Download loved tracks from last.fm for local database. Get album name and date for collections where tracks are grouped by albums

loved.py

import urllib2
import xml.etree.ElementTree as ET
import pickle
import time

def main():
    # Gather required info.
    lastfm_username = raw_input("Lastfm username: ").strip()
    lastfm_key = raw_input("Lastfm API key: ").strip()
    
    loved = []
    unknown = []
    page = 1
    while True:
        url = "http://ws.audioscrobbler.com/2.0/?method=user.getlovedtracks&user=%s&api_key=%s&page=%d" % \
              (lastfm_username, lastfm_key, page)
        print("Fetching: " + url)
        tree = ET.parse(urllib2.urlopen(url)).getroot()
        tracks = tree.findall('lovedtracks/track')
        for track in tracks:
            title = track.find('name').text
            artist = track.find('artist/name').text
            track_mbid = track.find('mbid').text
            if track_mbid == None:
                unknown.append((artist, title))
                print("UNKNOWN!" + artist + ", " + title)
                continue

            url = "http://ws.audioscrobbler.com/2.0/?method=track.getInfo&api_key=%s&mbid=%s" % \
              (lastfm_key, track_mbid)
            print("Fetching: " + url)
            track_tree = ET.parse(urllib2.urlopen(url)).getroot()
            album = track_tree.find('track/album')
            if album == None:
                unknown.append((artist, title))
                print("UNKNOWN!" + artist + ", " + title)
                continue
            track_number = album.get('position')
            album_title = album.find('title').text
            album_mbid = album.find('mbid').text
            
            url = "http://musicbrainz.org/ws/2/release/%s" % \
              (album_mbid)
            repeat = True
            while repeat:
                print("Fetching: " + url)
                try:
                    album_tree = ET.parse(urllib2.urlopen(url)).getroot()
                    repeat = False
                except urllib2.HTTPError as e:
                    if e.code == 503:
                        time.sleep(2)
                    else:
                        print ("Unknown code: " + str(e))
                        exit(1)

            try:
                album_date = album_tree.find('{http://musicbrainz.org/ns/mmd-2.0#}release/{http://musicbrainz.org/ns/mmd-2.0#}date').text
                dash_index = album_date.find("-")
                if dash_index > 0:
                    album_date = album_date[0: dash_index]
            except AttributeError:
                album_date = "NoDate"
            
            track_hash = {
                'artist': artist,
                'album_name': album_title,
                'album_date': album_date,
                'album_hash': album_mbid,
                'track_position': track_number,
                'track_title': title
            }
            print("Got track info: " + str(track_hash))

            loved.append(track_hash)
        if len(tracks) < 50:
            break
        page += 1

    print("Got " + str(len(loved)) + " loved tracks")

    output = open('loved.pkl', 'wb')
    pickle.dump(loved, output)
    output.close()
    
    print("Unknown:")
    for u in unknown:
        print(str(u))

if __name__ == '__main__':
    main()