hexapdf 0.2.0 → 0.3.0

data/lib/hexapdf/object.rb

@@ -4,7 +4,7 @@
 # This file is part of HexaPDF.
 #
 # HexaPDF - A Versatile PDF Creation and Manipulation Library For Ruby
-# Copyright (C) 2016 Thomas Leitner
+# Copyright (C) 2014-2017 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/parser.rb

@@ -4,7 +4,7 @@
 # This file is part of HexaPDF.
 #
 # HexaPDF - A Versatile PDF Creation and Manipulation Library For Ruby
-# Copyright (C) 2016 Thomas Leitner
+# Copyright (C) 2014-2017 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/rectangle.rb

@@ -4,7 +4,7 @@
 # This file is part of HexaPDF.
 #
 # HexaPDF - A Versatile PDF Creation and Manipulation Library For Ruby
-# Copyright (C) 2016 Thomas Leitner
+# Copyright (C) 2014-2017 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) 2016 Thomas Leitner
+# Copyright (C) 2014-2017 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/revision.rb

@@ -4,7 +4,7 @@
 # This file is part of HexaPDF.
 #
 # HexaPDF - A Versatile PDF Creation and Manipulation Library For Ruby
-# Copyright (C) 2016 Thomas Leitner
+# Copyright (C) 2014-2017 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/revisions.rb

@@ -4,7 +4,7 @@
 # This file is part of HexaPDF.
 #
 # HexaPDF - A Versatile PDF Creation and Manipulation Library For Ruby
-# Copyright (C) 2016 Thomas Leitner
+# Copyright (C) 2014-2017 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
@@ -62,24 +62,22 @@ module HexaPDF
 
         parser = Parser.new(io, document)
         object_loader = lambda {|xref_entry| parser.load_object(xref_entry)}
-        revision_loader = lambda do |offset|
-          xref_section, trailer = parser.load_revision(offset)
-          Revision.new(document.wrap(trailer, type: :XXTrailer), xref_section: xref_section,
-                       loader: object_loader)
-        end
 
-        revisions = [revision_loader.call(parser.startxref_offset)]
+        revisions = []
+        xref_section, trailer = parser.load_revision(parser.startxref_offset)
+        revisions << Revision.new(document.wrap(trailer, type: :XXTrailer),
+                                  xref_section: xref_section, loader: object_loader)
+
 
-        i = revisions.length - 1
-        while i >= 0
+        while (prev = revisions[0].trailer.value[:Prev])
           # PDF1.7 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.
-          xrefstm = revisions[i].trailer.value[:XRefStm]
-          prev = revisions[i].trailer.value[:Prev]
-          new_revisions = [(revision_loader.call(prev) if prev),
-                           (revision_loader.call(xrefstm) if xrefstm)].compact
-          revisions.insert(i, *new_revisions)
-          i += new_revisions.length - 1
+          xref_section, trailer = parser.load_revision(prev)
+          stm = revisions[0].trailer.value[:XRefStm]
+          stm_xref_section, = parser.load_revision(stm) if stm
+          xref_section.merge!(stm_xref_section) if stm
+          revisions.unshift(Revision.new(document.wrap(trailer, type: :XXTrailer),
+                                         xref_section: xref_section, loader: object_loader))
         end
 
         document.version = parser.file_header_version

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) 2016 Thomas Leitner
+# Copyright (C) 2014-2017 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
@@ -96,6 +96,7 @@ module HexaPDF
       @encrypter = false
       @io = nil
       @object = nil
+      @in_object = false
     end
 
     # Returns the serialized form of the given object.
@@ -267,10 +268,13 @@ module HexaPDF
     # Uses #serialize_hexapdf_reference if it is an indirect object, otherwise just serializes
     # the objects value.
     def serialize_hexapdf_object(obj)
-      if obj.indirect? && obj != @object
+      if obj.indirect? && (obj != @object || @in_object)
         serialize_hexapdf_reference(obj)
       else
-        __serialize(obj.value)
+        @in_object ||= (obj == @object)
+        str = __serialize(obj.value)
+        @in_object = false if obj == @object
+        str
       end
     end
 

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) 2016 Thomas Leitner
+# Copyright (C) 2014-2017 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) 2016 Thomas Leitner
+# Copyright (C) 2014-2017 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) 2016 Thomas Leitner
+# Copyright (C) 2014-2017 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/optimize.rb

@@ -4,7 +4,7 @@
 # This file is part of HexaPDF.
 #
 # HexaPDF - A Versatile PDF Creation and Manipulation Library For Ruby
-# Copyright (C) 2016 Thomas Leitner
+# Copyright (C) 2014-2017 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/tokenizer.rb

@@ -4,7 +4,7 @@
 # This file is part of HexaPDF.
 #
 # HexaPDF - A Versatile PDF Creation and Manipulation Library For Ruby
-# Copyright (C) 2016 Thomas Leitner
+# Copyright (C) 2014-2017 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
@@ -108,40 +108,40 @@ module HexaPDF
     def next_token
       prepare_string_scanner(20)
       prepare_string_scanner(20) while @ss.skip(WHITESPACE_MULTI_RE)
-      case (@ss.eos? ? -1 : @ss.string.getbyte(@ss.pos))
-      when 43, 45, 46, 48..57 # + - . 0..9
+      byte = @ss.string.getbyte(@ss.pos) || -1
+      if (48 <= byte && byte <= 57) || byte == 45 || byte == 43 || byte == 46 # 0..9 - + .
         parse_number
-      when 47 # /
+      elsif byte == 47 # /
         parse_name
-      when 40 # (
+      elsif byte == 40 # (
         parse_literal_string
-      when 60 # <
+      elsif byte == 60 # <
         if @ss.string.getbyte(@ss.pos + 1) != 60
           parse_hex_string
         else
           @ss.pos += 2
           TOKEN_DICT_START
         end
-      when 62 # >
+      elsif byte == 62 # >
         unless @ss.string.getbyte(@ss.pos + 1) == 62
           raise HexaPDF::MalformedPDFError.new("Delimiter '>' found at invalid position", pos: pos)
         end
         @ss.pos += 2
         TOKEN_DICT_END
-      when 91 # [
+      elsif byte == 91 # [
         @ss.pos += 1
         TOKEN_ARRAY_START
-      when 93 # ]
+      elsif byte == 93 # ]
         @ss.pos += 1
         TOKEN_ARRAY_END
-      when 123, 125 # { }
+      elsif byte == 123 || byte == 125 # { }
         Token.new(@ss.get_byte)
-      when 37 # %
+      elsif byte == 37 # %
         until @ss.skip_until(/(?=[\r\n])/)
           return NO_MORE_TOKENS unless prepare_string_scanner
         end
         next_token
-      when -1 # we reached the end of the file
+      elsif byte == -1 # we reached the end of the file
         NO_MORE_TOKENS
       else # everything else consisting of regular characters
         parse_keyword

data/lib/hexapdf/type.rb

@@ -4,7 +4,7 @@
 # This file is part of HexaPDF.
 #
 # HexaPDF - A Versatile PDF Creation and Manipulation Library For Ruby
-# Copyright (C) 2016 Thomas Leitner
+# Copyright (C) 2014-2017 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/type/catalog.rb

@@ -4,7 +4,7 @@
 # This file is part of HexaPDF.
 #
 # HexaPDF - A Versatile PDF Creation and Manipulation Library For Ruby
-# Copyright (C) 2016 Thomas Leitner
+# Copyright (C) 2014-2017 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/type/embedded_file.rb

@@ -4,7 +4,7 @@
 # This file is part of HexaPDF.
 #
 # HexaPDF - A Versatile PDF Creation and Manipulation Library For Ruby
-# Copyright (C) 2016 Thomas Leitner
+# Copyright (C) 2014-2017 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/type/file_specification.rb

@@ -4,7 +4,7 @@
 # This file is part of HexaPDF.
 #
 # HexaPDF - A Versatile PDF Creation and Manipulation Library For Ruby
-# Copyright (C) 2016 Thomas Leitner
+# Copyright (C) 2014-2017 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/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) 2016 Thomas Leitner
+# Copyright (C) 2014-2017 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/type/font_descriptor.rb

@@ -4,7 +4,7 @@
 # This file is part of HexaPDF.
 #
 # HexaPDF - A Versatile PDF Creation and Manipulation Library For Ruby
-# Copyright (C) 2016 Thomas Leitner
+# Copyright (C) 2014-2017 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/type/font_simple.rb

@@ -4,7 +4,7 @@
 # This file is part of HexaPDF.
 #
 # HexaPDF - A Versatile PDF Creation and Manipulation Library For Ruby
-# Copyright (C) 2016 Thomas Leitner
+# Copyright (C) 2014-2017 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/type/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) 2016 Thomas Leitner
+# Copyright (C) 2014-2017 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/type/font_type1.rb

@@ -4,7 +4,7 @@
 # This file is part of HexaPDF.
 #
 # HexaPDF - A Versatile PDF Creation and Manipulation Library For Ruby
-# Copyright (C) 2016 Thomas Leitner
+# Copyright (C) 2014-2017 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/type/form.rb

@@ -4,7 +4,7 @@
 # This file is part of HexaPDF.
 #
 # HexaPDF - A Versatile PDF Creation and Manipulation Library For Ruby
-# Copyright (C) 2016 Thomas Leitner
+# Copyright (C) 2014-2017 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/type/graphics_state_parameter.rb

@@ -4,7 +4,7 @@
 # This file is part of HexaPDF.
 #
 # HexaPDF - A Versatile PDF Creation and Manipulation Library For Ruby
-# Copyright (C) 2016 Thomas Leitner
+# Copyright (C) 2014-2017 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/type/image.rb

@@ -4,7 +4,7 @@
 # This file is part of HexaPDF.
 #
 # HexaPDF - A Versatile PDF Creation and Manipulation Library For Ruby
-# Copyright (C) 2016 Thomas Leitner
+# Copyright (C) 2014-2017 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
@@ -31,7 +31,10 @@
 # is created or manipulated using HexaPDF.
 #++
 
+require 'zlib'
+require 'hexapdf/error'
 require 'hexapdf/stream'
+require 'hexapdf/image_loader'
 
 module HexaPDF
   module Type
@@ -41,6 +44,10 @@ module HexaPDF
     # See: PDF1.7 s8.8
     class Image < Stream
 
+      # The structure that is returned by the Image#info method.
+      Info = Struct.new(:type, :width, :height, :color_space, :indexed, :components,
+                        :bits_per_component, :writable, :extension)
+
       define_field :Type,             type: Symbol,          default: :XObject
       define_field :Subtype,          type: Symbol,          required: true, default: :Image
       define_field :Width,            type: Integer,         required: true
@@ -67,6 +74,185 @@ module HexaPDF
       # facility and not when the image is part of a loaded PDF file.
       attr_accessor :source_path
 
+      # Returns an Info structure with information about the image.
+      #
+      # Available accessors:
+      #
+      # type::
+      #    The type of the image. Either :jpeg, :jp2, :jbig2, :ccitt or :png.
+      # width::
+      #    The width of the image.
+      # height::
+      #    The height of the image.
+      # color_space::
+      #    The color space the image uses. Either :rgb, :cmyk, :gray or :other.
+      # indexed::
+      #    Whether the image uses an indexed color space or not.
+      # components::
+      #    The number of color components of the color space, or -1 if the number couldn't be
+      #    determined.
+      # bits_per_component::
+      #    The number of bits per color component.
+      # writable::
+      #    Whether the image can be written by HexaPDF.
+      # extension::
+      #    The file extension that would be used when writing the file. Either jpg, jpx or png. Only
+      #    meaningful when writable is true.
+      def info
+        result = Info.new
+        result.width = self[:Width]
+        result.height = self[:Height]
+        result.bits_per_component = self[:BitsPerComponent]
+        result.indexed = false
+        result.writable = true
+
+        filter, rest = *self[:Filter]
+        case filter
+        when :DCTDecode
+          result.type = :jpeg
+          result.extension = 'jpg'.freeze
+        when :JPXDecode
+          result.type = :jp2
+          result.extension = 'jpx'.freeze
+        when :JBIG2Decode
+          result.type = :jbig2
+        when :CCITTFaxDecode
+          result.type = :ccitt
+        else
+          result.type = :png
+          result.extension = 'png'.freeze
+        end
+
+        if rest || ![:FlateDecode, :DCTDecode, :JPXDecode, nil].include?(filter)
+          result.writable = false
+        end
+
+        color_space, = *self[:ColorSpace]
+        if color_space == :Indexed
+          result.indexed = true
+          color_space, = *document.deref(self[:ColorSpace][1])
+        end
+        case color_space
+        when :DeviceRGB, :CalRGB
+          result.color_space = :rgb
+          result.components = 3
+        when :DeviceGray, :CalGray
+          result.color_space = :gray
+          result.components = 1
+        when :DeviceCMYK
+          result.color_space = :cmyk
+          result.components = 4
+          result.writable = false if result.type == :png
+        else
+          result.color_space = :other
+          result.components = -1
+          result.writable = false if result.type == :png
+        end
+
+        result
+      end
+
+      # :call-seq:
+      #   image.write(basename)
+      #   image.write(io)
+      #
+      # Saves this image XObject to the file with the given name and appends the correct extension
+      # (if the name already contains this extension, the name is used as is), or the given IO
+      # object.
+      #
+      # Raises an error if the image format is not supported.
+      #
+      # The output format and extension depends on the image type as returned by the #info method:
+      #
+      # :jpeg:: Saved as a JPEG file with the extension '.jpg'
+      # :jp2:: Saved as a JPEG2000 file with the extension '.jpx'
+      # :png:: Saved as a PNG file with the extension '.png'
+      def write(name_or_io)
+        info = self.info
+
+        unless info.writable
+          raise HexaPDF::Error, "PDF image format not supported for writing"
+        end
+
+        io = if name_or_io.kind_of?(String)
+               File.open(name_or_io.sub(/\.#{info.extension}\z/, '') + "." + info.extension, "wb")
+             else
+               name_or_io
+             end
+
+        if info.type == :jpeg || info.type == :jp2
+          source = stream_source
+          while source.alive? && (chunk = source.resume)
+            io << chunk
+          end
+        else
+          write_png(io, info)
+        end
+      ensure
+        io.close if io && name_or_io.kind_of?(String)
+      end
+
+      private
+
+      # Writes the image as PNG to the given IO stream.
+      def write_png(io, info)
+        io << ImageLoader::PNG::MAGIC_FILE_MARKER
+
+        color_type = if info.indexed
+                       ImageLoader::PNG::INDEXED
+                     elsif info.color_space == :rgb
+                       ImageLoader::PNG::TRUECOLOR
+                     else
+                       ImageLoader::PNG::GREYSCALE
+                     end
+
+        io << png_chunk('IHDR', [info.width, info.height, info.bits_per_component,
+                                 color_type, 0, 0, 0].pack('N2C5'))
+
+        if key?(:Intent)
+          # PNG s11.3.3.5
+          intent = ImageLoader::PNG::RENDERING_INTENT_MAP.rassoc(self[:Intent]).first
+          io << png_chunk('sRGB', intent.chr) <<
+            png_chunk('gAMA', [45455].pack('N')) <<
+            png_chunk('cHRM', [31270, 32900, 64000, 33000, 30000, 60000, 15000, 6000].pack('N8'))
+        end
+
+        if color_type == ImageLoader::PNG::INDEXED
+          palette_data = document.deref(self[:ColorSpace][3])
+          palette_data = palette_data.stream unless palette_data.kind_of?(String)
+          palette = ''.b
+          if info.color_space == :rgb
+            palette = palette_data[0, palette_data.length - palette_data.length % 3]
+          else
+            palette_data.each_byte {|byte| palette << byte << byte << byte}
+          end
+          io << png_chunk('PLTE', palette)
+        end
+
+        if self[:Mask].kind_of?(Array) && self[:Mask].each_slice(2).all? {|a, b| a == b} &&
+            (color_type == ImageLoader::PNG::TRUECOLOR || color_type == ImageLoader::PNG::GREYSCALE)
+          io << png_chunk('tRNS', self[:Mask].each_slice(2).map {|a, _| a}.pack('n*'))
+        end
+
+        filter, = *self[:Filter]
+        if filter == :FlateDecode && self[:DecodeParms] && self[:DecodeParms][:Predictor].to_i >= 10
+          data = stream_source
+        else
+          flate_decode = GlobalConfiguration.constantize('filter.map', :FlateDecode)
+          data = flate_decode.encoder(stream_decoder, Predictor: 15, Colors: 1, Columns: info.width,
+                                      BitsPerComponent: info.bits_per_component)
+        end
+        io << png_chunk('IDAT', Filter.string_from_source(data))
+
+        io << png_chunk('IEND', '')
+      end
+
+      # Returns the binary representation of the PNG chunk for the given chunk type and data.
+      def png_chunk(type, data = nil)
+        [data.to_s.length].pack("N") << type << data.to_s <<
+          [Zlib.crc32(data, Zlib.crc32(type))].pack("N")
+      end
+
     end
 
   end