Wikigame - Get to Philosophy from any given article

wikigame.py

import urllib2
import sys
from bs4 import BeautifulSoup
import re

# Used for formating wiki links
wiki_endpoint = "http://en.wikipedia.org/wiki/"
# Count how many links were visited on the way
counter = 0
# Regular Pattren for link
link_patt = re.compile(r'<a href="/wiki/[^:"].*?".*?>')
# Keep a list of visited articles to detect loops
visited = []
def remove_parentheses(data):
	''' We want the first link that is not in parentheses, so we make sure it's not '''
	first = data.split('href="')
	# If the paragraph dosen't contain parentheses, return the first wiki link
	if not ('(' in first[0] or ')' in first[0]):
		return [x for x in first if x.startswith('/wiki/') and not 'File:' in x][0]
	# Else, find the first link after the parentheses
	total = first[0]
	for i in range(len(first)):
		if total.count('(') == total.count(')'):
			# Make sure it's a wiki link
			return [x for x in first[i:] if x.startswith('/wiki/')][0]
		if i > 0:
			total += first[i]

def get_next(url):
	''' Get the first link from the article '''
	# Get the HTML
	html = urllib2.urlopen(url).read()
	# Get the article's paragraphs
	soup = BeautifulSoup(html)
	paragraphs = [x for x in soup.find(id='mw-content-text').contents if x.name == u'p']
	paragraph = ''
	for p in paragraphs:
		if re.search(link_patt, str(p)):
			paragraph = str(p)
			break
	# Sometimes we have to deal with a 'may refer to' page, which only has links in list items
	if paragraph == '':
		lists = [x for x in soup.find(id='mw-content-text').contents if x.name in [u'ul', u'ol']]
		for lst in lists:
			list_items = [x for x in lst.contents if x.name == 'li']
			for i in list_items:
				if re.search(link_patt, str(i)):
					paragraph = str(i)
					break
	link = remove_parentheses(paragraph).split('"')[0].split('/')[-1]
	return link

def next_url(title):
	''' URL formatting '''
	return wiki_endpoint + urllib2.quote(title).replace('%20', '_')

def nextPage(name):
	global counter
	''' Recursive function to get from a given article to the Philosophy article '''
	if name == "Philosophy":
		return
	counter += 1
	# Get the title of the next article
	url = next_url(name)
	# Format it to a URL
	title = get_next(url)
	# Log to the user
	print "{0} reached.".format(title.replace('_', ' '))
	# Check if we were here already and declare a loop if we were
	if title in visited:
		print 'Stuck in a loop, terminating.'
		return
	# Add it to the list of visited articles
	visited.append(title)
	# Make a recursive call
	nextPage(title)


def main():
	''' Get the first article's title from the user and run the code for it '''
	if len(sys.argv) < 2:
		print 'Please enter a title.'
		return
	title = ' '.join(sys.argv[1:])
	try:
		nextPage(title)
		print 'Number of articles visited: %s' % str(counter)
	except KeyboardInterrupt:
		print 'User has stopped.'
	except Exception, e:
		print 'ERROR: {0}'.format(str(e))

if __name__ == '__main__':
	main()

For a less reliable version which only use pure Python, please go here:
https://gist.github.com/OzTamir/4fd0ca8ff8ef8783412b/3b01e3d317a569235e909e7fca96ade50027c34f