Rekordbox 6 Database#

Pyrekordbox uses SQLALchemy as ORM-Framework to handle the Rekordbox database file (master.db). The table declarations can be found in pyrekordbox.db6.tables.

Querying the database#

A query can be executed on any table by calling the query() method. The result is a SQLAlchemy query object, which can be used to filter and sort the results.

from pyrekordbox.db6 import tables

query = db.query(tables.DjmdContent)
results = query.filter(tables.DjmdContent.Title == "My Song").all()

To simplify querying the database, the Rekordbox6Database class provides simple getters for executing queries on all the tables. The parameters of the getters are passed to the query.filter_by() method. If the query is filtered by a unique key (e.g. ID), the query will be executed using the query.one() method, returning the table instance directly:

# Return specific entry in DjmDContent table
content = db.get_content(ID=0)

In all other cases the query is returned, allowing to further filter and sort the results:

# Query and sort entries in DjmdHistory table
for history in db.get_history().order_by(tables.DjmdHistory.DateCreated):
    print(history)

This allows the user to make use of the full power of SQLAlchemy queries.

Relationships#

Some values of table entries are linked to other tables and can not be updated directly. For example, the djmdContent table contains an ArtistID column, which links to the djmdArtist table. The table declarations provide relationships to access the linked values. The artist of a song can be accessed as follows:

content = db.get_content(ID=0)
artist = content.Artist

A full list of linked tables can be found in the Rekordbox 6 Database Format documentation.

Updating the database#

Many values in the Rekordbox database can be updated by simply changing the corresponding attribute of the table instance and calling the db.commit() method:

content = db.get_content().first()
content.Title = "New title"
db.commit()

Since some values of table entries are linked to other tables these values can not be updated directly. Also, many values depend on other values on the table and have to be updated accordingly to ensure consistency of the database, for example the TrackNo of songs in playlists.

To simplify updating the database, the Rekordbox6Database class provides a set of high level methods for updating the database. These methods take care of updating all linked values and metadata.

Playlists#

A new playlist can be created by calling the db.create_playlist() method:

playlist = db.create_playlist("My Playlist")

It creates a new DjmdPlaylist instance and adds it to the djmdPlaylist table. By default, the playlist is inserted as last element of the parent playlist folder. Alternatively, the sequence number of the new playlist can be specified:

playlist = db.create_playlist("My Playlist", seq=2)

To add the playlist to a specific playlist folder, the parent parameter can be used to pass a playlist folder instance or ID:

folder = db.get_playlist(Name="My Folder").one()  # Query for unique playlist folder
playlist = db.create_playlist("My Playlist", parent=folder)

Playlist folders can also be created , which are also stored in the djmdPlaylist table. The method accepts the same parameters as the db.create_playlist() method:

folder = db.create_playlist_folder("My Folder")

Playlists and playlist folders can also be deleted:

playlist = db.get_playlist(Name="My Playlist").one()
db.delete_playlist(playlist)

folder = db.get_playlist(Name="My Folder").one()
db.delete_playlist(folder)

Caution

When deleting a playlist or playlist folder, all children are deleted as well. Deleting a playlist removes all songs from the playlist before deleting it. Similarly, deleting a playlist folder removes all playlists and playlist folders in the folder, including all songs in the sub-playlists.

Adding tracks to a playlist is done by calling the db.add_to_playlist() method. It accepts a DjmdContent instance or corresponding ID and creates a new entry in the djmdSongPlaylist table, which stores the contents of playlists:

content = db.get_content(ID=0)
playlist = db.get_playlist(Name="My Playlist").one()
song = db.add_to_playlist(playlist, content)

To delete a song from a playlist, the DjmdSongPlaylist instance or ID has to be passed, since a track can be contained in a plalyist more than once:

playlist = db.get_playlist(Name="My Playlist").one()
song = playlist.Songs[0]

db.remove_from_playlist(playlist, song)

Tracks#

New tracks can be added to your rekordbox collection by calling the db.add_content() method:

track = db.add_content("/Users/deadmau5/Downloads/Banger.mp3")

This creates a new DjmdContent instance and adds it to the djmdContent-table. This track will appear in your rekordbox collection, but the track metadata (title, artist, etc.) will be blank. To load this data, simply right click the track and select the Reload Tag menu option. rekordbox will correctly populate the metadata, including creating any related table data for things like artist and album.

The db.add_content() method accepts **kwargs which are passed on to the DjmdContent instance. You can use these keyword arguments if you want to populate some of the track metadata right away. However, things like DjmdArtist and [DjmdAlbum] are foreign keys to other tables, so make sure you create these objects first and pass the objects via kwargs, rather than passing plain strings. A plain string will work for Title, however.

db.add_content("/Users/deadmau5/Downloads/Banger.mp3", Title="Banger")

Populating the DjmdContent metadata directly from the track file is not currently supported, but it will hopefully be included in a future release.

Note

More coming soon!