Taming Your Music Collection: Album Artwork

Iterate The Albums
     Our first programming task will be to iterate over the albums in preparation for downloading the album artwork. The reason we want to iterate only the albums and not all the files is because we want our application to be efficient. We don't want to call the web service any more often than is absolutely necessary. Since the album artwork applies to all the songs in an album, we can do the lookup once and apply it to each file, rather than retrieve the album artwork one time for each track on the album.
Code: Note: I've dimmed code from the previous articles that has little to do with this step

import sys, os, fnmatch, shutil, argparse, re, urllib, urllib2, xml.dom.minidom, glob
from mutagen.easyid3 import EasyID3
from mutagen.mp3 import MP3

def getTrackNumber(track):
    if track.find("/") > -1:
        return getTrackNumber(track[0:track.find("/")])
    elif track.isdigit():
        return track.zfill(2)

def getID3FromFilename(file, id3info):
    filename = os.path.basename(file)
    m = re.match(r"(?P<trackNumber>\d{2})\. (?P<artist>((?! -).)+) - (?P<title>[^\.]+)\.mp3", filename)

    id3info["tracknumber"] = m.group('trackNumber')
    id3info["artist"] = m.group('artist')
    id3info["title"] = m.group('title')
    id3info["album"] = file.split('/')[-2]

    return id3info

def getID3FromWeb(file, id3info):
    # Get the Artist and Title from id3info to make the query
    _artist = id3info["artist"][0]
    _title = id3info["title"][0]

    # Query the MusicBrainz web service
        query = { 'query' : 'artist:' + _artist + ' AND recording:' + _title }
        response = urllib2.urlopen('http://musicbrainz.org/ws/2/recording?' + urllib.urlencode(query))
        x = xml.dom.minidom.parse(response)

        recordingList = x.getElementsByTagNameNS('http://musicbrainz.org/ns/mmd-2.0#', 'recording-list')[0]
        recording = recordingList.getElementsByTagName('recording')[0]
        if recording.nodeType == 1 and recording.attributes.get('ext:score').value == '100':
            id3info["tracknumber"] = getTrackNumber(str(int(recording.getElementsByTagName('track-list')[0].attributes.get('offset').value)+1))
            id3info["artist"] = recording.getElementsByTagName('name')[0].firstChild.nodeValue
            id3info["title"] = recording.getElementsByTagName('title')[0].firstChild.nodeValue

            # Check for multiple releases containing the title
            releaseList = recording.getElementsByTagName('release-list')[0]
            releases = releaseList.getElementsByTagName('release')

            releaseOpts = []

            for release in releases:
                releaseTitle = release.getElementsByTagName('title')[0].firstChild.nodeValue

                if releaseTitle not in releaseOpts:

            if len(releaseOpts) > 1:
                # Ask the user which option they'd prefer
                print file + " has multiple options:"
                index = 0
                print "0) Ignore file and discard edits."
                for opt in releaseOpts:
                    index += 1
                    print str(index) + ") " + opt

                choice = input("Choice: ")

                if choice == 0:
                    print "You chose to ignore the file."
                    id3info["album"] = releaseOpts[choice - 1]
                    id3info["tracknumber"] = getTrackNumber(str(int(releases[choice - 1].getElementsByTagName('track-list')[0].attributes.get('offset').value)+1))
                id3info["album"] = releaseOpts[0]

        print sys.exc_info()[0]

    return id3info

def move(output, file, dirMatch):
    audio = MP3(file, ID3=ID3)
    # Add ID3 tag if none exist

    id3info = EasyID3(file)

        _trackNumber = getTrackNumber(id3info["tracknumber"][0])
        _artist = id3info["artist"][0]
        _title = id3info["title"][0]
        _album = id3info["album"][0]
    except KeyError:
        if dirMatch:
            id3info = getID3FromFilename(file, id3info)
            id3info = getID3FromWeb(file, id3info)

        _trackNumber = id3info["tracknumber"][0]
        _artist = id3info["artist"][0]
        _title = id3info["title"][0]
        _album = id3info["album"][0]

    outputDir = output + "/" + _artist + "/" + _album + "/"
    outputFile = _trackNumber + ". " + _artist + " - " + _title + ".mp3"

    print "Saving file ", outputDir + outputFile

    if not os.path.exists(outputDir):

    shutil.move(file, outputDir + outputFile)

def main():
    parser = argparse.ArgumentParser()

    parser.add_argument('-d', '--directory', nargs=1, required=True, help='')
    parser.add_argument('-o', '--output', nargs=1, required=True, help='')

    args = parser.parse_args()

    directory = os.path.abspath(args.directory[0])
    output = os.path.abspath(args.output[0])

    dirMatch = (directory == output)

    for root, subFolders, filenames in os.walk(directory):
        for filename in fnmatch.filter(filenames, '*.mp3'):
            move(output, os.path.join(root, filename), dirMatch)
    # Build a list of all the album directories
    dirList = glob.glob(os.path.join(output, '*', '*'))
    dirList = filter(lambda f: os.path.isdir(f), dirList)
    for albumDir in dirList:
    	print "Album:", albumDir

$ ./mp3-tagger.py -d /UntamedMusic/ -o /Music/
Album: /Music/Coheed And Cambria/No World For Tomorrow
Album: /Music/Thrice/The Alchemy Index, Volumes III & IV

     So we've built a list of directories that points to each album and created a loop to iterate over each one using the glob library. You can learn more about glob at http://docs.python.org/library/glob.html. For the purposes of this article, I am only using 2 albums, as seen above. Now we can move on to the next step and actually retrieve the album artwork for each album.