combine_pdf 1.0.1 → 1.0.2

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: dab047c1632f06fe95ade20bd50476f66260c877
-  data.tar.gz: 32fb4875dfd46b48f50e262a4f0a00d8522861a2
+  metadata.gz: bb9d21d934a1001210f09268503f3ebc0282aedf
+  data.tar.gz: fc180c13f7789f87e32e06f0307093a485dfe560
 SHA512:
-  metadata.gz: 955acdc11d89d5e1f5d5033bfaa8438633f4068bc09ddadb3e77852e0c5cc6544ee4b38f4c67f6a3b24fdfec334c4e1ce4fb3cc6e257b11bd9584832138b7c37
-  data.tar.gz: f70ac4f27e87cf913b52be0dc60eb90e32002bedd1bb30cb2c4ee9452866d66e7d30828a516f20ab291effd2dfe0404b0fad9b649315c66f455b62effd9aee95
+  metadata.gz: 7b9d84a43b214ab50f9c9aca37ddb4d1532f2c72d5ec8b441119f0bfb33e2b295267def67acc6787254cd0760f8e44ba358054cd311eb6619a9d5edc125095b6
+  data.tar.gz: 7bed5345b025c0633c2cad6bd7b861d34e962f17f002d840f99a9159ec7d5d3529ebfff10026fe8d082bd47c4eeafaa404fedd828dd9ed5ff526c676d0a7c22c

data/CHANGELOG.md

@@ -1,5 +1,9 @@
 # Change Log
 
+**Fix**: Fix NilError when calling `fonts` for a page that has no fonts. Credit to Pavel Slabý (@paulslaby) for PR#110.
+
+**Fix**: Fix issue #109 where nested differences between objects weren't detected properly, causing loss of data if objects (specifically images that use image masks) would merge. The fix implements a manual equality checks with up to three (3) levels of recursion, protecting against stack overflow that can be caused by the combinations of complex PDF files and Ruby's limitless recursion on `eql?`. Credit to Ryan Scott (@Subtletree) exposing the issue.
+
 ***
 
 #### Change log v.1.0.1

data/README.md

@@ -18,7 +18,7 @@ Quick rundown:
 
 * When reading PDF Forms, some form data might be lost. I tried fixing this to the best of my ability, but I'm not sure it all works just yet.
 
-* When combining PDF Forms, form data might be unified. I couldn't fix this because this is how PDF forms work (filling a feild fills in the data in any field with the same name), but frankly, I kinda liked the issue... it's almost a feature.
+* When combining PDF Forms, form data might be unified. I couldn't fix this because this is how PDF forms work (filling a field fills in the data in any field with the same name), but frankly, I kinda liked the issue... it's almost a feature.
 
 * When unifying the same TOC data more then once, one of the references will be unified with the other (meaning that if the pages look the same, both references will link to the same page instead of linking to two different pages). You can fix this by adding content to the pages before merging the PDF files (i.e. add empty text boxes to all the pages).
 

data/lib/combine_pdf/pdf_protected.rb

@@ -33,8 +33,12 @@ module CombinePDF
         if obj.is_a?(Hash)
           referenced = obj[:referenced_object]
           if referenced && referenced.any?
-            #         tmp = resolved[referenced.object_id] || existing[referenced]
-            tmp = resolved[referenced.object_id] || (referenced[:raw_stream_content] && existing[referenced[:raw_stream_content]])
+            tmp = resolved[referenced.object_id]
+            if !tmp && referenced[:raw_stream_content]
+              tmp = existing[referenced[:raw_stream_content]]
+              # Avoid endless recursion by limiting it to a number of layers (default == 2)
+              tmp = nil unless equal_layers(tmp, referenced)
+            end
             if tmp
               obj[:referenced_object] = tmp
             else
@@ -339,6 +343,21 @@ module CombinePDF
 
     private
 
+    def equal_layers obj1, obj2, layer = 3
+      return true   if(layer == 0)
+      if obj1.is_a? Hash
+        return false unless obj2.is_a? Hash
+        keys = obj1.keys;
+        return false if (keys - obj2.keys).any?
+        keys.each {|k| return false unless equal_layers( obj1[k], obj2[k], layer-1) }
+      elsif obj1.is_a? Array
+        return false unless obj2.is_a? Array
+        (obj1-obj2).any?
+      else
+        obj1 == obj2
+      end
+    end
+
     def renaming_dictionary(object = nil, dictionary = {})
       object ||= @names
       case object

data/lib/combine_pdf/pdf_public.rb

@@ -253,10 +253,12 @@ module CombinePDF
     def fonts(limit_to_type0 = false)
       fonts_array = []
       pages.each do |p|
-        p[:Resources][:Font].values.each do |f|
-          f = f[:referenced_object] if f[:referenced_object]
-          if (limit_to_type0 || f[:Subtype] = :Type0) && f[:Type] == :Font && !fonts_array.include?(f)
-            fonts_array << f
+        if p[:Resources][:Font]
+          p[:Resources][:Font].values.each do |f|
+            f = f[:referenced_object] if f[:referenced_object]
+            if (limit_to_type0 || f[:Subtype] = :Type0) && f[:Type] == :Font && !fonts_array.include?(f)
+              fonts_array << f
+            end
           end
         end
       end

data/lib/combine_pdf/version.rb

@@ -1,3 +1,3 @@
 module CombinePDF
-  VERSION = '1.0.1'.freeze
+  VERSION = '1.0.2'.freeze
 end

data/test/automated

@@ -33,6 +33,9 @@ pdf.save '03_01_nil_value_conflict.pdf'
 pdf = CombinePDF.load './Ruby/test pdfs/space_after_streram_keyword.pdf'
 pdf.save '03_02_extra_space_after_stream_keyword.pdf'
 
+pdf = CombinePDF.load './Ruby/test pdfs/nested_difference.pdf'
+pdf.save '03_03_nested_difference.pdf'
+
 pdf = CombinePDF.load './Ruby/test pdfs/names_go_haywire_0.pdf'
 pdf << CombinePDF.load('./Ruby/test pdfs/names_go_haywire_1.pdf')
 pdf.save '04_check_view_and_names_reference.pdf'

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: combine_pdf
 version: !ruby/object:Gem::Version
-  version: 1.0.1
+  version: 1.0.2
 platform: ruby
 authors:
 - Boaz Segev
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2017-05-30 00:00:00.000000000 Z
+date: 2017-06-13 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: ruby-rc4