update w3c import script to actually work :).

Tools/Scripts/webkitpy/w3c/test_parser.py

 #!/usr/bin/env python
 
 # Copyright (C) 2013 Adobe Systems Incorporated. All rights reserved.
 #
 # Redistribution and use in source and binary forms, with or without
 # modification, are permitted provided that the following conditions
 # are met:
 #
 # 1. Redistributions of source code must retain the above
 #    copyright notice, this list of conditions and the following
@@ skipping 29 matching lines @@
         self.filename = filename
         self.host = Host()
         self.filesystem = self.host.filesystem
 
         self.test_doc = None
         self.ref_doc = None
         self.load_file(filename)
 
     def load_file(self, filename):
         if self.filesystem.exists(filename):
-            self.test_doc = Parser(self.filesystem.read_text_file(filename))
+            self.test_doc = Parser(self.filesystem.read_binary_file(filename))
         else:
             self.test_doc = None
         self.ref_doc = None
 
     def analyze_test(self, test_contents=None, ref_contents=None):
         """ Analyzes a file to determine if it's a test, what type of test, and what reference or support files it requires. Returns all of the test info """
 
         test_info = None
 
         if test_contents is None and self.test_doc is None:
@@ skipping 24 matching lines @@
             #
             # Note: The test files themselves are not checked for support files
             # outside their directories as the convention in the CSSWG is to
             # put all support files in the same dir or subdir as the test.
             #
             # All non-test files in the test's directory tree are normally
             # copied as part of the import as they are assumed to be required
             # support files.
             #
             # *But*, there is exactly one case in the entire css2.1 suite where
-            # at test depends on a file that lives in a different directory,
+            # a test depends on a file that lives in a different directory,
             # which depends on another file that lives outside of its
             # directory. This code covers that case :)
             if matches[0]['href'].startswith('..'):
                 support_files = self.support_files(self.ref_doc)
                 test_info['refsupport'] = support_files
 
         elif self.is_jstest():
             test_info = {'test': self.filename, 'jstest': True}
         elif self.options['all'] is True and not('-ref' in self.filename) and not('reference' in self.filename):
             test_info = {'test': self.filename}
@@ skipping 14 matching lines @@
         if doc is None:
             return support_files
 
         elements_with_src_attributes = doc.findAll(src=re.compile('.*'))
         elements_with_href_attributes = doc.findAll(href=re.compile('.*'))
 
         url_pattern = re.compile('url\(.*\)')
         urls = []
         for url in doc.findAll(text=url_pattern):
             url = re.search(url_pattern, url)
-            url = re.sub('url\([\'\"]', '', url.group(0))
-            url = re.sub('[\'\"]\)', '', url)
+            url = re.sub('url\([\'\"]?', '', url.group(0))
+            url = re.sub('[\'\"]?\)', '', url)
             urls.append(url)
 
         src_paths = [src_tag['src'] for src_tag in elements_with_src_attributes]
         href_paths = [href_tag['href'] for href_tag in elements_with_href_attributes]
 
         paths = src_paths + href_paths + urls
         for path in paths:
             if not(path.startswith('http:')) and not(path.startswith('mailto:')):
                 support_files.append(path)
 
         return support_files