diff --git a/lib/linguist/tokenizer.rb b/lib/linguist/tokenizer.rb
index f5f12981..0df2c56f 100644
--- a/lib/linguist/tokenizer.rb
+++ b/lib/linguist/tokenizer.rb
@@ -74,18 +74,22 @@ module Linguist
           s.skip_until(/[^\\]'/)
 
         # Skip number literals
-        elsif s.scan(/\d+/)
+        elsif s.scan(/(0x)?\d+/)
 
         # SGML style brackets
-        elsif token = s.scan(/<[^>]+>/)
+        elsif token = s.scan(/<[^\s<>][^<>]*>/)
           extract_sgml_tokens(token).each { |t| tokens << t }
 
         # Common programming punctuation
-        elsif token = s.scan(/;|\{|\}|\(|\)|<<?|\+/)
+        elsif token = s.scan(/;|\{|\}|\(|\)/)
           tokens << token
 
         # Regular token
-        elsif token = s.scan(/[\w\.@#\/]+/)
+        elsif token = s.scan(/[\w\.@#\/\*]+/)
+          tokens << token
+
+        # Common operators
+        elsif token = s.scan(/<<?|\+|\-|\*|\/|&&?|\|\|?/)
           tokens << token
 
         else
diff --git a/test/test_tokenizer.rb b/test/test_tokenizer.rb
index cea4e535..d38562c8 100644
--- a/test/test_tokenizer.rb
+++ b/test/test_tokenizer.rb
@@ -25,6 +25,7 @@ class TestTokenizer < Test::Unit::TestCase
   def test_skip_number_literals
     assert_equal %w(+), tokenize('1 + 1')
     assert_equal %w(add \( \)), tokenize('add(123, 456)')
+    assert_equal %w(|), tokenize('0x01 | 0x10')
   end
 
   def test_skip_comments
@@ -47,20 +48,33 @@ class TestTokenizer < Test::Unit::TestCase
     assert_equal %w(<?xml> version=), tokenize("<?xml version=\"1.0\"?>")
   end
 
+  def test_operators
+    assert_equal %w(+), tokenize("1 + 1")
+    assert_equal %w(-), tokenize("1 - 1")
+    assert_equal %w(*), tokenize("1 * 1")
+    assert_equal %w(/), tokenize("1 / 1")
+    assert_equal %w(&), tokenize("1 & 1")
+    assert_equal %w(&&), tokenize("1 && 1")
+    assert_equal %w(|), tokenize("1 | 1")
+    assert_equal %w(||), tokenize("1 || 1")
+    assert_equal %w(<), tokenize("1 < 0x01")
+    assert_equal %w(<<), tokenize("1 << 0x01")
+  end
+
   def test_c_tokens
     assert_equal %w(#ifndef HELLO_H #define HELLO_H void hello \( \) ; #endif), tokenize(:"c/hello.h")
     assert_equal %w(#include <stdio.h> int main \( \) { printf \( \) ; return ; }), tokenize(:"c/hello.c")
   end
 
   def test_cpp_tokens
-    assert_equal %w(class Bar { protected char name ; public void hello \( \) ; }), tokenize(:"cpp/bar.h")
+    assert_equal %w(class Bar { protected char *name ; public void hello \( \) ; }), tokenize(:"cpp/bar.h")
     assert_equal %w(#include <iostream> using namespace std ; int main \( \) { cout << << endl ; }), tokenize(:"cpp/hello.cpp")
   end
 
   def test_objective_c_tokens
     assert_equal %w(#import <Foundation/Foundation.h> @interface Foo NSObject { } @end), tokenize(:"objective-c/Foo.h")
     assert_equal %w(#import @implementation Foo @end), tokenize(:"objective-c/Foo.m")
-    assert_equal %w(#import <Cocoa/Cocoa.h> int main \( int argc char argv \) { NSLog \( @ \) ; return ; }), tokenize(:"objective-c/hello.m")
+    assert_equal %w(#import <Cocoa/Cocoa.h> int main \( int argc char *argv \) { NSLog \( @ \) ; return ; }), tokenize(:"objective-c/hello.m")
   end
 
   def test_javascript_tokens