Fix the regex flatting html to prevent misdetection

grit/format/html_inline.py

 #!/usr/bin/env python
 # Copyright (c) 2012 The Chromium Authors. All rights reserved.
 # Use of this source code is governed by a BSD-style license that can be
 # found in the LICENSE file.
 
 """Flattens a HTML file by inlining its external resources.
 
 This is a small script that takes a HTML file, looks for src attributes
 and inlines the specified file, producing one HTML file with no external
 dependencies. It recursively inlines the included files.
@@ skipping 275 matching lines @@
 
   # Check conditional elements, second pass. This catches conditionals in any
   # of the text we just inlined.
   flat_text = CheckConditionalElements(flat_text)
 
   # Allow custom modifications before inlining images.
   if rewrite_function:
     flat_text = rewrite_function(input_filepath, flat_text, distribution)
 
   flat_text = re.sub(
-      '<(?!script)[^>]+?src=(?P<quote>")(?P<filename>[^"\']*)\\1',
+      '<(?!script)(?:[^>]+?\s)src=(?P<quote>")(?P<filename>[^"\']*)\\1',
       SrcReplace, flat_text)
 
   # TODO(arv): Only do this inside <style> tags.
   flat_text = InlineCSSImages(flat_text)
 
   flat_text = re.sub(
-      '<link rel="icon".+?href=(?P<quote>")(?P<filename>[^"\']*)\\1',
+      '<link rel="icon"\s(?:[^>]+?\s)?href=(?P<quote>")(?P<filename>[^"\']*)\\1',
       SrcReplace, flat_text)
 
   if names_only:
     flat_text = None  # Will contains garbage if the flag is set anyway.
   return InlinedData(flat_text, inlined_files)
 
 
 def InlineToString(input_filename, grd_node, allow_external_script=False,
                    rewrite_function=None):
   """Inlines the resources in a specified file and returns it as a string.
@@ skipping 50 matching lines @@
 
 def main():
   if len(sys.argv) <= 2:
     print "Flattens a HTML file by inlining its external resources.\n"
     print "html_inline.py inputfile outputfile"
   else:
     InlineToFile(sys.argv[1], sys.argv[2], None)
 
 if __name__ == '__main__':
   main()