requires BeautifulSoup and requests (python-requests & python-beautifulsoup on debian)

scrapetorrents.py

#usage script.py removed.csv outputdirectory
import urllib, os, sys, random, time
from BeautifulSoup import BeautifulSoup
import requests
GOOGLE_CACHE = "http://webcache.googleusercontent.com/search?%s"
TRACKER_WIN = "cache:http://g.e-hentai.org/gallerytorrents.php?gid=%i&t=%s"
EH_TRACKER = "http://ehtracker.org/t/"

SESS = requests.Session(headers={"User-Agent": "Mozilla/5.0 (X11; Linux i686; rv:31.0) Gecko/20100101 Firefox/30.0"})
OUT_PATH = sys.argv[2]

def cache_url(gid, tok):
	return GOOGLE_CACHE % urllib.urlencode({"q": TRACKER_WIN % (gid, tok)})
	
def load_csv(filename):
	stuff = []
	with open(filename) as f:
		for l in f.readlines():
			gidtok = l.split(',')[0]
			try:
				spl = gidtok.split('/')
				stuff.append((int(spl[0]), spl[1]))
			except:
				pass
	print "Loaded from csv:", len(stuff)
	return stuff
	
def download_torrent(data, gid):
	soup = BeautifulSoup(data)
	r = soup.findAll('a')
	c = 1
	for i in r:
		if i['href'].startswith(EH_TRACKER):
			print gid,
			req = SESS.get(i['href'])
			if req.status_code == 200:
				with open(os.path.join(OUT_PATH, u"%i-%i-%s.torrent" % (gid, c, i.text)), 'wb')as outf:
					outf.write(req.content)
				c += 1
				print "Downloaded:", i.text
			else:
				print "Error:", req.status_code
				

def main():
	galleries = load_csv(sys.argv[1])
	for g in galleries:
		u = apply(cache_url, g)
		req = SESS.get(u)
		if req.status_code == 200:
			download_torrent(req.text, g[0])
			print g[0], "Data:"
			time.sleep(random.randint(0, 3))
		else:
			print g[0], "Error:", req.status_code
			time.sleep(random.randint(0, 10))

if __name__ == "__main__":
	main()