Script to retrieve the number of U.S. utility patents granted by week during 2009-2010

Two_years_of_weekly_grants_2009_10.py

import urllib2
import re
import datetime
import pprint
import time 

# Start by Getting Date of Last Tuesday of 2010
current = datetime.date(2010,12,31)
if current.weekday() == 0:
  #if New Years Eve falls on Monday, go back into previous week
  current = current - datetime.timedelta(days=1)
tuesday = current - datetime.timedelta(days=current.weekday()) + datetime.timedelta(days=1)

weeks = range(105)
#weeks = [1]

results = []
#Loop over weeks
for week in weeks:
  current = tuesday - datetime.timedelta(days=(week*7))
  month = str(current.month)
  day = str(current.day)
  year = str(current.year)

  # Create Search Address with Correct Search Options
  search_address = "http://patft.uspto.gov/netacgi/nph-Parser?Sect1=PTO2&Sect2=HITOFF&u=/netahtml/PTO/search-adv.htm"
  options = "&r=0&p=1&f=S&l=50&Query=+ISD/%s/%s/%s+AND+APT/1&d=PTXT" % (month, day, year)
  the_add = search_address + options

  # Query USPTO search
  response = urllib2.urlopen(the_add)
  html = response.read()

  # Extract Patent Numbers and Sort
  Number = re.findall(': (\d\d\d\d) patents\.', html)
  thedate = "%s/%s/%s" % (month, day, year)
  result = (thedate, Number[0])
  results.append(result)

  #Print out result
  print "The Total Number of U.S. Patents issued the week of %s/%s/%s is: %s" % \
         (month, day, year, Number[0])
  #Let's not flood the patent office
  time.sleep(4)       
#Save to File
outfile = open('Two_years_of_Grants_2009_2010.csv', 'w') 
for result in results:
  line = "[\'%s\', %s],\n" % (result[0], result[1]) 
  outfile.write(line)      

outfile.close()