update w3c import script to actually work :).

Tools/Scripts/webkitpy/w3c/test_importer.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 54 matching lines @@
       This can also be overridden by a -n or --no-overwrite flag
 
     - All files are converted to work in WebKit:
          1. .xht extensions are changed to .xhtml to make new-run-webkit-tests happy
          2. Paths to testharness.js files are modified point to Webkit's copy of them in
             LayoutTests/resources, using the correct relative path from the new location
          3. All CSS properties requiring the -webkit-vendor prefix are prefixed - this current
             list of what needs prefixes is read from Source/WebCore/CSS/CSSProperties.in
          4. Each reftest has its own copy of its reference file following the naming conventions
             new-run-webkit-tests expects
-         5. If a a reference files lives outside the directory of the test that uses it, it is checked
+         5. If a reference files lives outside the directory of the test that uses it, it is checked
             for paths to support files as it will be imported into a different relative position to the
             test file (in the same directory)
 
      - Upon completion, script outputs the total number tests imported, broken down by test type
 
      - Also upon completion, each directory where files are imported will have w3c-import.log written
        with a timestamp, the W3C Mercurial changeset if available, the list of CSS properties used that
        require prefixes, the list of imported files, and guidance for future test modification and
        maintenance.
 
@@ skipping 92 matching lines @@
             self.changeset = CHANGESET_NOT_AVAILABLE
 
     def find_importable_tests(self, directory):
         # FIXME: use filesystem
         for root, dirs, files in os.walk(directory):
             print 'Scanning ' + root + '...'
             total_tests = 0
             reftests = 0
             jstests = 0
 
-            # Ignore any repo stuff
-            if '.git' in dirs:
-                dirs.remove('.git')
-            if '.hg' in dirs:
-                dirs.remove('.hg')
-
-            # archive and data dirs are internal csswg things that live in every approved directory
-            if 'data' in dirs:
-                dirs.remove('data')
-            if 'archive' in dirs:
-                dirs.remove('archive')
+            # "archive" and "data" dirs are internal csswg things that live in every approved directory.
+            DIRS_TO_SKIP = ('.git', '.hg', 'data', 'archive')
+            for d in DIRS_TO_SKIP:
+                if d in dirs:
+                    dirs.remove(d)
 
             copy_list = []
 
             for filename in files:
                 # FIXME: This block should really be a separate function, but the early-continues make that difficult.
 
                 if filename.startswith('.') or filename.endswith('.pl'):
                     continue  # For some reason the w3c repo contains random perl scripts we don't care about.
 
                 fullpath = os.path.join(root, filename)
@@ skipping 42 matching lines @@
                     copy_list.append({'src': fullpath, 'dest': filename})
                 else:
                     total_tests += 1
                     copy_list.append({'src': fullpath, 'dest': filename})
 
             if not total_tests:
                 # We can skip the support directory if no tests were found.
                 if 'support' in dirs:
                     dirs.remove('support')
 
-                if copy_list:
-                    # Only add this directory to the list if there's something to import
-                    self.import_list.append({'dirname': root, 'copy_list': copy_list,
-                        'reftests': reftests, 'jstests': jstests, 'total_tests': total_tests})
+            if copy_list:
+                # Only add this directory to the list if there's something to import
+                self.import_list.append({'dirname': root, 'copy_list': copy_list,
+                    'reftests': reftests, 'jstests': jstests, 'total_tests': total_tests})
 
     def import_tests(self):
         if self.import_list:
             self.setup_destination_directory()
 
         converter = W3CTestConverter()
         total_imported_tests = 0
         total_imported_reftests = 0
         total_imported_jstests = 0
 
@@ skipping 53 matching lines @@
                 # FIXME: Eventually, so should js when support is added for this type of conversion
                 mimetype = mimetypes.guess_type(orig_filepath)
                 if 'html' in str(mimetype[0]) or 'xml' in str(mimetype[0])  or 'css' in str(mimetype[0]):
 
                     converted_file = converter.convert_for_webkit(new_path, filename=orig_filepath)
 
                     if not converted_file:
                         shutil.copyfile(orig_filepath, new_filepath)  # The file was unmodified.
                     else:
                         prefixed_properties.extend(set(converted_file[0]) - set(prefixed_properties))
-                        outfile = open(new_filepath, 'w')
+                        outfile = open(new_filepath, 'wb')
                         outfile.write(converted_file[1])
                         outfile.close()
                 else:
                     shutil.copyfile(orig_filepath, new_filepath)
 
                 copied_files.append(new_filepath.replace(self._webkit_root, ''))
 
             self.remove_deleted_files(new_path, copied_files)
             self.write_import_log(new_path, copied_files, prefixed_properties)
 
@@ skipping 76 matching lines @@
             for prop in prop_list:
                 import_log.write(prop + '\n')
         else:
             import_log.write('None\n')
         import_log.write('------------------------------------------------------------------------\n')
         import_log.write('List of files:\n')
         for item in file_list:
             import_log.write(item + '\n')
 
         import_log.close()