combine_pdf 1.0.20 → 1.0.21

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: b482b2bf36c858f6bb603a66536ee17682b21bc78381fa86a115a4e437f9160c
-  data.tar.gz: 29ab4a06d5ec597b806feeb346a2e972e98566442606f6e3e765348e5fee8ea4
+  metadata.gz: 6ee7f4bb8ca8b2a51a7f42251d9a3faa6c1d91d19ed3d91db8ce470a1b1995a9
+  data.tar.gz: 5be4868dcb9f1b834fc12e5316d5a46b6fee97104c8509cbcdb13908b499d79f
 SHA512:
-  metadata.gz: 615c0917cab8ad406d1eb61a81528656e596bc91f88ec96d8662446496bf9e465df917f3fcf1a9aa9ba8b3057e7ec8b64dc7a0799ca53597d3d071e3afc1fd36
-  data.tar.gz: 045ef9bc9ab9a29a1665df621f1f9f41e2744f6b285b68f6c76ec64e05493f29b423cfb1433c903267e5893d53de71558bf677912551e13fe86ccb66e827f89e
+  metadata.gz: 751d04258815d32076eccc5bb2905cfed641b2942721900dfa5826b0cf902a35bbf60178501a81617ce0979b1be87554666320423c4066b375dced68927e8aea
+  data.tar.gz: 7171c0c987646b6336984fff497fa40f02ae7b6079cd8a29b58cb7659c1b527179b44f0c7210edeb63dcc931c634320ae93d0513579852ee28beca23896a085e

data/CHANGELOG.md

@@ -2,6 +2,9 @@
 
 ***
 
+#### 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
 

data/lib/combine_pdf/parser.rb

@@ -358,25 +358,35 @@ module CombinePDF
         ##########################################
         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].to_i > 2)
+            @scanner.pos += out.last[:Length].to_i - 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

data/lib/combine_pdf/version.rb

@@ -1,3 +1,3 @@
 module CombinePDF
-  VERSION = '1.0.20'.freeze
+  VERSION = '1.0.21'.freeze
 end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: combine_pdf
 version: !ruby/object:Gem::Version
-  version: 1.0.20
+  version: 1.0.21
 platform: ruby
 authors:
 - Boaz Segev
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2020-12-07 00:00:00.000000000 Z
+date: 2020-12-12 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: ruby-rc4