#!/usr/bin/env python

"""
Extract common image urls from arbitrary text and download them to a directory.

  extract-images http://example.com /my/image/directory

"""

import re
import sys
import urllib
import os.path

url = sys.argv[1]
directory = sys.argv[2]

text = urllib.urlopen(url).read()
text = text.replace('"', '"')

if not os.path.isdir(directory):
    os.mkdir(directory)

for img in re.findall(r'"(.*?\.(?:jpg|png|gif))"', text):
    data = urllib.urlopen(img).read()
    filename = img.replace('http://', '').replace('/', '_')
    file(directory + '/' + filename, 'w').write(data)