Wayback Machine: finds historical New York Times front pages that include a certain string, and prints a link to those archived pages.

get_archive.py

import requests
import json
from bs4 import BeautifulSoup

site = 'nytimes.com'
for year in xrange(2010,2014+1):
	for month in xrange(1,12+1):
		url = 'http://archive.org/wayback/available?url='+site+'&timestamp='+str(year)+str(month).zfill(2)+str('01')
		r = requests.get(url)
		data = json.loads(r.text)
		newurl = data['archived_snapshots']['closest']['url']
		p = requests.get(newurl)
		soup = BeautifulSoup(p.text)
		summary = [h.text for h in soup.findAll('p', attrs={'class': 'summary'})] + [h.text for h in soup.findAll('dd', attrs={'class': 'summary'})]
		matching = [s for s in summary if 'Obama' in s]
		for item in matching:
			print month,year,newurl,item.replace('\r\n','').encode('utf-8')