def distance(items, info):
"""Determines how "significant" an album metadata change would be.
Returns a float in [0.0,1.0]. The list of items must be ordered.
"""
cur_artist, cur_album = current_metadata(items)
cur_artist = cur_artist or ''
cur_album = cur_album or ''
# These accumulate the possible distance components. The final
# distance will be dist/dist_max.
dist = 0.0
dist_max = 0.0
# Artist/album metadata.
dist += string_dist(cur_artist, info['artist']) * ARTIST_WEIGHT
dist_max += ARTIST_WEIGHT
dist += string_dist(cur_album, info['album']) * ALBUM_WEIGHT
dist_max += ALBUM_WEIGHT
# Track distances.
for i, (item, track_data) in enumerate(zip(items, info['tracks'])):
dist += track_distance(item, track_data, i+1) * TRACK_WEIGHT
dist_max += TRACK_WEIGHT
# Plugin distances.
plugin_d, plugin_dm = plugins.album_distance(items, info)
dist += plugin_d
dist_max += plugin_dm
# Normalize distance, avoiding divide-by-zero.
if dist_max == 0.0:
return 0.0
else:
return dist/dist_max
def distance(items, album_info, mapping):
"""Determines how "significant" an album metadata change would be.
Returns a float in [0.0,1.0]. `album_info` is an AlbumInfo object
reflecting the album to be compared. `items` is a sequence of all
Item objects that will be matched (order is not important).
`mapping` is a dictionary mapping Items to TrackInfo objects; the
keys are a subset of `items` and the values are a subset of
`album_info.tracks`.
"""
cur_artist, cur_album, _ = current_metadata(items)
cur_artist = cur_artist or ''
cur_album = cur_album or ''
# These accumulate the possible distance components. The final
# distance will be dist/dist_max.
dist = 0.0
dist_max = 0.0
# Artist/album metadata.
if not album_info.va:
dist += string_dist(cur_artist, album_info.artist) * ARTIST_WEIGHT
dist_max += ARTIST_WEIGHT
dist += string_dist(cur_album, album_info.album) * ALBUM_WEIGHT
dist_max += ALBUM_WEIGHT
# Matched track distances.
for item, track in mapping.iteritems():
dist += track_distance(item, track, album_info.va) * TRACK_WEIGHT
dist_max += TRACK_WEIGHT
# Extra and unmatched tracks.
for track in set(album_info.tracks) - set(mapping.values()):
dist += MISSING_WEIGHT
dist_max += MISSING_WEIGHT
for item in set(items) - set(mapping.keys()):
dist += UNMATCHED_WEIGHT
dist_max += UNMATCHED_WEIGHT
# Plugin distances.
plugin_d, plugin_dm = plugins.album_distance(items, album_info, mapping)
dist += plugin_d
dist_max += plugin_dm
# Normalize distance, avoiding divide-by-zero.
if dist_max == 0.0:
return 0.0
else:
return dist / dist_max
def distance(items, album_info, mapping):
"""Determines how "significant" an album metadata change would be.
Returns a float in [0.0,1.0]. `album_info` is an AlbumInfo object
reflecting the album to be compared. `items` is a sequence of all
Item objects that will be matched (order is not important).
`mapping` is a dictionary mapping Items to TrackInfo objects; the
keys are a subset of `items` and the values are a subset of
`album_info.tracks`.
"""
cur_artist, cur_album, _ = current_metadata(items)
cur_artist = cur_artist or ''
cur_album = cur_album or ''
# These accumulate the possible distance components. The final
# distance will be dist/dist_max.
dist = 0.0
dist_max = 0.0
# Artist/album metadata.
if not album_info.va:
dist += string_dist(cur_artist, album_info.artist) * ARTIST_WEIGHT
dist_max += ARTIST_WEIGHT
dist += string_dist(cur_album, album_info.album) * ALBUM_WEIGHT
dist_max += ALBUM_WEIGHT
# Matched track distances.
for item, track in mapping.iteritems():
dist += track_distance(item, track, album_info.va) * TRACK_WEIGHT
dist_max += TRACK_WEIGHT
# Extra and unmatched tracks.
for track in set(album_info.tracks) - set(mapping.values()):
dist += MISSING_WEIGHT
dist_max += MISSING_WEIGHT
for item in set(items) - set(mapping.keys()):
dist += UNMATCHED_WEIGHT
dist_max += UNMATCHED_WEIGHT
# Plugin distances.
plugin_d, plugin_dm = plugins.album_distance(items, album_info, mapping)
dist += plugin_d
dist_max += plugin_dm
# Normalize distance, avoiding divide-by-zero.
if dist_max == 0.0:
return 0.0
else:
return dist / dist_max
def distance(items, album_info):
"""Determines how "significant" an album metadata change would be.
Returns a float in [0.0,1.0]. The list of items must be ordered.
"""
cur_artist, cur_album, _ = current_metadata(items)
cur_artist = cur_artist or ''
cur_album = cur_album or ''
# These accumulate the possible distance components. The final
# distance will be dist/dist_max.
dist = 0.0
dist_max = 0.0
# Artist/album metadata.
if not album_info.va:
dist += string_dist(cur_artist, album_info.artist) * ARTIST_WEIGHT
dist_max += ARTIST_WEIGHT
dist += string_dist(cur_album, album_info.album) * ALBUM_WEIGHT
dist_max += ALBUM_WEIGHT
# Track distances.
for i, (item, track_info) in enumerate(zip(items, album_info.tracks)):
if item:
dist += track_distance(item, track_info, i+1, album_info.va) * \
TRACK_WEIGHT
dist_max += TRACK_WEIGHT
else:
dist += MISSING_WEIGHT
dist_max += MISSING_WEIGHT
# Plugin distances.
plugin_d, plugin_dm = plugins.album_distance(items, album_info)
dist += plugin_d
dist_max += plugin_dm
# Normalize distance, avoiding divide-by-zero.
if dist_max == 0.0:
return 0.0
else:
return dist / dist_max
def distance(items, album_info, mapping):
"""Determines how "significant" an album metadata change would be.
Returns a Distance object. `album_info` is an AlbumInfo object
reflecting the album to be compared. `items` is a sequence of all
Item objects that will be matched (order is not important).
`mapping` is a dictionary mapping Items to TrackInfo objects; the
keys are a subset of `items` and the values are a subset of
`album_info.tracks`.
"""
likelies, _ = current_metadata(items)
dist = hooks.Distance()
# Artist, if not various.
if not album_info.va:
dist.add_string('artist', likelies['artist'], album_info.artist)
# Album.
dist.add_string('album', likelies['album'], album_info.album)
# Current or preferred media.
if album_info.media:
# Preferred media options.
patterns = config['match']['preferred']['media'].as_str_seq()
options = [re.compile(r'(\d+x)?(%s)' % pat, re.I) for pat in patterns]
if options:
dist.add_priority('media', album_info.media, options)
# Current media.
elif likelies['media']:
dist.add_equality('media', album_info.media, likelies['media'])
# Mediums.
if likelies['disctotal'] and album_info.mediums:
dist.add_number('mediums', likelies['disctotal'], album_info.mediums)
# Prefer earliest release.
if album_info.year and config['match']['preferred']['original_year']:
# Assume 1889 (earliest first gramophone discs) if we don't know the
# original year.
original = album_info.original_year or 1889
diff = abs(album_info.year - original)
diff_max = abs(datetime.date.today().year - original)
dist.add_ratio('year', diff, diff_max)
# Year.
elif likelies['year'] and album_info.year:
if likelies['year'] in (album_info.year, album_info.original_year):
# No penalty for matching release or original year.
dist.add('year', 0.0)
elif album_info.original_year:
# Prefer matchest closest to the release year.
diff = abs(likelies['year'] - album_info.year)
diff_max = abs(datetime.date.today().year -
album_info.original_year)
dist.add_ratio('year', diff, diff_max)
else:
# Full penalty when there is no original year.
dist.add('year', 1.0)
# Preferred countries.
patterns = config['match']['preferred']['countries'].as_str_seq()
options = [re.compile(pat, re.I) for pat in patterns]
if album_info.country and options:
dist.add_priority('country', album_info.country, options)
# Country.
elif likelies['country'] and album_info.country:
dist.add_string('country', likelies['country'], album_info.country)
# Label.
if likelies['label'] and album_info.label:
dist.add_string('label', likelies['label'], album_info.label)
# Catalog number.
if likelies['catalognum'] and album_info.catalognum:
dist.add_string('catalognum', likelies['catalognum'],
album_info.catalognum)
# Disambiguation.
if likelies['albumdisambig'] and album_info.albumdisambig:
dist.add_string('albumdisambig', likelies['albumdisambig'],
album_info.albumdisambig)
# Album ID.
if likelies['mb_albumid']:
dist.add_equality('album_id', likelies['mb_albumid'],
album_info.album_id)
# Tracks.
dist.tracks = {}
for item, track in mapping.iteritems():
dist.tracks[track] = track_distance(item, track, album_info.va)
dist.add('tracks', dist.tracks[track].distance)
# Missing tracks.
for i in range(len(album_info.tracks) - len(mapping)):
dist.add('missing_tracks', 1.0)
# Unmatched tracks.
for i in range(len(items) - len(mapping)):
dist.add('unmatched_tracks', 1.0)
# Plugins.
dist.update(plugins.album_distance(items, album_info, mapping))
return dist
def distance(items, album_info, mapping):
"""Determines how "significant" an album metadata change would be.
Returns a Distance object. `album_info` is an AlbumInfo object
reflecting the album to be compared. `items` is a sequence of all
Item objects that will be matched (order is not important).
`mapping` is a dictionary mapping Items to TrackInfo objects; the
keys are a subset of `items` and the values are a subset of
`album_info.tracks`.
"""
likelies, _ = current_metadata(items)
dist = hooks.Distance()
# Artist, if not various.
if not album_info.va:
dist.add_string('artist', likelies['artist'], album_info.artist)
# Album.
dist.add_string('album', likelies['album'], album_info.album)
# Current or preferred media.
if album_info.media:
# Preferred media options.
patterns = config['match']['preferred']['media'].as_str_seq()
options = [re.compile(r'(\d+x)?(%s)' % pat, re.I) for pat in patterns]
if options:
dist.add_priority('media', album_info.media, options)
# Current media.
elif likelies['media']:
dist.add_equality('media', album_info.media, likelies['media'])
# Mediums.
if likelies['disctotal'] and album_info.mediums:
dist.add_number('mediums', likelies['disctotal'], album_info.mediums)
# Prefer earliest release.
if album_info.year and config['match']['preferred']['original_year']:
# Assume 1889 (earliest first gramophone discs) if we don't know the
# original year.
original = album_info.original_year or 1889
diff = abs(album_info.year - original)
diff_max = abs(datetime.date.today().year - original)
dist.add_ratio('year', diff, diff_max)
# Year.
elif likelies['year'] and album_info.year:
if likelies['year'] in (album_info.year, album_info.original_year):
# No penalty for matching release or original year.
dist.add('year', 0.0)
elif album_info.original_year:
# Prefer matchest closest to the release year.
diff = abs(likelies['year'] - album_info.year)
diff_max = abs(datetime.date.today().year -
album_info.original_year)
dist.add_ratio('year', diff, diff_max)
else:
# Full penalty when there is no original year.
dist.add('year', 1.0)
# Preferred countries.
patterns = config['match']['preferred']['countries'].as_str_seq()
options = [re.compile(pat, re.I) for pat in patterns]
if album_info.country and options:
dist.add_priority('country', album_info.country, options)
# Country.
elif likelies['country'] and album_info.country:
dist.add_string('country', likelies['country'], album_info.country)
# Label.
if likelies['label'] and album_info.label:
dist.add_string('label', likelies['label'], album_info.label)
# Catalog number.
if likelies['catalognum'] and album_info.catalognum:
dist.add_string('catalognum', likelies['catalognum'],
album_info.catalognum)
# Disambiguation.
if likelies['albumdisambig'] and album_info.albumdisambig:
dist.add_string('albumdisambig', likelies['albumdisambig'],
album_info.albumdisambig)
# Album ID.
if likelies['mb_albumid']:
dist.add_equality('album_id', likelies['mb_albumid'],
album_info.album_id)
# Tracks.
dist.tracks = {}
for item, track in mapping.iteritems():
dist.tracks[track] = track_distance(item, track, album_info.va)
dist.add('tracks', dist.tracks[track].distance)
# Missing tracks.
for i in range(len(album_info.tracks) - len(mapping)):
dist.add('missing_tracks', 1.0)
# Unmatched tracks.
for i in range(len(items) - len(mapping)):
dist.add('unmatched_tracks', 1.0)
# Plugins.
dist.update(plugins.album_distance(items, album_info, mapping))
return dist
def distance(items, album_info, mapping):
"""Determines how "significant" an album metadata change would be.
Returns a float in [0.0,1.0]. `album_info` is an AlbumInfo object
reflecting the album to be compared. `items` is a sequence of all
Item objects that will be matched (order is not important).
`mapping` is a dictionary mapping Items to TrackInfo objects; the
keys are a subset of `items` and the values are a subset of
`album_info.tracks`.
"""
likelies, _ = current_metadata(items)
# These accumulate the possible distance components. The final
# distance will be dist/dist_max.
dist = 0.0
dist_max = 0.0
# Artist/album metadata.
if not album_info.va:
dist += string_dist(likelies['artist'], album_info.artist) * \
weights['artist'].as_number()
dist_max += weights['artist'].as_number()
dist += string_dist(likelies['album'], album_info.album) * \
weights['album'].as_number()
dist_max += weights['album'].as_number()
# Year. No penalty for matching release or original year.
if likelies['year'] and album_info.year:
if likelies['year'] not in (album_info.year, album_info.original_year):
diff = abs(album_info.year - likelies['year'])
if diff:
dist += (1.0 - 1.0 / diff) * weights['year'].as_number()
dist_max += weights['year'].as_number()
# Actual or preferred media.
if album_info.media:
compare_media = likelies['media'] or \
config['match']['preferred_media'].get()
if compare_media and compare_media.lower() != album_info.media.lower():
dist += weights['media'].as_number()
dist_max += weights['media'].as_number()
# MusicBrainz album ID.
if likelies['mb_albumid']:
if likelies['mb_albumid'] != album_info.album_id:
dist += weights['album_id'].as_number()
dist_max += weights['album_id'].as_number()
# Apply a small penalty for differences across many minor metadata. This
# helps prioritise releases that are nearly identical.
if likelies['disctotal']:
if likelies['disctotal'] != album_info.mediums:
dist += weights['minor'].as_number()
dist_max += weights['minor'].as_number()
if likelies['label'] and album_info.label:
dist += string_dist(likelies['label'], album_info.label) * \
weights['minor'].as_number()
dist_max += weights['minor'].as_number()
if likelies['catalognum'] and album_info.catalognum:
dist += string_dist(likelies['catalognum'],
album_info.catalognum) * \
weights['minor'].as_number()
dist_max += weights['minor'].as_number()
if likelies['country'] and album_info.country:
dist += string_dist(likelies['country'],
album_info.country) * \
weights['minor'].as_number()
dist_max += weights['minor'].as_number()
if likelies['albumdisambig'] and album_info.albumdisambig:
dist += string_dist(likelies['albumdisambig'],
album_info.albumdisambig) * \
weights['minor'].as_number()
dist_max += weights['minor'].as_number()
# Matched track distances.
for item, track in mapping.iteritems():
dist += track_distance(item, track, album_info.va) * \
weights['track'].as_number()
dist_max += weights['track'].as_number()
# Extra and unmatched tracks.
for track in set(album_info.tracks) - set(mapping.values()):
dist += weights['missing'].as_number()
dist_max += weights['missing'].as_number()
for item in set(items) - set(mapping.keys()):
dist += weights['unmatched'].as_number()
dist_max += weights['unmatched'].as_number()
# Plugin distances.
plugin_d, plugin_dm = plugins.album_distance(items, album_info, mapping)
dist += plugin_d
dist_max += plugin_dm
# Normalize distance, avoiding divide-by-zero.
if dist_max == 0.0:
return 0.0
else:
return dist / dist_max
def distance(items, album_info, mapping):
"""Determines how "significant" an album metadata change would be.
Returns a float in [0.0,1.0]. `album_info` is an AlbumInfo object
reflecting the album to be compared. `items` is a sequence of all
Item objects that will be matched (order is not important).
`mapping` is a dictionary mapping Items to TrackInfo objects; the
keys are a subset of `items` and the values are a subset of
`album_info.tracks`.
"""
likelies, _ = current_metadata(items)
# These accumulate the possible distance components. The final
# distance will be dist/dist_max.
dist = 0.0
dist_max = 0.0
# Artist/album metadata.
if not album_info.va:
dist += string_dist(likelies['artist'], album_info.artist) * \
weights['artist'].as_number()
dist_max += weights['artist'].as_number()
dist += string_dist(likelies['album'], album_info.album) * \
weights['album'].as_number()
dist_max += weights['album'].as_number()
# Year. No penalty for matching release or original year.
if likelies['year'] and album_info.year:
if likelies['year'] not in (album_info.year, album_info.original_year):
diff = abs(album_info.year - likelies['year'])
if diff:
dist += (1.0 - 1.0 / diff) * weights['year'].as_number()
dist_max += weights['year'].as_number()
# Actual or preferred media.
preferred_media = config['match']['preferred_media'].get()
if likelies['media'] and album_info.media:
dist += string_dist(likelies['media'], album_info.media) * \
weights['media'].as_number()
dist_max += weights['media'].as_number()
elif album_info.media and preferred_media:
dist += string_dist(album_info.media, preferred_media) * \
weights['media'].as_number()
dist_max += weights['media'].as_number()
# MusicBrainz album ID.
if likelies['mb_albumid']:
if likelies['mb_albumid'] != album_info.album_id:
dist += weights['album_id'].as_number()
dist_max += weights['album_id'].as_number()
# Apply a small penalty for differences across many minor metadata. This
# helps prioritise releases that are nearly identical.
if likelies['disctotal']:
if likelies['disctotal'] != album_info.mediums:
dist += weights['minor'].as_number()
dist_max += weights['minor'].as_number()
if likelies['label'] and album_info.label:
dist += string_dist(likelies['label'], album_info.label) * \
weights['minor'].as_number()
dist_max += weights['minor'].as_number()
if likelies['catalognum'] and album_info.catalognum:
dist += string_dist(likelies['catalognum'],
album_info.catalognum) * \
weights['minor'].as_number()
dist_max += weights['minor'].as_number()
if likelies['country'] and album_info.country:
dist += string_dist(likelies['country'],
album_info.country) * \
weights['minor'].as_number()
dist_max += weights['minor'].as_number()
if likelies['albumdisambig'] and album_info.albumdisambig:
dist += string_dist(likelies['albumdisambig'],
album_info.albumdisambig) * \
weights['minor'].as_number()
dist_max += weights['minor'].as_number()
# Matched track distances.
for item, track in mapping.iteritems():
dist += track_distance(item, track, album_info.va) * \
weights['track'].as_number()
dist_max += weights['track'].as_number()
# Extra and unmatched tracks.
for track in set(album_info.tracks) - set(mapping.values()):
dist += weights['missing'].as_number()
dist_max += weights['missing'].as_number()
for item in set(items) - set(mapping.keys()):
dist += weights['unmatched'].as_number()
dist_max += weights['unmatched'].as_number()
# Plugin distances.
plugin_d, plugin_dm = plugins.album_distance(items, album_info, mapping)
dist += plugin_d
dist_max += plugin_dm
# Normalize distance, avoiding divide-by-zero.
if dist_max == 0.0:
return 0.0
else:
return dist / dist_max
def distance(items, album_info, mapping):
"""Determines how "significant" an album metadata change would be.
Returns a Distance object. `album_info` is an AlbumInfo object
reflecting the album to be compared. `items` is a sequence of all
Item objects that will be matched (order is not important).
`mapping` is a dictionary mapping Items to TrackInfo objects; the
keys are a subset of `items` and the values are a subset of
`album_info.tracks`.
"""
likelies, consensus = current_metadata(items)
dist = hooks.Distance()
# Artist, if not various.
if not album_info.va:
dist.add_string('artist', likelies['artist'], album_info.artist)
# Album.
dist.add_string('album', likelies['album'], album_info.album)
# Current or preferred media.
if album_info.media:
# Preferred media options.
patterns = config['match']['preferred']['media'].as_str_seq()
options = [re.compile(r'(\d+x)?(%s)' % pat, re.I) for pat in patterns]
if options:
dist.add_priority('media', album_info.media, options)
# Current media.
elif likelies['media']:
dist.add_equality('media', album_info.media, likelies['media'])
# Mediums.
if likelies['disctotal'] and album_info.mediums:
dist.add_number('mediums', likelies['disctotal'], album_info.mediums)
# Prefer earliest release.
if album_info.year and config['match']['preferred']['original_year']:
# Assume 1889 (earliest first gramophone discs) if we don't know the
# original year.
original = album_info.original_year or 1889
diff = abs(album_info.year - original)
diff_max = abs(datetime.date.today().year - original)
dist.add_ratio('year', diff, diff_max)
# Year.
elif likelies['year'] and album_info.year:
if likelies['year'] in (album_info.year, album_info.original_year):
# No penalty for matching release or original year.
dist.add('year', 0.0)
elif album_info.original_year:
# Prefer matchest closest to the release year.
diff = abs(likelies['year'] - album_info.year)
diff_max = abs(datetime.date.today().year -
album_info.original_year)
dist.add_ratio('year', diff, diff_max)
else:
# Full penalty when there is no original year.
dist.add('year', 1.0)
# Preferred countries.
patterns = config['match']['preferred']['countries'].as_str_seq()
options = [re.compile(pat, re.I) for pat in patterns]
if album_info.country and options:
dist.add_priority('country', album_info.country, options)
# Country.
elif likelies['country'] and album_info.country:
dist.add_string('country', likelies['country'], album_info.country)
# Label.
if likelies['label'] and album_info.label:
dist.add_string('label', likelies['label'], album_info.label)
# Catalog number.
if likelies['catalognum'] and album_info.catalognum:
dist.add_string('catalognum', likelies['catalognum'],
album_info.catalognum)
# Disambiguation.
if likelies['albumdisambig'] and album_info.albumdisambig:
dist.add_string('albumdisambig', likelies['albumdisambig'],
album_info.albumdisambig)
# Album ID.
if likelies['mb_albumid']:
dist.add_equality('album_id', likelies['mb_albumid'],
album_info.album_id)
# Tracks.
dist.tracks = {}
for item, track in mapping.items():
dist.tracks[track] = track_distance(item, track, album_info.va)
dist.add('tracks', dist.tracks[track].distance)
# Track totals
medium_totals = {}
for track in album_info.tracks:
if track.medium_total and track.medium not in medium_totals:
medium_totals[track.medium] = track.medium_total
medium_all_total = sum(medium_totals.values())
# If we have a consensus, and it's the same as the all-disc total,
# short-circuit the logic and take the simple approach.
if consensus and likelies['tracktotal'] and likelies['tracktotal'] == medium_all_total:
pass
else:
# The imported track may have tags for per disc numbering or not,
# so check for both.
def item_disc(i):
return i.disc or 0
items_by_disc = sorted(mapping.keys(), key=item_disc)
grouped_items = groupby(items_by_disc, item_disc)
for disc, disc_items in grouped_items:
medium_total = medium_totals.get(disc, 0)
tracktotals = set(i.tracktotal for i in disc_items if i.tracktotal)
for tracktotal in tracktotals:
if tracktotal != medium_all_total and medium_total:
dist.add_number('tracktotal', medium_total, tracktotal)
# Missing tracks.
for i in range(len(album_info.tracks) - len(mapping)):
dist.add('missing_tracks', 1.0)
# Unmatched tracks.
for i in range(len(items) - len(mapping)):
dist.add('unmatched_tracks', 1.0)
# Plugins.
dist.update(plugins.album_distance(items, album_info, mapping))
return dist
