lib/combine_pdf/parser.rb in combine_pdf-1.0.26 vs lib/combine_pdf/parser.rb in combine_pdf-1.0.27

@@ -260,11 +260,11 @@
         ##########################################
         ## parse a Literal String
         ##########################################
         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(/[\(\)]/)
             unless scn
               warn "Unknown error parsing string at #{@scanner.pos} for string: #{str}!"
@@ -367,21 +367,21 @@
           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:
-          
 
+
           # raise error if the stream doesn't end.
           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) : ''
+          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
@@ -630,21 +630,21 @@
     #
     # should be moved to the parser's workflow.
     #
     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
       @parsed.compact!
       objid_cache.clear
@@ -763,21 +763,21 @@
     #     new_data
     #   end
     # 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)
     # 		object.each {|obj| each_object(obj, limit_references, already_visited, &block)}
     # 	when object.is_a?(Hash)
     # 		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
     # end
   end