Class: Rpdf2txt::Stream

Inherits:
PdfObject show all
Defined in:
lib/rpdf2txt/object.rb

Direct Known Subclasses

CMap, Image, ObjStream

Constant Summary collapse

BT_PATTERN = 
/\bBT\b(?!(\\[()]|[^(\\])*\))/mn
ET_PATTERN = 
/\bET\b(?!(\\[()]|[^(\\])*\))/mn
FAIL_PTRN = 
/\((\\[()]|[^)])*\bET\b\s*$/mn
@@nontext_scan_pattern = 
%r!(?:#{dm_str})|(\b[qQ]\b)|#{xobj}|\bBI\b(.*?)\bID\b(.*?)\b(EI)\b!mn
@@hr_scan_pattern = 
/#{num}#{num}(\b[lm]\b)/mn

Instance Attribute Summary

Attributes inherited from PdfObject

#attributes, #decoder, #oid, #src

Instance Method Summary collapse

Methods inherited from PdfObject

#_parse_attributes, #build_tree, #catalogue_object, #extract_attribute_stream, #initialize, #parse_attributes, #revision_id

Constructor Details

This class inherits a constructor from Rpdf2txt::PdfObject

Instance Method Details

#append(decoded_stream) ⇒ Object 



702
703
704
 
# File 'lib/rpdf2txt/object.rb', line 702

def append(decoded_stream)
	(@decoded_stream  ||= '') << decoded_stream
end

#decode_raw_stream ⇒ Object 



816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
 
# File 'lib/rpdf2txt/object.rb', line 816

def decode_raw_stream
	@decrypted_stream = raw_stream
	unless(@decoder.nil?)
		@decrypted_stream = @decoder.decrypt(self)
	end
	stream = @decrypted_stream
	[@attributes[:filter]].flatten.compact.each { |filter|
      begin
        stream = case filter
                 when "/FlateDecode"
                   flate_decode stream
                 when "/LZWDecode"
                   lzw_decode stream
                 else
                   raise "Unimplemented filter: #{filter}"
                 end
      rescue StandardError => err
        warn "'#{err.message}' when filtering with #{filter}"
      end
	}
	stream
end

#decoded_stream ⇒ Object 



708
709
710
 
# File 'lib/rpdf2txt/object.rb', line 708

def decoded_stream
	@decoded_stream ||= decode_raw_stream
end

#decoded_stream=(decoded_stream) ⇒ Object 



705
706
707
 
# File 'lib/rpdf2txt/object.rb', line 705

def decoded_stream=(decoded_stream)
	@decoded_stream = decoded_stream
end

#extract_horizontal_rules(dm_src, dmatrix, result) ⇒ Object 



711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
 
# File 'lib/rpdf2txt/object.rb', line 711

def extract_horizontal_rules(dm_src, dmatrix, result)
  last_x = 0
  last_y = 0
  dm_src.scan(@@hr_scan_pattern) { |matches|
    case matches.last.to_s[-1]
    when ?l
      x = matches[0].to_f
      y = matches[1].to_f
      if(x != last_x && y == last_y)
        hr = HorizontalRule.new(x, y, dmatrix)
        hr.current_page, hr.text_state = @page, @text_state
        result.push(hr)
      end
      last_x = x
      last_y = y
    when ?m
      last_x = matches[0].to_f
      last_y = matches[1].to_f
    end
  }
end

#extract_nontext_objects(dm_src, dmatrix, stack, result) ⇒ Object 



732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
 
# File 'lib/rpdf2txt/object.rb', line 732

def extract_nontext_objects(dm_src, dmatrix, stack, result)
  dm_src.scan(@@nontext_scan_pattern) { |matches|
    matches = matches.compact
    case matches.last
    when 'q'
      stack.push(dmatrix)
    when 'Q'
      dmatrix = stack.pop
    when 'Do'
      x, y = (txt = result.last) ? [txt.x, txt.y] : [0, 0]
      ip = ImagePlacement.new(matches[-2], x, y, dmatrix)
      ip.current_page, ip.text_state = @page, @text_state
      result.push ip
    when 'EI'
      attrs, data, _ = matches
      im = InlineImage.new attrs, data.strip
      ip = ImagePlacement.new im, 0, 0, dmatrix
      ip.current_page, ip.text_state = @page, @text_state
      result.push ip
    else
      mmatrix = Matrix[[matches[0].to_f, matches[1].to_f,0],
                       [matches[2].to_f, matches[3].to_f,0],
                       [matches[4].to_f, matches[5].to_f,1]]
      dmatrix = dmatrix * mmatrix
    end
  }
  dmatrix
end

#extract_text_objects(page, text_state) ⇒ Object 



760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
 
# File 'lib/rpdf2txt/object.rb', line 760

def extract_text_objects(page, text_state)
    @page, @text_state = page, text_state
	stack = []
	result = []
	startpoint = decoded_stream.index(BT_PATTERN)
	endpoint = decoded_stream.index(ET_PATTERN)
    while FAIL_PTRN.match(decoded_stream[0..(endpoint+2)])
      endpoint = decoded_stream.index(ET_PATTERN, endpoint.next)
    end
	unless(startpoint && endpoint && (startpoint < endpoint))
		startpoint = 0
	end
    rotation = (page && Math::PI * page.attributes[:rotate].to_f / 180) || 0
	dmatrix = Matrix[[Math.cos(rotation),Math.sin(rotation),0],
                     [Math.sin(rotation),-Math.cos(rotation),0],
                     [0,0,1]]

	dm_src = decoded_stream[0...startpoint]
	while(endpoint && startpoint)
		### pick out the bits in between Text that are relevant to 
		### text positioning (such as the device-transformation-matrix)
		### NOTE: as far as I understand, the device matrix should 
		###       not be used to position text. However it is used 
		###       by some PDF-Creators and therefore we have to include
		###       it in our calculations.
      dmatrix = extract_nontext_objects(dm_src, dmatrix, stack, result)
      extract_horizontal_rules(dm_src, dmatrix, result)
		tsrc = decoded_stream[startpoint..(endpoint+2)]
      while FAIL_PTRN.match(tsrc)
        endpoint = decoded_stream.index(ET_PATTERN, endpoint + 2) || -1
        tsrc = decoded_stream[startpoint..(endpoint+2)]
      end
		text = Text.new(tsrc, @target_encoding, dmatrix)
		text.current_page = page
		text.text_state = text_state
      result.concat text.scan
		startpoint = decoded_stream.index(BT_PATTERN, endpoint)
		if(startpoint)
			dm_src = decoded_stream[endpoint...startpoint]
			endpoint = decoded_stream.index(ET_PATTERN, startpoint)
		end
	end
	result
end

#flate_decode(data) ⇒ Object 



838
839
840
 
# File 'lib/rpdf2txt/object.rb', line 838

def flate_decode(data)
  Zlib::Inflate.inflate(data)
end

#lzw_decode(data) ⇒ Object 



841
842
843
844
845
846
847
848
 
# File 'lib/rpdf2txt/object.rb', line 841

def lzw_decode(data)
  require 'rpdf2txt/lzw'
  earlychange = (parm = @attributes[:decodeparms]) && parm[:earlychange]
  if length = @attributes[:length]
    data = data[0, length.to_i]
  end
  LZW.decode data, (earlychange || 1).to_i
end

#raw_stream ⇒ Object 



804
805
806
807
808
809
810
811
812
813
814
815
 
# File 'lib/rpdf2txt/object.rb', line 804

def raw_stream
	#@raw_stream ||= @src.scan(/stream[\r\n]{1,2}(.*)endstream/mn).to_s
	#@raw_stream ||= @src.scan(/stream[\r\n]{1,2}(.*)endstream/mn)[0][0]
          unless(@raw_stream)
            if(src_scan = @src.scan(/stream[\r\n]{1,2}(.*)endstream/mn) and !src_scan.empty?)
	    @raw_stream = src_scan[0][0]
            else
              @raw_stream = src_scan.to_s
            end
          end
          return @raw_stream
end

#to_cmap ⇒ Object 



849
850
851
 
# File 'lib/rpdf2txt/object.rb', line 849

def to_cmap
	cmap = CMap.new(@src, @target_encoding)
end