combine_pdf 1.0.16 → 1.0.29

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: d071703a8903bc5a11d7beb5a0bcb284a8bced5763705de3a7dd9ae6be35b64d
-  data.tar.gz: 898069ad7ec79ad4fadea5383a721f28c5dfc3632fc20aaa40624954bfe70170
+  metadata.gz: 773cc37e05ad8fb02f148c12e3b7e1e24be8fc71ec946e4bca674bad107e1398
+  data.tar.gz: ecf125185a04ecdbced0856db01514d90036f93b315ff384b3b55cc794018032
 SHA512:
-  metadata.gz: d52868ff7a021801207ff17a1e87aca8fa1bd82e6fd0dacf91d39d57695b07b87238898238fa90755bfaaec65cb4df384b6db97f3f4e3bf70da7d3d808e55daa
-  data.tar.gz: f68133d14d5eb5f0428097b4421976da49471eed61f4ac0e574c4718b588458f6971fd107ccc397be004997f3c31ac6bfc122675992412fd521a7910d6f2abd9
+  metadata.gz: d8c2636e089c11b71f4a07d29a9eea6f7df8ca1f5466c2a447e8afb2598467d2fe3663d4e382c45a04dfd6ead9f19a6afe9ae7e01fae22b8c678db08ba6d1fe8
+  data.tar.gz: efc8160687a50fbf0df43cc5b0b85c013fc5fda3f2a053bc37716b403de18ad429caa4872832047a1ea1e2167797c0d8a5a773d8c341128752e38db9a87990ad

data/.github/workflows/main.yml

@@ -0,0 +1,32 @@
+name: Main
+on:
+  push:
+
+jobs:
+  tests:
+    name: Tests
+    runs-on: ubuntu-latest
+    strategy:
+      fail-fast: false
+      matrix:
+        ruby: ["2.7", "3.0", "3.1", "3.2", "3.3"]
+        rubyopt: [""]
+        include:
+          - ruby: "3.3"
+            rubyopt: "--enable-frozen-string-literal --debug-frozen-string-literal"
+
+    steps:
+      - name: Checkout code
+        uses: actions/checkout@v3
+
+      - name: Setup Ruby
+        uses: ruby/setup-ruby@v1
+        with:
+          ruby-version: ${{ matrix.ruby }}
+          bundler-cache: true
+
+      - name: Generate lockfile
+        run: bundle lock
+
+      - name: Run tests
+        run: bundle exec rake test RUBYOPT="${{ matrix.rubyopt }}"

data/.travis.yml

@@ -0,0 +1 @@
+language: ruby

data/CHANGELOG.md

@@ -1,6 +1,66 @@
 # Change Log
 
-***
+#### Change log v.1.0.29 (2024-12-07)
+
+**Fix**: frozen string literal support fix. Credit to @francescob (Francesco) for PR #245.
+
+#### Change log v.1.0.28 (2024-11-12)
+
+**Fix**: use `require` to load code (instead of `load`). Credit to @casperisfine (Jean byroot Boussier) for PR #216.
+
+#### Change log v.1.0.27 (2024-11-10)
+
+**Performance**: fix performance issues with `object_id` usage in Ruby 3+. Credit to @amomchilov (Alexander Momchilov) for PR #241.
+
+**Performance**: use frozen string literals. Credit to @casperisfine (Jean byroot Boussier) for PR #239.
+
+#### Change log v.1.0.26 (2023-12-22)
+
+**Performance**: possible performance bump. Credit to @denislavski (Denislav Naydenov) for opening PR #235.
+
+#### Change log v.1.0.25 (2023-12-19)
+
+**Fix**: possible improve memory usage. Credit to @denislavski (Denislav Naydenov) for opening PR #233 and suggesting this change.
+
+#### Change log v.1.0.24 (2023-10-19)
+
+**Fix**: possible `nil` in loop. Credit to @jkowens for PR #231 and adding a quick fix using a simple guard.
+
+**Fix**: preserve file creation date metadata where relevant.
+
+#### Change log v.1.0.23 (2023-04-04)
+
+**Feature**: merged PR #177 for the `raise_on_encrypted: true` option support. Credit to @leviwilson and @kimyu92 for the PR.
+
+#### Change log v.1.0.22
+
+**Fix**: fix `fonts` dereferencing issue (#203), credit to @MarcWeber (Marc Weber) for identifying the issue.
+
+**Fix**: fix `metrix` dependency, credit to @casperisfine (Jean byroot Boussier) for PR #195.
+
+#### Change log v.1.0.21
+
+**Fix**: possible fix for issue #184, where nested PDF files within an object stream could break the parser. Credit to Greg Sparrow (@hazelsparrow) for exposng the issue.
+
+#### Change log v.1.0.20
+
+**Fix**: merges PR #180, `TypeError: can't dup NilClass`. Credit to Adam Trepanier (@adam-e-trepanier) for the merge.
+
+#### Change log v.1.0.19
+
+**Fix**: fixes font height and width detection issue. Issue #179. Credit to @5anchezzz for opening the issue.
+
+**Fix**: fixes an indentation warning. Issue #173. Credit to @rubyFeedback for exposing this issue.
+
+#### Change log v.1.0.18
+
+**Fix**: fixed issue with the 1.0.17 release where `ProcSet` PDF Arrays should have been expected but where ignored and a PDF Object was assumed instead (issue #171) - credit to @chuchiperriman (Jesús Barbero Rodríguez).
+
+#### Change log v.1.0.17
+
+NB: yanked from RubyGems.org.
+
+**Fix**: fixed issue where nested structure equality tests might provide false positives, resulting in lost data (issue #166) - credit to @cschilbe (Conrad Schilbe).
 
 #### Change log v.1.0.16
 

data/README.md

@@ -1,27 +1,29 @@
 # CombinePDF - the ruby way for merging PDF files
 [![Gem Version](https://badge.fury.io/rb/combine_pdf.svg)](http://badge.fury.io/rb/combine_pdf)
 [![GitHub](https://img.shields.io/badge/GitHub-Open%20Source-blue.svg)](https://github.com/boazsegev/combine_pdf)
+[![Documentation](http://inch-ci.org/github/boazsegev/combine_pdf.svg?branch=master)](https://www.rubydoc.info/github/boazsegev/combine_pdf)
 [![Maintainers Wanted](https://img.shields.io/badge/maintainers-wanted-red.svg)](https://github.com/pickhardt/maintainers-wanted)
 
+
 CombinePDF is a nifty model, written in pure Ruby, to parse PDF files and combine (merge) them with other PDF files, watermark them or stamp them (all using the PDF file format and pure Ruby code).
 
-## Install
+## Unmaintained - Help Wanted(!)
 
-Install with ruby gems:
+I decided to stop maintaining this gem and hope someone could take over the PR reviews and maintenance of this gem (or simply open a successful fork).
 
-```ruby
-gem install combine_pdf
-```
+I wrote this gem because I needed to solve an issue with bates-numbering existing PDF documents.
 
-## Help Wanted
+However, since 2014 I have been maintaining the gem for free and for no reason at all, except that I enjoyed sharing it with the community.
 
-I need help maintaining the CombinePDF Ruby gem.
+I love this gem, but I cannot keep maintaining it as I have my own projects to focus own and I need both the time and (more importantly) the mindspace.
 
-I wrote this gem because I needed to solve an issue with bates-numbering existing PDF documents. However, during the last three years or so I have been maintaining the project for no reason at all, except that I enjoyed sharing it with the community.
+## Install
 
-I love this gem, but I feel it's time I took a step back from maintaining it and concentrate on my music and other things I want to develop.
+Install with ruby gems:
 
-Please hit me up if you would like to join in and eventually take over.
+```ruby
+gem install combine_pdf
+```
 
 ## Known Limitations
 
@@ -37,10 +39,12 @@ Quick rundown:
 
     Some links will be lost when ripping pages out of PDF files and merging them with another PDF.
 
-* Some encrypted PDF files (usually the ones you can't view without a password) will fail quietly instead of noisily.
+* Some encrypted PDF files (usually the ones you can't view without a password) will fail quietly instead of noisily. If you prefer to choose the noisy route, you can specify the `raise_on_encrypted` option using `CombinePDF.load(pdf_file, raise_on_encrypted: true)` which will raise a `CombinePDF::EncryptionError`.
 
 * Sometimes the CombinePDF will raise an exception even if the PDF could be parsed (i.e., when PDF optional content exists)... I find it better to err on the side of caution, although for optional content PDFs an exception is avoidable using `CombinePDF.load(pdf_file, allow_optional_content: true)`.
 
+* The CombinePDF gem runs recursive code to both parse and format the PDF files. Hence, PDF files that have heavily nested objects, as well as those that where combined in a way that results in cyclic nesting, might explode the stack - resulting in an exception or program failure.
+
 CombinePDF is written natively in Ruby and should (presumably) work on all Ruby platforms that follow Ruby 2.0 compatibility.
 
 However, PDF files are quite complex creatures and no guaranty is provided.
@@ -112,7 +116,42 @@ pdf.number_pages
 pdf.save "file_with_numbering.pdf"
 ```
 
-Numbering can be done with many different options, with different formating, with or without a box object, and even with opacity values - see documentation.
+Numbering can be done with many different options, with different formating, with or without a box object, and even with opacity values - [see documentation](https://www.rubydoc.info/github/boazsegev/combine_pdf/CombinePDF/PDF#number_pages-instance_method).
+
+For example, should you prefer to place the page number on the bottom right side of all PDF pages, do:
+
+```ruby
+pdf.number_pages(location: [:bottom_right])
+```
+
+As another example, the dashes around the number are removed and a box is placed around it. The numbering is semi-transparent and the first 3 pages are numbered using letters (a,b,c) rather than numbers:
+
+
+```ruby
+# number first 3 pages as "a", "b", "c"
+pdf.number_pages(number_format: " %s ",
+                 location: [:top, :bottom, :top_left, :top_right, :bottom_left, :bottom_right],
+                 start_at: "a",
+                 page_range: (0..2),
+                 box_color: [0.8,0.8,0.8],
+                 border_color: [0.4, 0.4, 0.4],
+                 border_width: 1,
+                 box_radius: 6,
+                 opacity: 0.75)
+# number the rest of the pages as 4, 5, ... etc'
+pdf.number_pages(number_format: " %s ",
+                 location: [:top, :bottom, :top_left, :top_right, :bottom_left, :bottom_right],
+                 start_at: 4,
+                 page_range: (3..-1),
+                 box_color: [0.8,0.8,0.8],
+                 border_color: [0.4, 0.4, 0.4],
+                 border_width: 1,
+                 box_radius: 6,
+                 opacity: 0.75)
+```
+
+    pdf.number_pages(number_format: " %s ", location: :bottom_right, font_size: 44)
+
 
 ## Loading and Parsing PDF data
 

data/Rakefile

@@ -1,2 +1,9 @@
-require "bundler/gem_tasks"
+require 'bundler/gem_tasks'
+require 'rake/testtask'
+
+task default: 'test'
+
+Rake::TestTask.new do |t|
+  t.test_files = FileList['test/**/*_test.rb']
+end
 

data/combine_pdf.gemspec

@@ -7,7 +7,7 @@ Gem::Specification.new do |spec|
   spec.name          = "combine_pdf"
   spec.version       = CombinePDF::VERSION
   spec.authors       = ["Boaz Segev"]
-  spec.email         = ["boaz@2be.co.il"]
+  spec.email         = ["bo@bowild.com"]
   spec.summary       = %q{Combine, stamp and watermark PDF files in pure Ruby.}
   spec.description   = %q{A nifty gem, in pure Ruby, to parse PDF files and combine (merge) them with other PDF files, number the pages, watermark them or stamp them, create tables, add basic text objects etc` (all using the PDF file format).}
   spec.homepage      = "https://github.com/boazsegev/combine_pdf"
@@ -19,8 +19,10 @@ Gem::Specification.new do |spec|
   spec.require_paths = ["lib"]
 
   spec.add_runtime_dependency 'ruby-rc4', '>= 0.1.5'
+  spec.add_runtime_dependency 'matrix'
 
   # spec.add_development_dependency "bundler", ">= 1.7"
-  spec.add_development_dependency "rake", "~> 10.0"
+  spec.add_development_dependency "rake", ">= 12.3.3"
   spec.add_development_dependency "minitest"
+  spec.add_development_dependency "minitest-around"
 end

data/lib/combine_pdf/api.rb

@@ -1,4 +1,5 @@
 # -*- encoding : utf-8 -*-
+# frozen_string_literal: true
 
 module CombinePDF
   module_function
@@ -24,11 +25,11 @@ module CombinePDF
     raise TypeError, "couldn't create PDF object, expecting type String" unless string.is_a?(String) || string.is_a?(Pathname)
     begin
       (begin
-     File.file? string
-   rescue
-     false
-   end) ? load(string) : parse(string)
-   rescue => _e
+        File.file? string
+      rescue
+        false
+      end) ? load(string) : parse(string)
+    rescue => _e
       raise 'General PDF error - Use CombinePDF.load or CombinePDF.parse for a non-general error message (the requested file was not found OR the string received is not a valid PDF stream OR the file was found but not valid).'
     end
   end

data/lib/combine_pdf/basic_writer.rb

@@ -1,4 +1,5 @@
 # -*- encoding : utf-8 -*-
+# frozen_string_literal: true
 ########################################################
 ## Thoughts from reading the ISO 32000-1:2008
 ## this file is part of the CombinePDF library and the code
@@ -35,7 +36,7 @@ module CombinePDF
     # mediabox:: the PDF page size in PDF points. defaults to [0, 0, 612.0, 792.0] (US Letter)
     def initialize(mediabox = [0, 0, 612.0, 792.0])
       # indirect_reference_id, :indirect_generation_number
-      @contents = ''
+      @contents = String.new
       @base_font_name = 'Writer' + SecureRandom.hex(7) + 'PDF'
       self[:Type] = :Page
       self[:indirect_reference_id] = 0

data/lib/combine_pdf/decrypt.rb

@@ -1,4 +1,5 @@
 # -*- encoding : utf-8 -*-
+# frozen_string_literal: true
 ########################################################
 ## Thoughts from reading the ISO 32000-1:2008
 ## this file is part of the CombinePDF library and the code
@@ -137,7 +138,7 @@ module CombinePDF
       object_key = @key.dup
       object_key << [encrypted_id].pack('i')[0..2]
       object_key << [encrypted_generation].pack('i')[0..1]
-      object_key << 'sAlT'.force_encoding(Encoding::ASCII_8BIT)
+      object_key << 'sAlT'.b
       key_length = object_key.length < 16 ? object_key.length : 16
 
       begin

data/lib/combine_pdf/exceptions.rb

@@ -1,3 +1,5 @@
+# frozen_string_literal: true
+
 module CombinePDF
   class EncryptionError < StandardError
   end

data/lib/combine_pdf/filter.rb

@@ -1,4 +1,5 @@
 # -*- encoding : utf-8 -*-
+# frozen_string_literal: true
 ########################################################
 ## Thoughts from reading the ISO 32000-1:2008
 ## this file is part of the CombinePDF library and the code

data/lib/combine_pdf/fonts.rb

@@ -1,4 +1,5 @@
 # -*- encoding : utf-8 -*-
+# frozen_string_literal: true
 ########################################################
 ## Thoughts from reading the ISO 32000-1:2008
 ## this file is part of the CombinePDF library and the code
@@ -138,12 +139,21 @@ module CombinePDF
       text.each_char do |c|
         metrics_array << (merged_metrics[c] || { wx: 0, boundingbox: [0, 0, 0, 0] })
       end
-      height = metrics_array.map { |m| m ? m[:boundingbox][3] : 0 } .max
-      height -= (metrics_array.map { |m| m ? m[:boundingbox][1] : 0 }).min
+      metrics_array_mapped_top = [].dup
+      metrics_array_mapped_bottom = [].dup
       width = 0.0
       metrics_array.each do |m|
-        width += (m[:wx] || m[:wy])
+        if (m && m[:boundingbox])
+          metrics_array_mapped_top << m[:boundingbox][3]
+          metrics_array_mapped_bottom << m[:boundingbox][1]
+        else
+          metrics_array_mapped_top << 0
+          metrics_array_mapped_bottom << 0
+        end
+        width += (m[:wx] || m[:wy] || 0) if m
       end
+      height = metrics_array_mapped_top.max
+      height -=metrics_array_mapped_bottom.min
       return [height.to_f / 1000 * size, width.to_f / 1000 * size] if metrics_array[0][:wy]
       [width.to_f / 1000 * size, height.to_f / 1000 * size]
     end

data/lib/combine_pdf/page_methods.rb

@@ -1,4 +1,5 @@
 # -*- encoding : utf-8 -*-
+# frozen_string_literal: true
 ########################################################
 ## Thoughts from reading the ISO 32000-1:2008
 ## this file is part of the CombinePDF library and the code
@@ -94,7 +95,7 @@ module CombinePDF
       # end
 
       # set ProcSet to recommended value
-      resources[:ProcSet] = [:PDF, :Text, :ImageB, :ImageC, :ImageI] # this was recommended by the ISO. 32000-1:2008
+      resources[:ProcSet] ||= [:PDF, :Text, :ImageB, :ImageC, :ImageI] # this was recommended by the ISO. 32000-1:2008
 
       if top # if this is a stamp (overlay)
         insert_content CONTENT_CONTAINER_START, 0
@@ -147,7 +148,7 @@ module CombinePDF
 
     # This method adds a simple text box to the Page represented by the PDFWriter class.
     # This function takes two values:
-    # text:: the text to potin the box.
+    # text:: the text to write in the box.
     # properties:: a Hash of box properties.
     # the symbols and values in the properties Hash could be any or all of the following:
     # x:: the left position of the box.
@@ -214,7 +215,7 @@ module CombinePDF
       options[:text_padding] = 0 if options[:text_padding].to_f >= 1
 
       # create box stream
-      box_stream = ''
+      box_stream = +''
       # set graphic state for box
       if options[:box_color] || (options[:border_width].to_i > 0 && options[:border_color])
         # compute x and y position for text
@@ -290,7 +291,7 @@ module CombinePDF
       # reset x,y by text alignment - x,y are calculated from the bottom left
       # each unit (1) is 1/72 Inch
       # create text stream
-      text_stream = ''
+      text_stream = +''
       if !text.to_s.empty? && options[:font_size] != 0 && (options[:font_color] || options[:stroke_color])
         # compute x and y position for text
         x = options[:x] + (options[:width] * options[:text_padding])
@@ -679,7 +680,7 @@ module CombinePDF
       insert_content 'Q'
 
       # Prep content
-      @contents = ''
+      @contents = +''
       insert_content @contents
       @contents
     end
@@ -788,7 +789,7 @@ module CombinePDF
           # add to array
           if out.last.nil? || out.last[0] != fonts[i]
             out.last[1] << '>' unless out.last.nil?
-            out << [fonts[i], '<', 0, 0]
+            out << [fonts[i], (+'<'), 0, 0]
           end
           out.last[1] << (fonts_array[i].cmap.nil? ? (c.unpack('H*')[0]) : fonts_array[i].cmap[c])
           if fonts_array[i].metrics[c]

data/lib/combine_pdf/parser.rb

@@ -33,7 +33,7 @@ module CombinePDF
     # they are mainly to used to know if the file is (was) encrypted and to get more details.
     attr_reader :info_object, :root_object, :names_object, :forms_object, :outlines_object, :metadata
 
-    attr_reader :allow_optional_content
+    attr_reader :allow_optional_content, :raise_on_encrypted
     # when creating a parser, it is important to set the data (String) we wish to parse.
     #
     # <b>the data is required and it is not possible to set the data at a later stage</b>
@@ -58,6 +58,7 @@ module CombinePDF
       @version = nil
       @scanner = nil
       @allow_optional_content = options[:allow_optional_content]
+      @raise_on_encrypted = options[:raise_on_encrypted]
     end
 
     # parse the data in the new parser (the data already set through the initialize / new method)
@@ -96,6 +97,7 @@ module CombinePDF
       end
 
       if @root_object[:Encrypt]
+        raise EncryptionError, 'the file is encrypted' if @raise_on_encrypted
         # change_references_to_actual_values @root_object
         warn 'PDF is Encrypted! Attempting to decrypt - not yet fully supported.'
         decryptor = PDFDecrypt.new @parsed, @root_object
@@ -233,16 +235,18 @@ module CombinePDF
         # all characters that aren't white space or special: /[^\x00\x09\x0a\x0c\x0d\x20\x28\x29\x3c\x3e\x5b\x5d\x7b\x7d\x2f\x25]+
         elsif str = @scanner.scan(/\/[^\x00\x09\x0a\x0c\x0d\x20\x28\x29\x3c\x3e\x5b\x5d\x7b\x7d\x2f\x25]*/)
           out << (str[1..-1].gsub(/\#[0-9a-fA-F]{2}/) { |a| a[1..2].hex.chr }).to_sym
+          # warn "CombinePDF detected name: #{out.last.to_s}"
         ##########################################
         ## Parse a Number
         ##########################################
         elsif str = @scanner.scan(/[\+\-\.\d]+/)
           str =~ /\./ ? (out << str.to_f) : (out << str.to_i)
+          # warn "CombinePDF detected number: #{out.last.to_s}"
         ##########################################
         ## parse a Hex String
         ##########################################
         elsif str = @scanner.scan(/\<[0-9a-fA-F]*\>/)
-          # warn "Found a hex string"
+          # warn "Found a hex string #{str}"
           str = str.slice(1..-2).force_encoding(Encoding::ASCII_8BIT)
           # str = "0#{str}" if str.length.odd?
           out << unify_string([str].pack('H*').force_encoding(Encoding::ASCII_8BIT))
@@ -258,7 +262,7 @@ module CombinePDF
         ##########################################
         elsif @scanner.scan(/\(/)
           # warn "Found a literal string"
-          str = ''.force_encoding(Encoding::ASCII_8BIT)
+          str = ''.b
           count = 1
           while count > 0 && @scanner.rest?
             scn = @scanner.scan_until(/[\(\)]/)
@@ -319,8 +323,8 @@ module CombinePDF
                 str << 12
               when 48..57 # octal notation for byte?
                 rep -= 48
-                rep = (rep << 3) + (str_bytes.shift-48) if str_bytes[0].between?(48, 57)
-                rep = (rep << 3) + (str_bytes.shift-48) if str_bytes[0].between?(48, 57) && (((rep << 3) + (str_bytes[0] - 48)) <= 255)
+                rep = (rep << 3) + (str_bytes.shift-48) if str_bytes[0]&.between?(48, 57)
+                rep = (rep << 3) + (str_bytes.shift-48) if str_bytes[0]&.between?(48, 57) && (((rep << 3) + (str_bytes[0] - 48)) <= 255)
                 str << rep
               when 10 # new line, ignore
                 str_bytes.shift if str_bytes[0] == 13
@@ -336,6 +340,7 @@ module CombinePDF
             end
           end
           out << unify_string(str.pack('C*').force_encoding(Encoding::ASCII_8BIT))
+          # warn "Found Literal String: #{out.last}"
         ##########################################
         ## parse a Dictionary
         ##########################################
@@ -348,29 +353,42 @@ module CombinePDF
         ## return content of array or dictionary
         ##########################################
         elsif @scanner.scan(/\]/) || @scanner.scan(/>>/)
+          # warn "Dictionary / Array ended with #{@scanner.peek(5)}"
           return out
         ##########################################
         ## parse a Stream
         ##########################################
         elsif @scanner.scan(/stream[ \t]*[\r\n]/)
           @scanner.pos += 1 if @scanner.peek(1) == "\n".freeze && @scanner.matched[-1] != "\n".freeze
+          # advance by the publshed stream length (if any)
+          old_pos = @scanner.pos
+          if(out.last.is_a?(Hash) && out.last[:Length].is_a?(Integer) && out.last[:Length] > 2)
+            @scanner.pos += out.last[:Length] - 2
+          end
+
           # the following was dicarded because some PDF files didn't have an EOL marker as required
           # str = @scanner.scan_until(/(\r\n|\r|\n)endstream/)
           # instead, a non-strict RegExp is used:
-          str = @scanner.scan_until(/endstream/)
+
 
           # raise error if the stream doesn't end.
-          unless str
+          unless @scanner.skip_until(/endstream/)
             raise ParsingError, "Parsing Error: PDF file error - a stream object wasn't properly closed using 'endstream'!"
           end
+          length = @scanner.pos - (old_pos + 9)
+          length = 0 if(length < 0)
+          length -= 1 if(@scanner.string[old_pos + length - 1] == "\n") 
+          length -= 1 if(@scanner.string[old_pos + length - 1] == "\r") 
+          str = (length > 0) ? @scanner.string.slice(old_pos, length) : +''
+
+          # warn "CombinePDF parser: detected Stream #{str.length} bytes long #{str[0..3]}...#{str[-4..-1]}"
 
           # need to remove end of stream
           if out.last.is_a? Hash
-            # out.last[:raw_stream_content] = str[0...-10] #cuts only one EON char (\n or \r)
-            out.last[:raw_stream_content] = unify_string str.sub(/(\r\n|\n|\r)?endstream\z/, '').force_encoding(Encoding::ASCII_8BIT)
+            out.last[:raw_stream_content] = unify_string str.force_encoding(Encoding::ASCII_8BIT)
           else
             warn 'Stream not attached to dictionary!'
-            out << str.sub(/(\r\n|\n|\r)?endstream\z/, '').force_encoding(Encoding::ASCII_8BIT)
+            out << str.force_encoding(Encoding::ASCII_8BIT)
           end
         ##########################################
         ## parse an Object after finished
@@ -528,6 +546,14 @@ module CombinePDF
               inheritance_hash[:Resources] ||= { referenced_object: {}, is_reference_only: true }.dup
               (inheritance_hash[:Resources][:referenced_object] || inheritance_hash[:Resources]).update((catalogs[:Resources][:referenced_object] || catalogs[:Resources]), &HASH_UPDATE_PROC_FOR_OLD)
             end
+            if catalogs[:ProcSet].is_a?(Array)
+              if(inheritance_hash[:ProcSet])
+                inheritance_hash[:ProcSet][:referenced_object].concat(catalogs[:ProcSet])
+                inheritance_hash[:ProcSet][:referenced_object].uniq!
+              else
+                inheritance_hash[:ProcSet] ||= { referenced_object: catalogs[:ProcSet], is_reference_only: true }.dup
+              end
+            end
             if catalogs[:ColorSpace]
               inheritance_hash[:ColorSpace] ||= { referenced_object: {}, is_reference_only: true }.dup
               (inheritance_hash[:ColorSpace][:referenced_object] || inheritance_hash[:ColorSpace]).update((catalogs[:ColorSpace][:referenced_object] || catalogs[:ColorSpace]), &HASH_UPDATE_PROC_FOR_OLD)
@@ -556,6 +582,18 @@ module CombinePDF
               catalogs[:ColorSpace] = { referenced_object: catalogs[:ColorSpace], is_reference_only: true } unless catalogs[:ColorSpace][:referenced_object]
               catalogs[:ColorSpace][:referenced_object].update((inheritance_hash[:ColorSpace][:referenced_object] || inheritance_hash[:ColorSpace]), &HASH_UPDATE_PROC_FOR_OLD)
             end
+            if inheritance_hash[:ProcSet]
+              if(catalogs[:ProcSet])
+                if catalogs[:ProcSet].is_a?(Array)
+                  catalogs[:ProcSet] = { referenced_object: catalogs[:ProcSet], is_reference_only: true }
+                end
+                catalogs[:ProcSet][:referenced_object].concat(inheritance_hash[:ProcSet][:referenced_object])
+                catalogs[:ProcSet][:referenced_object].uniq!
+              else
+                catalogs[:ProcSet] = { is_reference_only: true }.dup
+                catalogs[:ProcSet][:referenced_object] = []
+              end
+            end
             # (catalogs[:ColorSpace] ||= {}).update(inheritance_hash[:ColorSpace], &HASH_UPDATE_PROC_FOR_OLD) if inheritance_hash[:ColorSpace]
             # catalogs[:Order] ||= inheritance_hash[:Order] if inheritance_hash[:Order]
             # catalogs[:AS] ||= inheritance_hash[:AS] if inheritance_hash[:AS]
@@ -594,17 +632,17 @@ module CombinePDF
     #
     def serialize_objects_and_references
       obj_dir = {}
-      objid_cache = {}
+      objid_cache = {}.compare_by_identity
       # create a dictionary for referenced objects (no value resolution at this point)
       # at the same time, delete duplicates and old versions when objects have multiple versions
       @parsed.uniq!
       @parsed.length.times do |i|
         o = @parsed[i]
-        objid_cache[o.object_id] = i
+        objid_cache[o] = i
         tmp_key = [o[:indirect_reference_id], o[:indirect_generation_number]]
         if tmp_found = obj_dir[tmp_key]
           tmp_found.clear
-          @parsed[objid_cache[tmp_found.object_id]] = nil
+          @parsed[objid_cache[tmp_found]] = nil
         end
         obj_dir[tmp_key] = o
       end
@@ -727,9 +765,9 @@ module CombinePDF
     # end
 
     # # run block of code on evey PDF object (PDF objects are class Hash)
-    # def each_object(object, limit_references = true, already_visited = {}, &block)
+    # def each_object(object, limit_references = true, already_visited = {}.compare_by_identity, &block)
     # 	unless limit_references
-    # 		already_visited[object.object_id] = true
+    # 		already_visited[object] = true
     # 	end
     # 	case
     # 	when object.is_a?(Array)
@@ -738,7 +776,7 @@ module CombinePDF
     # 		yield(object)
     # 		unless limit_references && object[:is_reference_only]
     # 			object.each do |k,v|
-    # 				each_object(v, limit_references, already_visited, &block) unless already_visited[v.object_id]
+    # 				each_object(v, limit_references, already_visited, &block) unless already_visited[v]
     # 			end
     # 		end
     # 	end