hexapdf 0.46.0 → 1.6.0

data/lib/hexapdf/font/cmap/writer.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
@@ -40,9 +40,7 @@ module HexaPDF
   module Font
     class CMap
 
-      # Creates a CMap file.
-      #
-      # Currently only ToUnicode CMaps are supported.
+      # Creates a CMap file, either a ToUnicode CMap or a CID CMap.
       class Writer
 
         # Maximum number of entries in one section.
@@ -74,6 +72,28 @@ module HexaPDF
           to_unicode_template % result.chop!
         end
 
+        # Returns a CID CMap for the given input code to CID mapping which needs to be sorted by
+        # input codes.
+        #
+        # Note that the returned CMap always uses a 16-bit input code space!
+        def create_cid_cmap(mapping)
+          return cid_template % '' if mapping.empty?
+
+          chars, ranges = compute_section_entries(mapping)
+
+          result = create_sections("cidchar", chars.size / 2) do |index|
+            index *= 2
+            sprintf("<%04X>", chars[index]) << " #{chars[index + 1]}\n"
+          end
+
+          result << create_sections("cidrange", ranges.size / 3) do |index|
+            index *= 3
+            sprintf("<%04X><%04X>", ranges[index], ranges[index + 1]) << " #{ranges[index + 2]}\n"
+          end
+
+          cid_template % result.chop!
+        end
+
         private
 
         # Computes the entries for the "char" and "range" sections based on the given mapping.
@@ -146,7 +166,7 @@ module HexaPDF
           result
         end
 
-        # Returns the CMap file template for a ToUnicode CMap.
+        # Returns the template for a ToUnicode CMap.
         def to_unicode_template
           <<~TEMPLATE
             /CIDInit /ProcSet findresource begin
@@ -170,6 +190,40 @@ module HexaPDF
           TEMPLATE
         end
 
+        # Returns the template for a CID CMap.
+        def cid_template
+          <<~TEMPLATE
+            %%!PS-Adobe-3.0 Resource-CMap
+            %%%%DocumentNeededResources: ProcSet (CIDInit)
+            %%%%IncludeResource: ProcSet (CIDInit)
+            %%%%BeginResource: CMap (Custom)
+            %%%%Title: (Custom Adobe Identity 0)
+            %%%%Version: 1
+            /CIDInit /ProcSet findresource begin
+            12 dict begin
+            begincmap
+            /CIDSystemInfo 3 dict dup begin
+              /Registry (Adobe) def
+              /Ordering (Identity) def
+              /Supplement 0 def
+            end def
+            /CMapName /Custom def
+            /CMapType 1 def
+            /CMapVersion 1 def
+            /WMode 0 def
+            1 begincodespacerange
+            <0000> <FFFF>
+            endcodespacerange
+            %s
+            endcmap
+            CMapName currentdict /CMap defineresource pop
+            end
+            end
+            %%%%EndResource
+            %%%%EOF
+          TEMPLATE
+        end
+
       end
 
     end

data/lib/hexapdf/font/cmap.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
@@ -85,6 +85,13 @@ module HexaPDF
         Writer.new.create_to_unicode_cmap(mapping)
       end
 
+      # Returns a string containing a CID CMap that represents the given code to CID mapping.
+      #
+      # See: Writer#create_cid_cmap
+      def self.create_cid_cmap(mapping)
+        Writer.new.create_cid_cmap(mapping)
+      end
+
       # The registry part of the CMap version.
       attr_accessor :registry
 
@@ -136,10 +143,13 @@ module HexaPDF
       # An error is raised if the string contains invalid bytes.
       def read_codes(string)
         codes = []
-        bytes = string.each_byte
+        bytes = string.bytes
+        length = bytes.length
+        i = 0
 
-        loop do
-          byte = bytes.next
+        while i < length
+          byte = bytes[i]
+          i += 1
           code = 0
 
           found = @codespace_ranges.any? do |first_byte_range, rest_ranges|
@@ -147,9 +157,10 @@ module HexaPDF
 
             code = (code << 8) + byte
             valid = rest_ranges.all? do |range|
-              begin
-                byte = bytes.next
-              rescue StopIteration
+              if i < length
+                byte = bytes[i]
+                i += 1
+              else
                 raise HexaPDF::Error, "Missing bytes while reading codes via CMap"
               end
               code = (code << 8) + byte

data/lib/hexapdf/font/encoding/base.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
@@ -81,6 +81,33 @@ module HexaPDF
           @code_to_name.key(name)
         end
 
+        # Returns the encoding in a compact array form.
+        #
+        # If the optional +base_encoding+ argument is specified, all codes that have the same value
+        # in the base encoding are ignored.
+        #
+        # The returned array is of the form:
+        #
+        #   code1 name1 name2 ... code2 name3 name4 ...
+        #
+        # This means that name1 is associated with code1, name2 with code1 + 1 and so on.
+        #
+        # See: PDF 2.0 s9.6.5.1
+        def to_compact_array(base_encoding: nil)
+          result = []
+          last_code = -3
+          @code_to_name.sort.each do |code, name|
+            next if base_encoding&.name(code) == name
+            if last_code + 1 == code
+              result << name
+            else
+              result << code << name
+            end
+            last_code = code
+          end
+          result
+        end
+
       end
 
     end

data/lib/hexapdf/font/encoding/difference_encoding.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/font/encoding/glyph_list.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/font/encoding/mac_expert_encoding.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/font/encoding/mac_roman_encoding.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/font/encoding/standard_encoding.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/font/encoding/symbol_encoding.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/font/encoding/win_ansi_encoding.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/font/encoding/zapf_dingbats_encoding.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/font/encoding.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/font/invalid_glyph.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/font/true_type/builder.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/font/true_type/font.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/font/true_type/optimizer.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/font/true_type/subsetter.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/font/true_type/table/cmap.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/font/true_type/table/cmap_subtable.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/font/true_type/table/directory.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/font/true_type/table/glyf.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/font/true_type/table/head.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/font/true_type/table/hhea.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/font/true_type/table/hmtx.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/font/true_type/table/kern.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/font/true_type/table/loca.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/font/true_type/table/maxp.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/font/true_type/table/name.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/font/true_type/table/os2.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/font/true_type/table/post.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/font/true_type/table.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
@@ -63,7 +63,12 @@ module HexaPDF
 
         # Calculates the checksum for the given data.
         def self.calculate_checksum(data)
-          data.unpack('N*').inject(0) {|sum, long| sum + long } % 2**32
+          checksum = 0
+          if (remainder_length = data.length % 4) != 0
+            checksum = (data[-remainder_length, remainder_length] << "\0" * (4 - remainder_length)).
+              unpack1('N')
+          end
+          checksum + data.unpack('N*').inject(0) {|sum, long| sum + long } % 2**32
         end
 
         # The TrueType font object associated with this table.

data/lib/hexapdf/font/true_type.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/font/true_type_wrapper.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
@@ -57,6 +57,10 @@ module HexaPDF
     class TrueTypeWrapper
 
       # Represents a single glyph of the wrapped font.
+      #
+      # Since some characters/strings may be mapped to the same glyph id by the font's builtin cmap
+      # table, it is possible that different Glyph instances with the same #id but different #str
+      # exist.
       class Glyph
 
         # The associated TrueTypeWrapper object.
@@ -152,6 +156,7 @@ module HexaPDF
         @id_to_glyph = {}
         @codepoint_to_glyph = {}
         @encoded_glyphs = {}
+        @last_char_code = 0
       end
 
       # Returns the type of the font, i.e. :TrueType.
@@ -179,14 +184,15 @@ module HexaPDF
         !@subsetter.nil?
       end
 
-      # Returns a Glyph object for the given glyph ID.
+      # Returns a Glyph object for the given glyph ID and +str+ pair.
       #
-      # The optional argument +str+ should be the string representation of the glyph. Only use it if
-      # it is known,
+      # The optional argument +str+ should be the string representation of the glyph. It is possible
+      # that multiple strings map to the same glyph (e.g. hyphen and soft-hyphen could be
+      # represented by the same glyph).
       #
       # Note: Although this method is public, it should normally not be used by application code!
       def glyph(id, str = nil)
-        @id_to_glyph[id] ||=
+        @id_to_glyph[[id, str]] ||=
           if id >= 0 && id < @wrapped_font[:maxp].num_glyphs
             Glyph.new(self, id, str || (+'' << (@cmap.gid_to_code(id) || 0xFFFD)))
           else
@@ -228,14 +234,17 @@ module HexaPDF
 
       # Encodes the glyph and returns the code string.
       def encode(glyph)
-        (@encoded_glyphs[glyph.id] ||=
+        (@encoded_glyphs[glyph] ||=
           begin
             raise HexaPDF::MissingGlyphError.new(glyph) if glyph.kind_of?(InvalidGlyph)
-            if @subsetter
-              [[@subsetter.use_glyph(glyph.id)].pack('n'), glyph]
-            else
-              [[glyph.id].pack('n'), glyph]
+            @subsetter.use_glyph(glyph.id) if @subsetter
+            @last_char_code += 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_char_code == 13 || @last_char_code == 40 || @last_char_code == 92
+              @last_char_code += (@last_char_code == 40 ? 2 : 1)
             end
+            [[@last_char_code].pack('n'), @last_char_code]
           end)[0]
       end
 
@@ -286,14 +295,16 @@ module HexaPDF
                                                  Supplement: 0},
                                  CIDToGIDMap: :Identity})
         dict = document.add({Type: :Font, Subtype: :Type0, BaseFont: cid_font[:BaseFont],
-                             Encoding: :'Identity-H', DescendantFonts: [cid_font]})
+                             DescendantFonts: [cid_font]})
         dict.font_wrapper = self
 
         document.register_listener(:complete_objects) do
+          next if dict.null?
           update_font_name(dict)
           embed_font(dict, document)
           complete_width_information(dict)
           create_to_unicode_cmap(dict, document)
+          add_encoding_information_cmap(dict, document)
         end
 
         dict
@@ -306,7 +317,7 @@ module HexaPDF
         return unless @subsetter
 
         tag = +''
-        data = @encoded_glyphs.each_with_object(''.b) {|(id, v), s| s << id.to_s << v[0] }
+        data = @encoded_glyphs.each_with_object(''.b) {|(g, v), s| s << g.id.to_s << v[0] }
         hash = Digest::MD5.hexdigest(data << @wrapped_font.font_name).to_i(16)
         while hash != 0 && tag.length < 6
           hash, mod = hash.divmod(UPPERCASE_LETTERS.length)
@@ -336,8 +347,8 @@ module HexaPDF
       # Adds the /DW and /W fields to the CIDFont dictionary.
       def complete_width_information(dict)
         default_width = glyph(3, " ").width.to_i
-        widths = @encoded_glyphs.reject {|_, v| v[1].width == default_width }.map do |id, v|
-          [(@subsetter ? @subsetter.subset_glyph_id(id) : id), v[1].width]
+        widths = @encoded_glyphs.reject {|g, _| g.width == default_width }.map do |g, _|
+          [(@subsetter ? @subsetter.subset_glyph_id(g.id) : g.id), g.width]
         end.sort!
         dict[:DescendantFonts].first.set_widths(widths, default_width: default_width)
       end
@@ -346,9 +357,10 @@ module HexaPDF
       # correctly.
       def create_to_unicode_cmap(dict, document)
         stream = HexaPDF::StreamData.new do
-          mapping = @encoded_glyphs.keys.map! do |id|
+          mapping = @encoded_glyphs.map do |glyph, (_, char_code)|
             # Using 0xFFFD as mentioned in Adobe #5411, last line before section 1.5
-            [(@subsetter ? @subsetter.subset_glyph_id(id) : id), @cmap.gid_to_code(id) || 0xFFFD]
+            # TODO: glyph.str assumed to consist of single char, No support for multiple chars
+            [char_code, glyph.str.ord || 0xFFFD]
           end.sort_by!(&:first)
           HexaPDF::Font::CMap.create_to_unicode_cmap(mapping)
         end
@@ -357,6 +369,29 @@ module HexaPDF
         dict[:ToUnicode] = stream_obj
       end
 
+      # Adds the /Encoding entry to the +dict+.
+      #
+      # This can either be the identity mapping or, if some Unicode codepoints are mapped to the
+      # same glyph, a custom CMap.
+      def add_encoding_information_cmap(dict, document)
+        mapping = @encoded_glyphs.map do |glyph, (_, char_code)|
+          # Using 0xFFFD as mentioned in Adobe #5411, last line before section 1.5
+          [char_code, (@subsetter ? @subsetter.subset_glyph_id(glyph.id) : glyph.id)]
+        end.sort_by!(&:first)
+        if mapping.all? {|char_code, cid| char_code == cid }
+          dict[:Encoding] = :'Identity-H'
+        else
+          stream = HexaPDF::StreamData.new { HexaPDF::Font::CMap.create_cid_cmap(mapping) }
+          stream_obj = document.add({Type: :CMap,
+                                     CMapName: :Custom,
+                                     CIDSystemInfo: {Registry: "Adobe", Ordering: "Identity",
+                                                     Supplement: 0},
+                                    }, stream: stream)
+          stream_obj.set_filter(:FlateDecode)
+          dict[:Encoding] = stream_obj
+        end
+      end
+
     end
 
   end