
+      def self.merge_form_dictionary(target_form, source_form, root_field)
+        target_resources = target_form.default_resources
+        font_name_mapping = {}
+        serializer = HexaPDF::Serializer.new
+
+        source_form.default_resources[:Font].each do |font_name, value|
+          new_name = target_resources.add_font(target_form.document.import(value))
+          font_name_mapping[serializer.serialize(font_name)] = serializer.serialize(new_name)
+        end
+
+        root_field[:DA] = target_form.document.import(source_form[:DA])
+        root_field[:Q] = target_form.document.import(source_form[:Q])
+
+        font_name_mapping
+      end
+
+      # Fixes the calculate actions listed in the /CO entry of the main AcroForm dictionary to use
+      # the new names of the fields.
+      def self.fix_calculate_actions(acro_form, source_form, import_name)
+        if source_form[:CO]
+          acro_form[:CO] ||= []
+          acro_form[:CO].value.concat(acro_form.document.import(source_form[:CO]).value)
+          acro_form[:CO].each do |field|
+            next unless (action = field[:AA]&.[](:C))
+            action[:JS].gsub!(/"(.*?)"/) do |match|
+              if source_form.field_by_name($1)
+                "\"#{import_name}.#{$1}\""
+              else
+                match
+              end
+            end
+          end
+        end
+      end
+
+    end
+
+  end
+end

data/lib/hexapdf/task/optimize.rb

@@ -4,7 +4,7 @@
 # This file is part of HexaPDF.
 #
 # HexaPDF - A Versatile PDF Creation and Manipulation Library For Ruby
-# Copyright (C) 2014-2024 Thomas Leitner
+# Copyright (C) 2014-2025 Thomas Leitner
 #
 # HexaPDF is free software: you can redistribute it and/or modify it
 # under the terms of the GNU Affero General Public License version 3 as
@@ -214,13 +214,13 @@ module HexaPDF
         end
       end
 
-      # Deletes field entries of the object that are optional and currently set to their default
-      # value.
+      # Deletes field entries (except for /Type) of the object that are optional and currently set
+      # to their default value.
       def self.delete_fields_with_defaults(obj)
         return unless obj.kind_of?(HexaPDF::Dictionary) && !obj.null?
         obj.each do |name, value|
-          if (field = obj.class.field(name)) && !field.required? && field.default? &&
-              value == field.default
+          if name != :Type && (field = obj.class.field(name)) && !field.required? &&
+             field.default? && value == field.default
             obj.delete(name)
           end
         end

data/lib/hexapdf/task/pdfa.rb

@@ -4,7 +4,7 @@
 # This file is part of HexaPDF.
 #
 # HexaPDF - A Versatile PDF Creation and Manipulation Library For Ruby
-# Copyright (C) 2014-2024 Thomas Leitner
+# Copyright (C) 2014-2025 Thomas Leitner
 #
 # HexaPDF is free software: you can redistribute it and/or modify it
 # under the terms of the GNU Affero General Public License version 3 as

data/lib/hexapdf/task.rb

@@ -4,7 +4,7 @@
 # This file is part of HexaPDF.
 #
 # HexaPDF - A Versatile PDF Creation and Manipulation Library For Ruby
-# Copyright (C) 2014-2024 Thomas Leitner
+# Copyright (C) 2014-2025 Thomas Leitner
 #
 # HexaPDF is free software: you can redistribute it and/or modify it
 # under the terms of the GNU Affero General Public License version 3 as
@@ -65,6 +65,7 @@ module HexaPDF
     autoload(:Optimize, 'hexapdf/task/optimize')
     autoload(:Dereference, 'hexapdf/task/dereference')
     autoload(:PDFA, 'hexapdf/task/pdfa')
+    autoload(:MergeAcroForm, 'hexapdf/task/merge_acro_form')
 
   end
 

data/lib/hexapdf/test_utils.rb

@@ -4,7 +4,7 @@
 # This file is part of HexaPDF.
 #
 # HexaPDF - A Versatile PDF Creation and Manipulation Library For Ruby
-# Copyright (C) 2014-2024 Thomas Leitner
+# Copyright (C) 2014-2025 Thomas Leitner
 #
 # HexaPDF is free software: you can redistribute it and/or modify it
 # under the terms of the GNU Affero General Public License version 3 as
@@ -92,8 +92,9 @@ module HexaPDF
     # Creates a fiber that yields the given string in +len+ length parts.
     def feeder(string, len = string.length)
       Fiber.new do
+        string = string.b
         until string.empty?
-          Fiber.yield(string.slice!(0, len).force_encoding('BINARY'))
+          Fiber.yield(string.slice!(0, len))
         end
       end
     end

data/lib/hexapdf/tokenizer.rb

@@ -4,7 +4,7 @@
 # This file is part of HexaPDF.
 #
 # HexaPDF - A Versatile PDF Creation and Manipulation Library For Ruby
-# Copyright (C) 2014-2024 Thomas Leitner
+# Copyright (C) 2014-2025 Thomas Leitner
 #
 # HexaPDF is free software: you can redistribute it and/or modify it
 # under the terms of the GNU Affero General Public License version 3 as
@@ -118,42 +118,43 @@ module HexaPDF
     def next_token
       prepare_string_scanner(20)
       prepare_string_scanner(20) while @ss.skip(WHITESPACE_MULTI_RE)
-      byte = @ss.string.getbyte(@ss.pos) || -1
-      if (48 <= byte && byte <= 57) || byte == 45 || byte == 43 || byte == 46 # 0..9 - + .
+      case (byte = @ss.scan_byte || -1)
+      when 43, 45, 46, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57 # + - . 0..9
+        @ss.pos -= 1
         parse_number
-      elsif byte == 47 # /
+      when 47 # /
         parse_name
-      elsif byte == 40 # (
+      when 40 # (
         parse_literal_string
-      elsif byte == 60 # <
-        if @ss.string.getbyte(@ss.pos + 1) == 60
-          @ss.pos += 2
+      when 60 # <
+        if @ss.peek_byte == 60
+          @ss.pos += 1
           TOKEN_DICT_START
         else
           parse_hex_string
         end
-      elsif byte == 62 # >
-        unless @ss.string.getbyte(@ss.pos + 1) == 62
-          raise HexaPDF::MalformedPDFError.new("Delimiter '>' found at invalid position", pos: pos)
+      when 62 # >
+        unless @ss.scan_byte == 62
+          raise HexaPDF::MalformedPDFError.new("Delimiter '>' found at invalid position", pos: pos - 1)
         end
-        @ss.pos += 2
         TOKEN_DICT_END
-      elsif byte == 91 # [
-        @ss.pos += 1
+      when 91 # [
         TOKEN_ARRAY_START
-      elsif byte == 93 # ]
-        @ss.pos += 1
+      when 93 # ]
         TOKEN_ARRAY_END
-      elsif byte == 123 || byte == 125 # { }
-        Token.new(@ss.get_byte)
-      elsif byte == 37 # %
+      when 41 # )
+        raise HexaPDF::MalformedPDFError.new("Delimiter ')' found at invalid position", pos: pos - 1)
+      when 123, 125 # { }
+        Token.new(byte.chr.b)
+      when 37 # %
         until @ss.skip_until(/(?=[\r\n])/)
           return NO_MORE_TOKENS unless prepare_string_scanner
         end
         next_token
-      elsif byte == -1 # we reached the end of the file
+      when -1 # we reached the end of the file
         NO_MORE_TOKENS
       else # everything else consisting of regular characters
+        @ss.pos -= 1
         parse_keyword
       end
     end
@@ -207,12 +208,13 @@ module HexaPDF
     # Note: This is a special method meant for use with reconstructing the cross-reference table!
     def next_integer_or_keyword
       skip_whitespace
-      byte = @ss.string.getbyte(@ss.pos) || -1
-      if 48 <= byte && byte <= 57
+      byte = @ss.peek_byte || -1
+      case byte
+      when 48, 49, 50, 51, 52, 53, 54, 55, 56, 57
         parse_number
-      elsif (97 <= byte && byte <= 122) || (65 <= byte && byte <= 90)
+      when 97..122, 65..90
         parse_keyword
-      elsif byte == -1 # we reached the end of the file
+      when -1 # we reached the end of the file
         NO_MORE_TOKENS
       else
         nil
@@ -222,8 +224,7 @@ module HexaPDF
     # Reads the byte (an integer) at the current position and advances the scan pointer.
     def next_byte
       prepare_string_scanner(1)
-      @ss.pos += 1
-      @ss.string.getbyte(@ss.pos - 1)
+      @ss.scan_byte
     end
 
     # Reads the cross-reference subsection entry at the current position and advances the scan
@@ -277,25 +278,35 @@ 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
     def parse_number
-      val = scan_until(WHITESPACE_OR_DELIMITER_RE) || @ss.scan(/.*/)
-      if val.match?(/\A[+-]?\d++(?!\.)\z/)
-        tmp = val.to_i
-        # Handle object references, see PDF2.0 s7.3.10
-        prepare_string_scanner(10)
-        if @ss.scan(REFERENCE_RE)
-          tmp = if tmp > 0
-                  Reference.new(tmp, @ss[1].to_i)
-                else
-                  maybe_raise("Invalid indirect object reference (#{tmp},#{@ss[1].to_i})")
-                  nil
-                end
+      prepare_string_scanner(40)
+      pos = self.pos
+      if (tmp = @ss.scan_integer)
+        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)
+            tmp = if tmp > 0
+                    Reference.new(tmp, @ss[1].to_i)
+                  else
+                    maybe_raise("Invalid indirect object reference (#{tmp},#{@ss[1].to_i})")
+                    nil
+                  end
+          end
+          return tmp
+        else
+          self.pos = pos
         end
-        tmp
-      elsif val.match?(/\A[+-]?(?:\d+\.\d*|\.\d+)\z/)
+      end
+
+      val = scan_until(WHITESPACE_OR_DELIMITER_RE) || @ss.scan(/.*/)
+      if val.match?(/\A[+-]?(?:\d+\.\d*|\.\d+)\z/)
         val << '0' if val.getbyte(-1) == 46 # dot '.'
         Float(val)
       else
@@ -318,7 +329,6 @@ module HexaPDF
     #
     # See: PDF2.0 s7.3.4.2
     def parse_literal_string
-      @ss.pos += 1
       str = "".b
       parentheses = 1
 
@@ -335,7 +345,7 @@ module HexaPDF
         when 40 then parentheses += 1 # (
         when 13 # \r
           str[-1] = "\n"
-          @ss.pos += 1 if @ss.peek(1) == "\n"
+          @ss.pos += 1 if @ss.peek_byte == 10 # \n
         when 92 # \\
           str.chop!
           prepare_string_scanner(3)
@@ -361,7 +371,6 @@ module HexaPDF
     #
     # See: PDF2.0 s7.3.4.3
     def parse_hex_string
-      @ss.pos += 1
       data = scan_until(/(?=>)/)
       unless data
         raise HexaPDF::MalformedPDFError.new("Unclosed hex string found", pos: pos)
@@ -376,7 +385,6 @@ module HexaPDF
     #
     # See: PDF2.0 s7.3.5
     def parse_name
-      @ss.pos += 1
       str = scan_until(WHITESPACE_OR_DELIMITER_RE) || @ss.scan(/.*/)
       str.gsub!(/#[A-Fa-f0-9]{2}/) {|m| m[1, 2].hex.chr }
       if str.force_encoding(Encoding::UTF_8).valid_encoding?