pdf-reader 0.8.2 → 0.8.3

data/CHANGELOG

@@ -1,3 +1,7 @@
+v0.8.3 (14th February 2010)
+- Fix a bug in tokenising of hex strings inside dictionaries
+  - Thanks to Brad Ediger for detecting the issue and proposing a solution
+
 v0.8.2 (1st January 2010)
 - Fix parsing of files that use Form XObjects behind an indirect reference
   (thanks Cornelius Illi and Patrick Crosby)

data/Rakefile

@@ -6,7 +6,7 @@ require 'rake/testtask'
 require "rake/gempackagetask"
 require 'spec/rake/spectask'
 
-PKG_VERSION = "0.8.2"
+PKG_VERSION = "0.8.3"
 PKG_NAME = "pdf-reader"
 PKG_FILE_NAME = "#{PKG_NAME}-#{PKG_VERSION}"
 
@@ -57,27 +57,25 @@ end
 # RSpec files aren't included, as they depend on the PDF files,
 # which will make the gem filesize irritatingly large
 spec = Gem::Specification.new do |spec|
-	spec.name = PKG_NAME
-	spec.version = PKG_VERSION
-	spec.platform = Gem::Platform::RUBY
-	spec.summary = "A library for accessing the content of PDF files"
-	spec.files =  Dir.glob("{examples,lib}/**/**/*") +
-                      ["Rakefile"]
-  
+  spec.name = PKG_NAME
+  spec.version = PKG_VERSION
+  spec.platform = Gem::Platform::RUBY
+  spec.summary = "A library for accessing the content of PDF files"
+  spec.files =  Dir.glob("{examples,lib}/**/**/*") + ["Rakefile"]
   spec.require_path = "lib"
   spec.bindir = "bin"
   spec.executables << "pdf_object"
   spec.executables << "pdf_text"
   spec.executables << "pdf_list_callbacks"
-	spec.has_rdoc = true
-	spec.extra_rdoc_files = %w{README.rdoc TODO CHANGELOG MIT-LICENSE }
-	spec.rdoc_options << '--title' << 'PDF::Reader Documentation' <<
-	                     '--main'  << 'README.rdoc' << '-q'
-  spec.author = "Peter Jones"
-	spec.email = "pjones@pmade.com"
-	spec.rubyforge_project = "pdf-reader"
-	spec.homepage = "http://software.pmade.com/pdfreader"
-	spec.description = "The PDF::Reader library implements a PDF parser conforming as much as possible to the PDF specification from Adobe"
+  spec.has_rdoc = true
+  spec.extra_rdoc_files = %w{README.rdoc TODO CHANGELOG MIT-LICENSE }
+  spec.rdoc_options << '--title' << 'PDF::Reader Documentation' <<
+                       '--main'  << 'README.rdoc' << '-q'
+  spec.author = "James Healy"
+  spec.email = "jimmy@deefa.com"
+  spec.rubyforge_project = "pdf-reader"
+  spec.homepage = "http://github.com/yob/pdf-reader"
+  spec.description = "The PDF::Reader library implements a PDF parser conforming as much as possible to the PDF specification from Adobe"
   spec.add_dependency('Ascii85', '>=0.9')
 end
 

data/lib/pdf/reader.rb

@@ -137,8 +137,7 @@ class PDF::Reader
   ################################################################################
   # Given an IO object that contains PDF data, return the contents of a single object
   def object (io, id, gen)
-    @buffer   = Buffer.new(io)
-    @xref     = XRef.new(@buffer)
+    @xref     = XRef.new(io)
     @xref.load
 
     @xref.object(Reference.new(id, gen))

data/lib/pdf/reader/buffer.rb

@@ -27,7 +27,7 @@
 
 class PDF::Reader
 
-  # A string tokeniser that recognises PDF grammer. When passed an IO stream or a
+  # A string tokeniser that recognises PDF grammar. When passed an IO stream or a
   # string, repeated calls to token() will return the next token from the source.
   #
   # This is very low level, and getting the raw tokens is not very useful in itself.
@@ -99,7 +99,7 @@ class PDF::Reader
     end
 
     # return raw bytes from the underlying IO stream. All bytes up to the first
-    # occurance of needle will be returned. The match (if any) is not returned.
+    # occurrence of needle will be returned. The match (if any) is not returned.
     # The IO stream cursor is left on the first byte of the match.
     #
     #   needle - a string to search the IO stream for
@@ -129,7 +129,7 @@ class PDF::Reader
       reset_pos
       prepare_tokens if @tokens.size < 3
       merge_indirect_reference
-      merge_tokens
+      prepare_tokens if @tokens.size < 3
 
       @tokens.shift
     end
@@ -206,10 +206,14 @@ class PDF::Reader
     # them, replace the tokens with a PDF::Reader::Reference instance.
     #
     # Merging them into a single string was another option, but that would mean
-    # code further up the stact would need to check every token  to see if it looks
+    # code further up the stack would need to check every token  to see if it looks
     # like an indirect object. For optimisation reasons, I'd rather avoid
     # that extra check.
     #
+    # It's incredibly likely that the next 3 tokens in the buffer are NOT an
+    # indirect reference, so test for that case first and avoid the relatively
+    # expensive regexp checks if possible.
+    #
     def merge_indirect_reference
       return if @tokens.size < 3
       return if @tokens[2] != "R"
@@ -222,26 +226,8 @@ class PDF::Reader
       end
     end
 
-    # merge any consequtive tokens that are actually 1 token. The only current
-    # time this is the case is << and >>. < and > are valid tokens (they indicate
-    # a hex string) but so are << and >> (they indicate a dictionary).
-    #
-    def merge_tokens
-      @tokens.each_with_index do |tok, idx|
-        if tok == "<" && @tokens[idx+1] == "<"
-          @tokens.inspect
-          @tokens[idx] = "<<"
-          @tokens[idx+1] = nil
-        elsif tok == ">" && @tokens[idx+1] == ">"
-          @tokens[idx] = ">>"
-          @tokens[idx+1] = nil
-        end
-      end
-      @tokens.compact!
-    end
-
     # if we're currently inside a literal string we more or less just read bytes until
-    # we find the closes ) delimiter. Lots of bytes that would otherwise indicate the
+    # we find the closing ) delimiter. Lots of bytes that would otherwise indicate the
     # start of a new token in regular mode are left untouched when inside a literal
     # string.
     #
@@ -294,13 +280,27 @@ class PDF::Reader
           @tokens << tok if tok.size > 0
           tok = ""
           break
-        when "\x28", "\x3C", "\x5B", "\x7B", "\x2F"
+        when "\x3C"
+          # opening delimiter '<', start of new token
+          @tokens << tok if tok.size > 0
+          chr << @io.read(1) if peek_char == "\x3C" # check if token is actually '<<'
+          @tokens << chr
+          tok = ""
+          break
+        when "\x3E"
+          # closing delimiter '>', start of new token
+          @tokens << tok if tok.size > 0
+          chr << @io.read(1) if peek_char == "\x3E" # check if token is actually '>>'
+          @tokens << chr
+          tok = ""
+          break
+        when "\x28", "\x5B", "\x7B", "\x2F"
           # opening delimiter, start of new token
           @tokens << tok if tok.size > 0
           @tokens << chr
           tok = ""
           break
-        when "\x29", "\x3E", "\x5D", "\x7D"
+        when "\x29", "\x5D", "\x7D"
           # closing delimiter
           @tokens << tok if tok.size > 0
           @tokens << chr
@@ -313,5 +313,14 @@ class PDF::Reader
 
       @tokens << tok if tok.size > 0
     end
+
+    # peek at the next character in the io stream, leaving the stream position
+    # untouched
+    #
+    def peek_char
+      chr = @io.read(1)
+      @io.seek(-1, IO::SEEK_CUR) unless chr.nil?
+      chr
+    end
   end
 end

metadata

@@ -1,15 +1,15 @@
 --- !ruby/object:Gem::Specification 
 name: pdf-reader
 version: !ruby/object:Gem::Version 
-  version: 0.8.2
+  version: 0.8.3
 platform: ruby
 authors: 
-- Peter Jones
+- James Healy
 autorequire: 
 bindir: bin
 cert_chain: []
 
-date: 2010-01-01 00:00:00 +11:00
+date: 2010-02-14 00:00:00 +11:00
 default_executable: 
 dependencies: 
 - !ruby/object:Gem::Dependency 
@@ -23,7 +23,7 @@ dependencies:
         version: "0.9"
     version: 
 description: The PDF::Reader library implements a PDF parser conforming as much as possible to the PDF specification from Adobe
-email: pjones@pmade.com
+email: jimmy@deefa.com
 executables: 
 - pdf_object
 - pdf_text
@@ -76,7 +76,7 @@ files:
 - CHANGELOG
 - MIT-LICENSE
 has_rdoc: true
-homepage: http://software.pmade.com/pdfreader
+homepage: http://github.com/yob/pdf-reader
 licenses: []
 
 post_install_message: