A reddit imgur ripper. Prevents downloading same files repeatedly if ran multiple times.

redditrip.py

#!/usr/local/bin/python3

# Raymond Ho
# November 29, 2014

# Downloads imgur uploads from reddit and saves them in a new directory.
# Rerun every so often to download new pictures without downloading the
# same pictures multiple times. A text file keeps track of what has 
# already been downloaded.

import requests			# To access Reddit
import urllib.request   # To download files
import re 				# To find RegEx,
import os			 	# To check if directory exists

def rip_from_reddit(subreddit='MechanicalKeyboards', post_limit=100):

	"""
	Returns a list of unseen pics from specified subreddit.
	Updates text file's link.
	"""

	url = 'http://www.reddit.com/r/' + subreddit + '/.json?' +\
	 'limit=' + str(post_limit)
	 
	authorize = ('User-Agent', 'Raymonds redditrip')
	r = requests.get(url, auth=authorize)

	# RegExp to find image links.
	images = re.findall("http://i.imgur.com/\w+.(?:jpg|gif|png)", r.text)

	unseen_pics = [] # These are going to be the pics we download later.

	# Updates the text file with new links
	with open(file_name, 'a+') as f:
		for link in images:
			if link not in seen_pics:
				unseen_pics.append(link)
				f.write(link + '\n')
				print('Adding', link)

	if not unseen_pics:
		print('No new files to download.\
			\nCheck subreddit spelling / Increase limit.')

	#Create path if it doesn't exist.	
	PICDIR = 'reddit_' + subreddit.lower() + '/'
	if not os.path.exists(PICDIR):
		os.makedirs(PICDIR)

	# Download everything we haven't seen already.
	for pic in unseen_pics:
		urllib.request.urlretrieve(pic, PICDIR + pic[pic.rfind('/')+1:])
		print('Downloading..', pic)

if __name__ == '__main__':

	seen_pics = []
	file_name = 'redditlinks.txt'

	# Read file into a list to see what we already downloaded
	# If file doesn't exist, create it.
	try:
		with open(file_name, 'r') as f:
			for line in f:
				seen_pics.append(line.strip())
	except FileNotFoundError:
		open(file_name, 'w')

	rip_from_reddit()