Discogs Label Disambiguation Fix + config option. (#6035)

Fixes #5366 . 

Adds removal of disambiguation from label names, in addition this PR
moves the Discogs disambiguation function out of the
MetadataSourcePlugin, and puts it in the Discogs plugin, keeping the
parent class more generic.

A config option has been added to allow disabling Discogs disambiguation
removal. Tests and docs are written for the feature, and shows no side
effects in other plugins that rely on the MetadataSourcePlugin.

beets/metadata_plugins.py

@@ -271,10 +271,9 @@ class MetadataSourcePlugin(BeetsPlugin, metaclass=abc.ABCMeta):
         """Returns an artist string (all artists) and an artist_id (the main
         artist) for a list of artist object dicts.
 
-        For each artist, this function moves articles (such as 'a', 'an',
+        For each artist, this function moves articles (such as 'a', 'an', and 'the')
-        and 'the') to the front and strips trailing disambiguation numbers. It
+        to the front. It returns a tuple containing the comma-separated string
-        returns a tuple containing the comma-separated string of all
+        of all normalized artists and the ``id`` of the main/first artist.
-        normalized artists and the ``id`` of the main/first artist.
         Alternatively a keyword can be used to combine artists together into a
         single string by passing the join_key argument.
 
@@ -298,8 +297,6 @@ class MetadataSourcePlugin(BeetsPlugin, metaclass=abc.ABCMeta):
             if not artist_id:
                 artist_id = artist[id_key]
             name = artist[name_key]
-            # Strip disambiguation number.
-            name = re.sub(r" \(\d+\)$", "", name)
             # Move articles to the front.
             name = re.sub(r"^(.*?), (a|an|the)$", r"\2 \1", name, flags=re.I)
             # Use a join keyword if requested and available.

beetsplug/discogs.py

@@ -76,6 +76,8 @@ TRACK_INDEX_RE = re.compile(
     re.VERBOSE,
 )
 
+DISAMBIGUATION_RE = re.compile(r" \(\d+\)")
+
 
 class ReleaseFormat(TypedDict):
     name: str
@@ -96,6 +98,7 @@ class DiscogsPlugin(MetadataSourcePlugin):
                 "separator": ", ",
                 "index_tracks": False,
                 "append_style_genre": False,
+                "strip_disambiguation": True,
             }
         )
         self.config["apikey"].redact = True
@@ -362,15 +365,20 @@ class DiscogsPlugin(MetadataSourcePlugin):
 
         label = catalogno = labelid = None
         if result.data.get("labels"):
-            label = result.data["labels"][0].get("name")
+            label = self.strip_disambiguation(
+                result.data["labels"][0].get("name")
+            )
             catalogno = result.data["labels"][0].get("catno")
             labelid = result.data["labels"][0].get("id")
 
         cover_art_url = self.select_cover_art(result)
 
-        # Additional cleanups (various artists name, catalog number, media).
+        # Additional cleanups
+        # (various artists name, catalog number, media, disambiguation).
         if va:
             artist = config["va_name"].as_str()
+        else:
+            artist = self.strip_disambiguation(artist)
         if catalogno == "none":
             catalogno = None
         # Explicitly set the `media` for the tracks, since it is expected by
@@ -622,6 +630,14 @@ class DiscogsPlugin(MetadataSourcePlugin):
 
         return tracklist
 
+    def strip_disambiguation(self, text: str) -> str:
+        """Removes discogs specific disambiguations from a string.
+        Turns 'Label Name (5)' to 'Label Name' or 'Artist (1) & Another Artist (2)'
+        to 'Artist & Another Artist'. Does nothing if strip_disambiguation is False."""
+        if not self.config["strip_disambiguation"]:
+            return text
+        return DISAMBIGUATION_RE.sub("", text)
+
     def get_track_info(self, track, index, divisions):
         """Returns a TrackInfo object for a discogs track."""
         title = track["title"]
@@ -634,6 +650,7 @@ class DiscogsPlugin(MetadataSourcePlugin):
         artist, artist_id = self.get_artist(
             track.get("artists", []), join_key="join"
         )
+        artist = self.strip_disambiguation(artist)
         length = self.get_track_length(track["duration"])
         return TrackInfo(
             title=title,

docs/changelog.rst

@@ -11,6 +11,8 @@ New features:
 
 - :doc:`plugins/lastgenre`: Add a ``--pretend`` option to preview genre changes
   without storing or writing them.
+- :doc:`plugins/discogs`: New config option `strip_disambiguation` to toggle
+  stripping discogs numeric disambiguation on artist and label fields.
 
 Bug fixes:
 
@@ -21,6 +23,8 @@ Bug fixes:
 - :doc:`plugins/spotify` Removed old and undocumented config options
   `artist_field`, `album_field` and `track` that were causing issues with track
   matching. :bug:`5189`
+- :doc:`plugins/discogs` Fixed inconsistency in stripping disambiguation from
+  artists but not labels. :bug:`5366`
 
 For packagers:
 
@@ -35,6 +39,8 @@ Other changes:
 - Moved ``art.py`` utility module from ``beets`` into ``beetsplug`` namespace as
   it is not used in the core beets codebase. It can now be found in
   ``beetsplug._utils``.
+- :class:`beets.metadata_plugin.MetadataSourcePlugin`: Remove discogs specific
+  disambiguation stripping.
 
 2.4.0 (September 13, 2025)
 --------------------------

docs/plugins/discogs.rst

@@ -109,6 +109,9 @@ Other configurations available under ``discogs:`` are:
 - **search_limit**: The maximum number of results to return from Discogs. This
   is useful if you want to limit the number of results returned to speed up
   searches. Default: ``5``
+- **strip_disambiguation**: 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 it. Default: ``True``
 
 .. _discogs guidelines: https://support.discogs.com/hc/en-us/articles/360005055373-Database-Guidelines-12-Tracklisting#Index_Tracks_And_Headings
 

test/plugins/test_discogs.py

@@ -374,6 +374,83 @@ class DGAlbumInfoTest(BeetsTestCase):
         assert d.genre == "GENRE1, GENRE2"
         assert d.style is None
 
+    def test_strip_disambiguation(self):
+        """Test removing disambiguation from all disambiguated fields."""
+        data = {
+            "id": 123,
+            "uri": "https://www.discogs.com/release/123456-something",
+            "tracklist": [
+                {
+                    "title": "track",
+                    "position": "A",
+                    "type_": "track",
+                    "duration": "5:44",
+                    "artists": [
+                        {"name": "TEST ARTIST (5)", "tracks": "", "id": 11146}
+                    ],
+                }
+            ],
+            "artists": [
+                {"name": "ARTIST NAME (2)", "id": 321, "join": "&"},
+                {"name": "OTHER ARTIST (5)", "id": 321, "join": ""},
+            ],
+            "title": "title",
+            "labels": [
+                {
+                    "name": "LABEL NAME (5)",
+                    "catno": "catalog number",
+                }
+            ],
+        }
+        release = Bag(
+            data=data,
+            title=data["title"],
+            artists=[Bag(data=d) for d in data["artists"]],
+        )
+        d = DiscogsPlugin().get_album_info(release)
+        assert d.artist == "ARTIST NAME & OTHER ARTIST"
+        assert d.tracks[0].artist == "TEST ARTIST"
+        assert d.label == "LABEL NAME"
+
+    def test_strip_disambiguation_false(self):
+        """Test disabling disambiguation removal from all disambiguated fields."""
+        config["discogs"]["strip_disambiguation"] = False
+        data = {
+            "id": 123,
+            "uri": "https://www.discogs.com/release/123456-something",
+            "tracklist": [
+                {
+                    "title": "track",
+                    "position": "A",
+                    "type_": "track",
+                    "duration": "5:44",
+                    "artists": [
+                        {"name": "TEST ARTIST (5)", "tracks": "", "id": 11146}
+                    ],
+                }
+            ],
+            "artists": [
+                {"name": "ARTIST NAME (2)", "id": 321, "join": "&"},
+                {"name": "OTHER ARTIST (5)", "id": 321, "join": ""},
+            ],
+            "title": "title",
+            "labels": [
+                {
+                    "name": "LABEL NAME (5)",
+                    "catno": "catalog number",
+                }
+            ],
+        }
+        release = Bag(
+            data=data,
+            title=data["title"],
+            artists=[Bag(data=d) for d in data["artists"]],
+        )
+        d = DiscogsPlugin().get_album_info(release)
+        assert d.artist == "ARTIST NAME (2) & OTHER ARTIST (5)"
+        assert d.tracks[0].artist == "TEST ARTIST (5)"
+        assert d.label == "LABEL NAME (5)"
+
 
 @pytest.mark.parametrize(
     "formats, expected_media, expected_albumtype",