combine_pdf 0.1.7 → 0.1.8

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: b2dad9414d26c8b0a8e197cba983fce4f01aa395
-  data.tar.gz: c13bcf4b4ca5b8eb1bcd64c2f06bd86a24a6fb1a
+  metadata.gz: 01bc76f1d54a27d2c7eb98b55a37655d48edd496
+  data.tar.gz: 809d1607c8142ce61b31b078b3f3b24ff2716a05
 SHA512:
-  metadata.gz: 9426188b388f66219a205367e0560883905deb20d7d4bd2041dff0b98cd71126edf46a889caf776830d2af6337fd2b2fef93c503fbbc3a24ad1dde1ea93e004f
-  data.tar.gz: b7089d932bfba7936735ae8e01e0aebf75da0b945b3e9b00033d28e3fe9619d218fd26cf0e011c01f960cd713d45714326370816b78fa7e146d8e13f3d64b95d
+  metadata.gz: ce3db7d462887c3ef9e38e675efae4d8ddd5f4a18636566f695ccb3e5fa357e722c463f5d383889bcd67e5f15da65838549fe3d034aa8fd0adf22f3a9b2deca8
+  data.tar.gz: 0744547ed7fcdbb415925248ffa0b9e766269b90c236c30904a396dbfdd121dfe6ad3ac8f8fac72ea371dbf2ed47cb29fcdc9f9b8d956e6a03fff1354dbfe668

data/CHANGELOG.md

@@ -2,6 +2,12 @@
 
 ***
 
+Change log v.0.1.8
+
+**fix**: Fixed an issue reported by Saba, where PDF files that were written using bad practices (namely, without wrapping their content streams correctly) would not be stamped correctly due to changes in the space matrix (CTM). Fixed by wrapping all existing streams.
+
+***
+
 Change log v.0.1.7
 
 **fix**: PDF `insert` had a typo in the code that would cause failure when unsupported object insertion was attempted - fixed by Nathan Keyes (nkeyes).

data/lib/combine_pdf.rb

@@ -17,6 +17,8 @@ load "combine_pdf/combine_pdf_parser.rb"
 load "combine_pdf/combine_pdf_pdf.rb"
 require "combine_pdf/version"
 
+load "combine_pdf/combine_pdf_methods.rb"
+
 
 # This is a pure ruby library to combine/merge, stmap/overlay and number PDF files - as well as to create tables (ment for indexing combined files).
 #
@@ -103,185 +105,7 @@ require "combine_pdf/version"
 #
 # MIT
 module CombinePDF
-	module_function
-
-	# Create an empty PDF object or create a PDF object from a file (parsing the file).
-	# file_name:: is the name of a file to be parsed.
-	def load(file_name = "")
-		raise TypeError, "couldn't parse and data, expecting type String" unless file_name.is_a?(String) || file_name.is_a?(Pathname)
-		return PDF.new() if file_name == ''
-		PDF.new( PDFParser.new(  IO.read(file_name).force_encoding(Encoding::ASCII_8BIT) ) )
-	end
-	def new(file_name = "")
-		load(file_name)
-	end
-
-	# Create a PDF object from a raw PDF data (parsing the data).
-	# data:: is a string that represents the content of a PDF file.
-	def parse(data)
-		raise TypeError, "couldn't parse and data, expecting type String" unless data.is_a? String
-		PDF.new( PDFParser.new(data) )
-	end
-	# makes a PDFWriter object
-	#
-	# PDFWriter objects reresent an empty page and have the method "textbox"
-	# that adds content to that page.
-	#
-	# PDFWriter objects are used internally for numbering pages (by creating a PDF page
-	# with the page number and "stamping" it over the existing page).
-	#
-	# ::mediabox an Array representing the size of the PDF document. defaults to: [0.0, 0.0, 612.0, 792.0]
-	#
-	# if the page is PDFWriter object as a stamp, the final size will be that of the original page.
-	def create_page(mediabox = [0, 0, 595.3, 841.9])
-		PDFWriter.new mediabox
-	end
-
-	# makes a PDF object containing a table
-	#
-	# all the pages in this PDF object are PDFWriter objects and are
-	# writable using the texbox function (should you wish to add a title, or more info)
-	#
-	# the main intended use of this method is to create indexes (a table of contents) for merged data.
-	#
-	# example:
-	#   pdf = CombinePDF.create_table headers: ["header 1", "another header"], table_data: [ ["this is one row", "with two columns"] , ["this is another row", "also two columns", "the third will be ignored"] ]
-	#   pdf.save "table_file.pdf"
-	#
-	# accepts a Hash with any of the following keys as well as any of the PDFWriter#textbox options:
-	# headers:: an Array of strings with the headers (will be repeated every page).
-	# table_data:: as Array of Arrays, each containing a string for each column. the first row sets the number of columns. extra columns will be ignored.
-	# font:: a registered or standard font name (see PDFWriter). defaults to nil (:Helvetica).
-	# header_font:: a registered or standard font name for the headers (see PDFWriter). defaults to nil (the font for all the table rows).
-	# max_font_size:: the maximum font size. if the string doesn't fit, it will be resized. defaults to 14.
-	# column_widths:: an array of relative column widths ([1,2] will display only the first two columns, the second twice as big as the first). defaults to nil (even widths).
-	# header_color:: the header color. defaults to [0.8, 0.8, 0.8] (light gray).
-	# main_color:: main row color. defaults to nil (transparent / white).
-	# alternate_color:: alternate row color. defaults to [0.95, 0.95, 0.95] (very light gray).
-	# font_color:: font color. defaults to [0,0,0] (black).
-	# border_color:: border color. defaults to [0,0,0] (black).
-	# border_width:: border width in PDF units. defaults to 1.
-	# header_align:: the header text alignment within each column (:right, :left, :center). defaults to :center.
-	# row_align:: the row text alignment within each column. defaults to :left (:right for RTL table).
-	# direction:: the table's writing direction (:ltr or :rtl). this reffers to the direction of the columns and doesn't effect text (rtl text is automatically recognized). defaults to :ltr.
-	# rows_per_page:: the number of rows per page, INCLUDING the header row. deafults to 25.
-	# page_size:: the size of the page in PDF points. defaults to [0, 0, 595.3, 841.9] (A4).
-	def create_table(options = {})
-		defaults = {
-			headers: nil,
-			table_data: [[]],
-			font: nil,
-			header_font: nil,
-			max_font_size: 14,
-			column_widths: nil,
-			header_color: [0.8, 0.8, 0.8],
-			main_color: nil,
-			alternate_color: [0.95, 0.95, 0.95],
-			font_color: [0,0,0],
-			border_color: [0,0,0],
-			border_width: 1,
-			header_align: :center,
-			row_align: nil,
-			direction: :ltr,
-			rows_per_page: 25,
-			page_size: [0, 0, 595.3, 841.9] #A4
-		}
-		options = defaults.merge options
-		options[:header_font] = options[:font] unless options[:header_font]
-		options[:row_align] ||= ( (options[:direction] == :rtl) ? :right : :left )
-		# assert table_data is an array of arrays
-		return false unless (options[:table_data].select {|r| !r.is_a?(Array) }).empty?
-		# compute sizes
-		page_size = options[:page_size]
-		top = page_size[3] * 0.9
-		height = page_size[3] * 0.8 / options[:rows_per_page]
-		from_side = page_size[2] * 0.1
-		width = page_size[2] * 0.8
-		columns = options[:table_data][0].length
-		column_widths = []
-		columns.times {|i| column_widths << (width/columns) }
-		if options[:column_widths]
-			scale = 0
-			options[:column_widths].each {|w| scale += w}
-			column_widths = []
-			options[:column_widths].each { |w|  column_widths << (width*w/scale) }
-		end
-		column_widths = column_widths.reverse if options[:direction] == :rtl
-		# set pdf object and start writing the data
-		table = PDF.new()
-		page = nil
-		rows_per_page = options[:rows_per_page]
-		row_number = rows_per_page + 1
-
-		options[:table_data].each do |row_data|
-			if row_number > rows_per_page
-				page = create_page page_size
-				table << page
-				row_number = 1
-				# add headers
-				if options[:headers]
-					x = from_side
-					headers = options[:headers]
-					headers = headers.reverse if options[:direction] == :rtl
-					column_widths.each_index do |i|
-						text = headers[i].to_s
-						page.textbox text, {x: x, y: (top - (height*row_number)), width: column_widths[i], height: height, box_color: options[:header_color], text_align: options[:header_align] }.merge(options).merge({font: options[:header_font]})
-						x += column_widths[i]
-					end
-					row_number += 1
-				end
-			end
-			x = from_side
-			row_data = row_data.reverse if options[:direction] == :rtl
-			column_widths.each_index do |i|
-				text = row_data[i].to_s
-				box_color = options[:main_color]
-				box_color = options[:alternate_color] if options[:alternate_color] && row_number.odd?
-				page.textbox text, {x: x, y: (top - (height*row_number)), width: column_widths[i], height: height, box_color: box_color, text_align: options[:row_align]}.merge(options)
-				x += column_widths[i]
-			end			
-			row_number += 1
-		end
-		table
-	end
-	def new_table(options = {})
-		create_table options
-	end
-
-	# adds a correctly formatted font object to the font library.
-	#
-	# registered fonts will remain in the library and will only be embeded in
-	# PDF objects when they are used by PDFWriter objects (for example, for numbering pages).
-	#
-	# this function enables plug-ins to expend the font functionality of CombinePDF.
-	#
-	# font_name:: a Symbol with the name of the font. if the fonts exists in the library, it will be overwritten!
-	# font_metrics:: a Hash of font metrics, of the format char => {wx: char_width, boundingbox: [left_x, buttom_y, right_x, top_y]} where char == character itself (i.e. " " for space). The Hash should contain a special value :missing for the metrics of missing characters. an optional :wy might be supported in the future, for up to down fonts.
-	# font_pdf_object:: a Hash in the internal format recognized by CombinePDF, that represents the font object.
-	# font_cmap:: a CMap dictionary Hash) which maps unicode characters to the hex CID for the font (i.e. {"a" => "61", "z" => "7a" }).
-	def register_font(font_name, font_metrics, font_pdf_object, font_cmap = nil)
-		Fonts.register_font font_name, font_metrics, font_pdf_object, font_cmap
-	end
 
-	# adds an existing font (from any PDF Object) to the font library.
-	#
-	# returns the font on success or false on failure.
-	#
-	# example:
-	#   fonts = CombinePDF.new("japanese_fonts.pdf").fonts(true)
-	#   CombinePDF.register_font_from_pdf_object :david, fonts[0]
-	#
-	# VERY LIMITTED SUPPORT:
-	# - at the moment it only imports Type0 fonts.
-	# - also, to extract the Hash of the actual font object you were looking for, is not a trivial matter. I do it on the console.
-	# font_name:: a Symbol with the name of the font registry. if the fonts exists in the library, it will be overwritten! 
-	# font_object:: a Hash in the internal format recognized by CombinePDF, that represents the font object.
-	def register_existing_font font_name, font_object
-		Fonts.register_font_from_pdf_object font_name, font_object
-	end
-	def register_font_from_pdf_object font_name, font_object
-		register_existing_font font_name, font_object
-	end
 end
 
 
@@ -301,7 +125,7 @@ end
 ## test performance with:
 ## puts Benchmark.measure { pdf = CombinePDF.new(file); pdf.save "test.pdf" } # PDFEditor.new_pdf
 ## demo: file_name = "~/Ruby/pdfs/encrypted.pdf"; pdf=0; puts Benchmark.measure { pdf = CombinePDF.new(file_name); pdf.save "test.pdf" }
-## at the moment... my code it terribly slow for larger files... :(
+## at the moment... my code is terribly slow for larger files... :(
 ## The file saving is solved (I hope)... but file loading is an issue.
 ##  pdf.each_object {|obj| puts "Stream length: #{obj[:raw_stream_content].length} was registered as #{obj[:Length].is_a?(Hash)? obj[:Length][:referenced_object][:indirect_without_dictionary] : obj[:Length]}" if obj[:raw_stream_content] }
 ##  pdf.objects.each {|obj| puts "#{obj.class.name}: #{obj[:indirect_reference_id]}, #{obj[:indirect_generation_number]} is: #{obj[:Type] || obj[:indirect_without_dictionary]}" }

data/lib/combine_pdf/combine_pdf_basic_writer.rb

@@ -114,7 +114,8 @@ module CombinePDF
 				border_color: nil,
 				border_width: 0,
 				box_radius: 0,
-				opacity: 1
+				opacity: 1,
+				ctm: [1,0,0,1,0,0]
 			}
 			options.update properties
 			# reset the length and height to meaningful values, if negative
@@ -134,7 +135,7 @@ module CombinePDF
 
 				# set graphic state for the box
 				box_stream << "q\n"
-				box_graphic_state = { ca: options[:opacity], CA: options[:opacity], LW: options[:border_width], LC: 0, LJ: 0,  LD: 0 }
+				box_graphic_state = { ca: options[:opacity], CA: options[:opacity], LW: options[:border_width], LC: 0, LJ: 0,  LD: 0, CTM: options[:ctm]}
 				if options[:box_radius] != 0 # if the text box has rounded corners
 					box_graphic_state[:LC], box_graphic_state[:LJ] =  2, 1
 				end
@@ -228,7 +229,7 @@ module CombinePDF
 
 				# set graphic state for text
 				text_stream << "q\n"
-				text_graphic_state = graphic_state({ca: options[:opacity], CA: options[:opacity], LW: options[:stroke_width].to_f, LC: 2, LJ: 1,  LD: 0})
+				text_graphic_state = graphic_state({ca: options[:opacity], CA: options[:opacity], LW: options[:stroke_width].to_f, LC: 2, LJ: 1,  LD: 0, CTM: options[:ctm]})
 				text_stream << "#{PDFOperations._object_to_pdf text_graphic_state} gs\n"
 				text_stream << "DeviceRGB CS\nDeviceRGB cs\n"
 				# set text render mode

data/lib/combine_pdf/combine_pdf_decrypt.rb

@@ -45,6 +45,19 @@ module CombinePDF
 			when 1,2
 				# raise_encrypted_error
 				_perform_decrypt_proc_ @objects, self.method(:decrypt_RC4)
+			when 4
+				# raise unsupported error for now
+				raise_encrypted_error
+				# make sure CF is a Hash (as required by the PDF standard for this type of encryption).
+				raise_encrypted_error unless @encryption_dictionary[:CF].is_a?(Hash)
+
+				# do nothing if there is no data to decrypt except embeded files...?
+				return true unless (@encryption_dictionary[:CF].values.select { |v| !v[:AuthEvent] || v[:AuthEvent] == :DocOpen } ).empty?
+
+				# attempt to decrypt all strings?
+				# attempt to decrypy all streams
+				# attempt to decrypt all embeded files?
+
 			else
 				raise_encrypted_error
 			end

data/lib/combine_pdf/combine_pdf_methods.rb

@@ -0,0 +1,187 @@
+# -*- encoding : utf-8 -*-
+
+
+
+
+
+module CombinePDF
+	module_function
+
+	# Create an empty PDF object or create a PDF object from a file (parsing the file).
+	# file_name:: is the name of a file to be parsed.
+	def load(file_name = "")
+		raise TypeError, "couldn't parse and data, expecting type String" unless file_name.is_a?(String) || file_name.is_a?(Pathname)
+		return PDF.new() if file_name == ''
+		PDF.new( PDFParser.new(  IO.read(file_name).force_encoding(Encoding::ASCII_8BIT) ) )
+	end
+	def new(file_name = "")
+		load(file_name)
+	end
+
+	# Create a PDF object from a raw PDF data (parsing the data).
+	# data:: is a string that represents the content of a PDF file.
+	def parse(data)
+		raise TypeError, "couldn't parse and data, expecting type String" unless data.is_a? String
+		PDF.new( PDFParser.new(data) )
+	end
+	# makes a PDFWriter object
+	#
+	# PDFWriter objects reresent an empty page and have the method "textbox"
+	# that adds content to that page.
+	#
+	# PDFWriter objects are used internally for numbering pages (by creating a PDF page
+	# with the page number and "stamping" it over the existing page).
+	#
+	# ::mediabox an Array representing the size of the PDF document. defaults to: [0.0, 0.0, 612.0, 792.0]
+	#
+	# if the page is PDFWriter object as a stamp, the final size will be that of the original page.
+	def create_page(mediabox = [0, 0, 595.3, 841.9])
+		PDFWriter.new mediabox
+	end
+
+	# makes a PDF object containing a table
+	#
+	# all the pages in this PDF object are PDFWriter objects and are
+	# writable using the texbox function (should you wish to add a title, or more info)
+	#
+	# the main intended use of this method is to create indexes (a table of contents) for merged data.
+	#
+	# example:
+	#   pdf = CombinePDF.create_table headers: ["header 1", "another header"], table_data: [ ["this is one row", "with two columns"] , ["this is another row", "also two columns", "the third will be ignored"] ]
+	#   pdf.save "table_file.pdf"
+	#
+	# accepts a Hash with any of the following keys as well as any of the PDFWriter#textbox options:
+	# headers:: an Array of strings with the headers (will be repeated every page).
+	# table_data:: as Array of Arrays, each containing a string for each column. the first row sets the number of columns. extra columns will be ignored.
+	# font:: a registered or standard font name (see PDFWriter). defaults to nil (:Helvetica).
+	# header_font:: a registered or standard font name for the headers (see PDFWriter). defaults to nil (the font for all the table rows).
+	# max_font_size:: the maximum font size. if the string doesn't fit, it will be resized. defaults to 14.
+	# column_widths:: an array of relative column widths ([1,2] will display only the first two columns, the second twice as big as the first). defaults to nil (even widths).
+	# header_color:: the header color. defaults to [0.8, 0.8, 0.8] (light gray).
+	# main_color:: main row color. defaults to nil (transparent / white).
+	# alternate_color:: alternate row color. defaults to [0.95, 0.95, 0.95] (very light gray).
+	# font_color:: font color. defaults to [0,0,0] (black).
+	# border_color:: border color. defaults to [0,0,0] (black).
+	# border_width:: border width in PDF units. defaults to 1.
+	# header_align:: the header text alignment within each column (:right, :left, :center). defaults to :center.
+	# row_align:: the row text alignment within each column. defaults to :left (:right for RTL table).
+	# direction:: the table's writing direction (:ltr or :rtl). this reffers to the direction of the columns and doesn't effect text (rtl text is automatically recognized). defaults to :ltr.
+	# rows_per_page:: the number of rows per page, INCLUDING the header row. deafults to 25.
+	# page_size:: the size of the page in PDF points. defaults to [0, 0, 595.3, 841.9] (A4).
+	def create_table(options = {})
+		defaults = {
+			headers: nil,
+			table_data: [[]],
+			font: nil,
+			header_font: nil,
+			max_font_size: 14,
+			column_widths: nil,
+			header_color: [0.8, 0.8, 0.8],
+			main_color: nil,
+			alternate_color: [0.95, 0.95, 0.95],
+			font_color: [0,0,0],
+			border_color: [0,0,0],
+			border_width: 1,
+			header_align: :center,
+			row_align: nil,
+			direction: :ltr,
+			rows_per_page: 25,
+			page_size: [0, 0, 595.3, 841.9] #A4
+		}
+		options = defaults.merge options
+		options[:header_font] = options[:font] unless options[:header_font]
+		options[:row_align] ||= ( (options[:direction] == :rtl) ? :right : :left )
+		# assert table_data is an array of arrays
+		return false unless (options[:table_data].select {|r| !r.is_a?(Array) }).empty?
+		# compute sizes
+		page_size = options[:page_size]
+		top = page_size[3] * 0.9
+		height = page_size[3] * 0.8 / options[:rows_per_page]
+		from_side = page_size[2] * 0.1
+		width = page_size[2] * 0.8
+		columns = options[:table_data][0].length
+		column_widths = []
+		columns.times {|i| column_widths << (width/columns) }
+		if options[:column_widths]
+			scale = 0
+			options[:column_widths].each {|w| scale += w}
+			column_widths = []
+			options[:column_widths].each { |w|  column_widths << (width*w/scale) }
+		end
+		column_widths = column_widths.reverse if options[:direction] == :rtl
+		# set pdf object and start writing the data
+		table = PDF.new()
+		page = nil
+		rows_per_page = options[:rows_per_page]
+		row_number = rows_per_page + 1
+
+		options[:table_data].each do |row_data|
+			if row_number > rows_per_page
+				page = create_page page_size
+				table << page
+				row_number = 1
+				# add headers
+				if options[:headers]
+					x = from_side
+					headers = options[:headers]
+					headers = headers.reverse if options[:direction] == :rtl
+					column_widths.each_index do |i|
+						text = headers[i].to_s
+						page.textbox text, {x: x, y: (top - (height*row_number)), width: column_widths[i], height: height, box_color: options[:header_color], text_align: options[:header_align] }.merge(options).merge({font: options[:header_font]})
+						x += column_widths[i]
+					end
+					row_number += 1
+				end
+			end
+			x = from_side
+			row_data = row_data.reverse if options[:direction] == :rtl
+			column_widths.each_index do |i|
+				text = row_data[i].to_s
+				box_color = options[:main_color]
+				box_color = options[:alternate_color] if options[:alternate_color] && row_number.odd?
+				page.textbox text, {x: x, y: (top - (height*row_number)), width: column_widths[i], height: height, box_color: box_color, text_align: options[:row_align]}.merge(options)
+				x += column_widths[i]
+			end			
+			row_number += 1
+		end
+		table
+	end
+	def new_table(options = {})
+		create_table options
+	end
+
+	# adds a correctly formatted font object to the font library.
+	#
+	# registered fonts will remain in the library and will only be embeded in
+	# PDF objects when they are used by PDFWriter objects (for example, for numbering pages).
+	#
+	# this function enables plug-ins to expend the font functionality of CombinePDF.
+	#
+	# font_name:: a Symbol with the name of the font. if the fonts exists in the library, it will be overwritten!
+	# font_metrics:: a Hash of font metrics, of the format char => {wx: char_width, boundingbox: [left_x, buttom_y, right_x, top_y]} where char == character itself (i.e. " " for space). The Hash should contain a special value :missing for the metrics of missing characters. an optional :wy might be supported in the future, for up to down fonts.
+	# font_pdf_object:: a Hash in the internal format recognized by CombinePDF, that represents the font object.
+	# font_cmap:: a CMap dictionary Hash) which maps unicode characters to the hex CID for the font (i.e. {"a" => "61", "z" => "7a" }).
+	def register_font(font_name, font_metrics, font_pdf_object, font_cmap = nil)
+		Fonts.register_font font_name, font_metrics, font_pdf_object, font_cmap
+	end
+
+	# adds an existing font (from any PDF Object) to the font library.
+	#
+	# returns the font on success or false on failure.
+	#
+	# example:
+	#   fonts = CombinePDF.new("japanese_fonts.pdf").fonts(true)
+	#   CombinePDF.register_font_from_pdf_object :david, fonts[0]
+	#
+	# VERY LIMITTED SUPPORT:
+	# - at the moment it only imports Type0 fonts.
+	# - also, to extract the Hash of the actual font object you were looking for, is not a trivial matter. I do it on the console.
+	# font_name:: a Symbol with the name of the font registry. if the fonts exists in the library, it will be overwritten! 
+	# font_object:: a Hash in the internal format recognized by CombinePDF, that represents the font object.
+	def register_existing_font font_name, font_object
+		Fonts.register_font_from_pdf_object font_name, font_object
+	end
+	def register_font_from_pdf_object font_name, font_object
+		register_existing_font font_name, font_object
+	end
+end

data/lib/combine_pdf/combine_pdf_operations.rb

@@ -16,6 +16,13 @@ module CombinePDF
 	# differentiate between complex PDF objects.
 	PRIVATE_HASH_KEYS = [:indirect_reference_id, :indirect_generation_number, :raw_stream_content, :is_reference_only, :referenced_object, :indirect_without_dictionary]
 
+	# holds a simple content stream that starts a PDF graphic state container - used for wrapping malformed PDF content streams.
+	CONTENT_CONTAINER_START = { is_reference_only: true , referenced_object: {indirect_reference_id: 0, raw_stream_content: 'q'} }
+	# holds a simple content stream that ends a PDF graphic state container - used for wrapping malformed PDF content streams.
+	CONTENT_CONTAINER_MIDDLE = { is_reference_only: true , referenced_object: {indirect_reference_id: 0, raw_stream_content: "Q\nq"} }
+	# holds a simple content stream that ends a PDF graphic state container - used for wrapping malformed PDF content streams.
+	CONTENT_CONTAINER_END = { is_reference_only: true , referenced_object: {indirect_reference_id: 0, raw_stream_content: 'Q'} }
+
 	# @private
 	# @!visibility private
 	#:nodoc: all
@@ -75,10 +82,16 @@ module CombinePDF
 
 			if top # if this is a stamp (overlay)
 				page[:Contents] = original_contents
+				page[:Contents].unshift CONTENT_CONTAINER_START.dup
+				page[:Contents].push CONTENT_CONTAINER_MIDDLE.dup
 				page[:Contents].push *stream_contents
+				page[:Contents].push CONTENT_CONTAINER_END.dup
 			else #if this was a watermark (underlay? would be lost if the page was scanned, as white might not be transparent)
 				page[:Contents] = stream_contents
+				page[:Contents].unshift CONTENT_CONTAINER_START.dup
+				page[:Contents].push CONTENT_CONTAINER_MIDDLE.dup
 				page[:Contents].push *original_contents
+				page[:Contents].push CONTENT_CONTAINER_END.dup
 			end
 
 			page
@@ -139,7 +152,8 @@ module CombinePDF
 					stream[:raw_stream_content].gsub! _object_to_pdf(old_key), _object_to_pdf(new_key)  ##### PRAY(!) that the parsed datawill be correctly reproduced! 
 				end
 				# patch back to PDF defaults, for OCRed PDF files.
-				stream[:raw_stream_content] = "q\nq\nq\nDeviceRGB CS\nDeviceRGB cs\n0 0 0 rg\n0 0 0 RG\n0 Tr\n%s\nQ\nQ\nQ\n" % stream[:raw_stream_content]
+				# stream[:raw_stream_content] = "q\nq\nq\nDeviceRGB CS\nDeviceRGB cs\n0 0 0 rg\n0 0 0 RG\n0 Tr\n%s\nQ\nQ\nQ\n" % stream[:raw_stream_content]
+				stream[:raw_stream_content] = "q\nq\nq\nDeviceRGB CS\nDeviceRGB cs\n0 0 0 rg\n0 0 0 RG\n0 Tr\n1 0 0 1 0 0 cm\n%s\nQ\nQ\nQ\n" % stream[:raw_stream_content]
 			end
 
 			new_page
@@ -158,6 +172,13 @@ module CombinePDF
 			end
 		end
 
+		# returns the PDF Object Hash holding the acutal data (if exists) or the original hash (if it wasn't a reference)
+		#
+		# works only AFTER references have been connected.
+		def get_referenced object
+			object[:referenced_object] || object
+		end
+
 
 		# Ruby normally assigns pointes.
 		# noramlly:

data/lib/combine_pdf/combine_pdf_pdf.rb

@@ -531,7 +531,7 @@ module CombinePDF
 			each_object do |obj|
 				if obj[:is_reference_only]
 					obj[:referenced_object] = objects_reference_hash[ [obj[:indirect_reference_id], obj[:indirect_generation_number] ]   ]
-					warn "couldn't connect a reference!!! could be a null object, Silent error!!!" unless obj[:referenced_object]
+					warn "couldn't connect a reference!!! could be a null or removed (empty) object, Silent error!!!" unless obj[:referenced_object]
 				end
 			end
 

data/lib/combine_pdf/version.rb

@@ -1,3 +1,3 @@
 module CombinePdf
-  VERSION = "0.1.7"
+  VERSION = "0.1.8"
 end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: combine_pdf
 version: !ruby/object:Gem::Version
-  version: 0.1.7
+  version: 0.1.8
 platform: ruby
 authors:
 - Boaz Segev
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2014-11-18 00:00:00.000000000 Z
+date: 2014-12-18 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: ruby-rc4
@@ -73,6 +73,7 @@ files:
 - lib/combine_pdf/combine_pdf_decrypt.rb
 - lib/combine_pdf/combine_pdf_filter.rb
 - lib/combine_pdf/combine_pdf_fonts.rb
+- lib/combine_pdf/combine_pdf_methods.rb
 - lib/combine_pdf/combine_pdf_operations.rb
 - lib/combine_pdf/combine_pdf_parser.rb
 - lib/combine_pdf/combine_pdf_pdf.rb
@@ -102,4 +103,3 @@ signing_key:
 specification_version: 4
 summary: Combine, stamp and watermark PDF files in pure Ruby.
 test_files: []
-has_rdoc: