Parse-USPTO-Patents.py

# Load USPTO .xml document
xml_text = html.unescape(open(filename, 'r').read())
	
# Split out patent applications / grants
for patent in xml_text.split("<?xml version=\"1.0\" encoding=\"UTF-8\"?>"):
  
  # Skip if it doesn't exist
  if patent is None or patent == "":
    continue 
  
  # Load patent text as HTML document
  bs = BeautifulSoup(patent)                                     

  # Search patent for application 
  application = bs.find('us-patent-application')
  
  # If no application, search for grant
  if application is None: 
    application = bs.find('us-patent-grant')