#!/usr/local/bin/python

#  itunes-url-decoder.py
#
#  A quick and dirty Python hack to retrieve the RSS feed URL from an
#  iTunes Music Store (ITMS) podcast URL.
#
#  Written in August 2007 by Andrew Clarke <mail@ozzmosis.com>
#  and released to the public domain.

__MYVERSION__ = "1.3"
__MYDATE__ = "2007-08-14"

import sys, urllib2, gzip, StringIO, xml.parsers.expat

feedURL = ''
have_key = False

def usage():
    print "Usage: " + sys.argv[0] + " url [url...]"
    sys.exit(1)

if len(sys.argv) < 2:
    usage()

def http_get(uri):
    request = urllib2.Request(uri)
    request.add_header("Accept-encoding", "gzip")
    usock = urllib2.urlopen(request)
    data = usock.read()
    if usock.headers.get('content-encoding', None) == 'gzip':
        data = gzip.GzipFile(fileobj=StringIO.StringIO(data)).read()
    return data

def get_xml_url(data):
    #  HTML scraping...

    open_str = "itmsOpen('"
    start = data.find(open_str)
    if start == -1:
        return ''
    data = data[start + len(open_str):]
    data = data.replace("itms://", "http://")
    end = data.find("'")
    return data[:end]

def char_data(data):
    # This feels like a terrible hack

    global have_key, feedURL

    if feedURL != '':
        return

    if have_key:
        feedURL = data
    else:
        if data == 'feedURL':
            have_key = True

for arg in sys.argv[1:]:
    feedURL = ''
    have_key = False

    xml_url = get_xml_url(http_get(arg))

    if xml_url == '':
        print >> sys.stderr, sys.argv[0] + ' error: itmsOpen() not found in ' + arg
        continue

    xml_data = http_get(xml_url)
    x = xml.parsers.expat.ParserCreate()
    x.CharacterDataHandler = char_data
    x.Parse(xml_data, True)

    if feedURL == '':
        print >> sys.stderr, sys.argv[0] + ' error: feedURL not found in XML file ' + xml_url
    else:
        print feedURL