danvics/soulsync
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions 1

Add ignore cross-album duplicates setting to Duplicate Detector

Browse source 
Tracks on different albums (e.g., same song on a studio album and a
compilation) are no longer flagged as duplicates by default, keeping
albums complete. Setting is enabled by default and configurable via
the job settings UI.
This commit is contained in:
Broque Thomas 2026-03-18 17:42:49 -07:00
parent 0428a4565d
commit e1b203405f
1 changed files with 9 additions and 1 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

10
core/repair_jobs/duplicate_detector.py
View file

@ -24,7 +24,9 @@ class DuplicateDetectorJob(RepairJob):
'(file path, format, bitrate) so you can decide which to keep.\n\n' '(file path, format, bitrate) so you can decide which to keep.\n\n'
'Settings:\n' 'Settings:\n'
'- Title Similarity: How closely titles must match to be considered duplicates (0.0 - 1.0)\n' '- Title Similarity: How closely titles must match to be considered duplicates (0.0 - 1.0)\n'
'- Artist Similarity: How closely artist names must match (0.0 - 1.0)' '- Artist Similarity: How closely artist names must match (0.0 - 1.0)\n'
'- Ignore Cross-Album: When enabled, tracks on different albums are not flagged as duplicates '
'(keeps your albums complete even if the same song appears on multiple albums)'
) )
icon = 'repair-icon-duplicate' icon = 'repair-icon-duplicate'
default_enabled = False default_enabled = False
@ -32,6 +34,7 @@ class DuplicateDetectorJob(RepairJob):
default_settings = { default_settings = {
'title_similarity': 0.85, 'title_similarity': 0.85,
'artist_similarity': 0.80, 'artist_similarity': 0.80,
'ignore_cross_album': True,
} }
auto_fix = False auto_fix = False
@ -41,6 +44,7 @@ class DuplicateDetectorJob(RepairJob):
settings = self._get_settings(context) settings = self._get_settings(context)
title_threshold = settings.get('title_similarity', 0.85) title_threshold = settings.get('title_similarity', 0.85)
artist_threshold = settings.get('artist_similarity', 0.80) artist_threshold = settings.get('artist_similarity', 0.80)
ignore_cross_album = settings.get('ignore_cross_album', True)
# Fetch all tracks with artist/album names via JOIN # Fetch all tracks with artist/album names via JOIN
tracks = [] tracks = []
@ -140,6 +144,10 @@ class DuplicateDetectorJob(RepairJob):
if artist_sim < artist_threshold: if artist_sim < artist_threshold:
continue continue
# Skip cross-album duplicates — same song on different albums is intentional
if ignore_cross_album and t1['album'] and t2['album'] and t1['album'] != t2['album']:
continue
group.append(t2) group.append(t2)
if len(group) >= 2: if len(group) >= 2: