Mutagen is a Python module to handle audio metadata.

UFID: Unique File Identifier

The unique file identifier is a reference to an external database that may contain more information about the work tagged with it. Like many ID3 tags, it can have multiple values. For example, you can link the song "Black Sabbath" by "Black Sabbath" as mt0030286828 in AllMusic, and as 153a3596-af03-36ed-976a-83fec8745732 in MusicBrainz.

Note that according to the definition of the UFID ID3 frame, the URLs should not be queries themselves. I guess that it might be a feature of the software that reads this tag to transform it to direct links. In the example used, those links would be: http://www.allmusic.com/song/black-sabbath-mt0030286828 and https://musicbrainz.org/work/153a3596-af03-36ed-976a-83fec8745732.

Let's see how to write UFID and read it back in Python, with Mutagen.

{% highlight python %}

from mutagen import id3 song = id3.ID3('test.mp3')

Note that there is no UFID tag in the beginning:

song.keys() dict_keys(['TPE1', 'TALB', 'APIC:', 'TDRC', 'TRCK', 'TPE2', 'TPOS', 'TIT2'])

Let's add two references:

amurl = 'http://www.allmusic.com/' amref = 'mt0030286828' mburl = 'https://musicbrainz.org/' mbref = '153a3596-af03-36ed-976a-83fec8745732' song.add(id3.UFID(owner=amurl, data=amref.encode())) song.add(id3.UFID(owner=mburl, data=mbref.encode()))

There are now two UFIDs:

song.keys() dict_keys(['UFID:https://musicbrainz.org/', 'TRCK', 'TPE1', 'UFID:http://www.allmusic.com/', 'TPE2', 'TALB', 'TIT2', 'APIC:', 'TPOS', 'TDRC'])

Read them back:

song['UFID:https://musicbrainz.org/'] UFID(owner='https://musicbrainz.org/', data=b'153a3596-af03-36ed-976a-83fec8745732') song['UFID:https://musicbrainz.org/'].owner 'https://musicbrainz.org/' song['UFID:https://musicbrainz.org/'].data b'153a3596-af03-36ed-976a-83fec8745732' song['UFID:https://musicbrainz.org/'].data.decode() '153a3596-af03-36ed-976a-83fec8745732' {% endhighlight %}

Note the use of encode() and decode() methods of unicode used for the data part of the frame, as it expects its contents to be bytes.