Discogs Plugin#
The discogs plugin extends the autotagger's search capabilities to include
matches from the Discogs database.
Files can be imported as albums or as singletons. Since Discogs matches are always based on Discogs releases, the album tag is written even to singletons. This enhances the importers results when reimporting as (full or partial) albums later on.
Installation#
To use the discogs plugin, first enable it in your configuration (see
Using Plugins). Then, install beets with discogs extra
pip install "beets[discogs]"
You will also need to register for a Discogs account, and provide authentication credentials via a personal access token or an OAuth2 authorization.
Matches from Discogs will now show up during import alongside matches from MusicBrainz. The search terms sent to the Discogs API are based on the artist and album tags of your tracks. If those are empty no query will be issued.
If you have a Discogs ID for an album you want to tag, you can also enter it at the "enter Id" prompt in the importer.
Authentication via Personal Access Token#
As an alternative to OAuth, you can get a token from Discogs and add it to your configuration. To get a personal access token (called a "user token" in the python3-discogs-client documentation):
login to Discogs;
visit the Developer settings page;
press the Generate new token button;
copy the generated token;
place it in your configuration in the
discogssection as theuser_tokenoption:discogs: user_token: "xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx"
Configuration#
This plugin can be configured like other metadata source plugins as described in Using Metadata Source Plugins.
Default#
discogs:
apikey: REDACTED
apisecret: REDACTED
tokenfile: discogs_token.json
user_token:
index_tracks: no
append_style_genre: no
separator: ', '
strip_disambiguation: yes
featured_string: Feat.
anv:
artist_credit: yes
artist: no
album_artist: no
data_source_mismatch_penalty: 0.5
search_limit: 5
-
index_tracks (default:
no)# Index tracks (see the Discogs guidelines) along with headers, mark divisions between distinct works on the same release or within works. When enabled, beets will incorporate the names of the divisions containing each track into the imported track's title.
For example, importing divisions album would result in track names like:
Messiah, Part I: No.1: Sinfony Messiah, Part II: No.22: Chorus- Behold The Lamb Of God Athalia, Act I, Scene I: Sinfonia
whereas with
index_tracksdisabled you'd get:No.1: Sinfony No.22: Chorus- Behold The Lamb Of God Sinfonia
This option is useful when importing classical music.
-
append_style_genre (default:
no)# Appends the Discogs style (if found) to the
genrestag. This can be useful if you want more granular genres to categorize your music. For example, a release in Discogs might have a genre of "Electronic" and a style of "Techno": enabling this setting would append "Techno" to thegenreslist.
-
separator (default:
", ")# How to join multiple style values from Discogs into a string.
Changed in version 2.7.0: This option now only applies to the
stylefield as beets now only handles lists ofgenres.
-
strip_disambiguation (default:
yes)# Discogs uses strings like
"(4)"to mark distinct artists and labels with the same name. If you'd like to use the Discogs disambiguation in your tags, you can disable this option.
-
featured_string (default:
Feat.)# Configure the string used for noting featured artists. Useful if you prefer
Featuringorft..
- anv#
This configuration option is dedicated to handling Artist Name Variations (ANVs). Sometimes a release credits artists differently compared to the majority of their work. For example, "Basement Jaxx" may be credited as "Tha Jaxx" or "The Basement Jaxx". You can select any combination of these config options to control where beets writes and stores the variation credit. The default, shown below, writes variations to the artist_credit field.
discogs: anv: artist_credit: yes artist: no album_artist: no
-
data_source_mismatch_penalty (default:
0.5)# Penalty applied when the data source of a match candidate differs from the original source of your existing tracks. Any decimal number between 0.0 and 1.0
This setting controls how much to penalize matches from different metadata sources during import. The penalty is applied when beets detects that a match candidate comes from a different data source than what appears to be the original source of your music collection.
Example configurations:
# Prefer MusicBrainz over Discogs when sources don't match plugins: musicbrainz discogs musicbrainz: data_source_mismatch_penalty: 0.3 # Lower penalty = preferred discogs: data_source_mismatch_penalty: 0.8 # Higher penalty = less preferred
# Do not penalise candidates from Discogs at all plugins: musicbrainz discogs musicbrainz: data_source_mismatch_penalty: 0.5 discogs: data_source_mismatch_penalty: 0.0
# Disable cross-source penalties entirely plugins: musicbrainz discogs musicbrainz: data_source_mismatch_penalty: 0.0 discogs: data_source_mismatch_penalty: 0.0
Tip
The last configuration is equivalent to setting:
match: distance_weights: data_source: 0.0 # Disable data source matching
-
source_weight (default:
0.5)# Deprecated since version 2.5: Use data_source_mismatch_penalty instead.
-
search_limit (default:
5)# Maximum number of search results to return.
Troubleshooting#
Several issues have been encountered with the Discogs API. If you have one, please start by searching for a similar issue on the repo.
Here are two things you can try:
Try deleting the token file (
~/.config/beets/discogs_token.jsonby default) to force re-authorization.Make sure that your system clock is accurate. The Discogs servers can reject your request if your clock is too out of sync.
Matching tracks by Discogs ID is not yet supported. The --group-albums
option in album import mode provides an alternative to singleton mode for
autotagging tracks that are not in album-related folders.