def track_distance(self, item, info):
last_data = last_match(item.path)
if not last_data:
# Match failed.
return 0.0, 0.0
dist, dist_max = 0.0, 0.0
# Track title distance.
dist += match.string_dist(last_data['title'],
info.title) \
* match.TRACK_TITLE_WEIGHT
dist_max += match.TRACK_TITLE_WEIGHT
# MusicBrainz track ID.
if last_data['track_mbid']:
# log.debug('Last track ID match: %s/%s' %
# (last_data['track_mbid'], track_data['id']))
if last_data['track_mbid'] != last_data['id']:
dist += match.TRACK_ID_WEIGHT
dist_max += match.TRACK_ID_WEIGHT
# log.debug('Last data: %s; distance: %f' %
# (str(last_data), dist/dist_max if dist_max > 0.0 else 0.0))
return dist * DISTANCE_SCALE, dist_max * DISTANCE_SCALE
def track_distance(self, item, info):
last_data = last_match(item.path)
if not last_data:
# Match failed.
return 0.0, 0.0
dist, dist_max = 0.0, 0.0
# Track title distance.
dist += match.string_dist(last_data['title'],
info.title) \
* match.TRACK_TITLE_WEIGHT
dist_max += match.TRACK_TITLE_WEIGHT
# MusicBrainz track ID.
if last_data['track_mbid']:
# log.debug('Last track ID match: %s/%s' %
# (last_data['track_mbid'], track_data['id']))
if last_data['track_mbid'] != last_data['id']:
dist += match.TRACK_ID_WEIGHT
dist_max += match.TRACK_ID_WEIGHT
# log.debug('Last data: %s; distance: %f' %
# (str(last_data), dist/dist_max if dist_max > 0.0 else 0.0))
return dist * DISTANCE_SCALE, dist_max * DISTANCE_SCALE
def album_distance(self, items, info):
last_artist, last_artist_id = get_cur_artist(
[item for item in items if item])
# Compare artist to MusicBrainz metadata.
dist, dist_max = 0.0, 0.0
if last_artist:
dist += match.string_dist(last_artist, info.artist) \
* match.ARTIST_WEIGHT
dist_max += match.ARTIST_WEIGHT
log.debug('Last artist (%s/%s) distance: %f' %
(last_artist, info.artist,
dist / dist_max if dist_max > 0.0 else 0.0))
#fixme: artist MBID currently ignored (as in vanilla tagger)
return dist, dist_max
def album_distance(self, items, info):
last_artist, last_artist_id = get_cur_artist(
[item for item in items if item]
)
# Compare artist to MusicBrainz metadata.
dist, dist_max = 0.0, 0.0
if last_artist:
dist += match.string_dist(last_artist, info.artist) \
* match.ARTIST_WEIGHT
dist_max += match.ARTIST_WEIGHT
log.debug('Last artist (%s/%s) distance: %f' %
(last_artist, info.artist,
dist/dist_max if dist_max > 0.0 else 0.0))
#fixme: artist MBID currently ignored (as in vanilla tagger)
return dist, dist_max
def test_parens_have_lower_weight(self):
dist1 = match.string_dist(u'One .Two.', u'One')
dist2 = match.string_dist(u'One (Two)', u'One')
self.assert_(dist2 < dist1)
def test_case_ignored(self):
dist = match.string_dist(u'Some String', u'sOME sTring')
self.assertEqual(dist, 0.0)
def test_different_strings(self):
dist = match.string_dist(u'Some String', u'Totally Different')
self.assertNotEqual(dist, 0.0)
def test_accented_characters(self):
dist = match.string_dist(u"\xe9\xe1\xf1", u"ean")
self.assertEqual(dist, 0.0)
def test_different_strings(self):
dist = match.string_dist('Some String', 'Totally Different')
self.assertNotEqual(dist, 0.0)
def test_ampersand_expansion(self):
dist = match.string_dist(u'And', u'&')
self.assertEqual(dist, 0.0)
def test_postfix_an(self):
dist = match.string_dist(u'An Album Title', u'Album Title, An')
self.assertEqual(dist, 0.0)
def test_postfix_an(self):
dist = match.string_dist('An Album Title', 'Album Title, An')
self.assertEqual(dist, 0.0)
def test_postfix_a(self):
dist = match.string_dist('A Song Title', 'Song Title, A')
self.assertEqual(dist, 0.0)
def test_postfix_the(self):
dist = match.string_dist('The Song Title', 'Song Title, The')
self.assertEqual(dist, 0.0)
def test_brackets_have_lower_weight(self):
dist1 = match.string_dist('One .Two.', 'One')
dist2 = match.string_dist('One [Two]', 'One')
self.assert_(dist2 < dist1)
def test_parens_have_lower_weight(self):
dist1 = match.string_dist('One .Two.', 'One')
dist2 = match.string_dist('One (Two)', 'One')
self.assert_(dist2 < dist1)
def test_leading_the_has_lower_weight(self):
dist1 = match.string_dist('XXX Band Name', 'Band Name')
dist2 = match.string_dist('The Band Name', 'Band Name')
self.assert_(dist2 < dist1)
def test_case_ignored(self):
dist = match.string_dist('Some String', 'sOME sTring')
self.assertEqual(dist, 0.0)
def test_ep_label_has_zero_weight(self):
dist = match.string_dist(u'My Song (EP)', u'My Song')
self.assertEqual(dist, 0.0)
def test_postfix_the(self):
dist = match.string_dist(u'The Song Title', u'Song Title, The')
self.assertEqual(dist, 0.0)
def test_postfix_a(self):
dist = match.string_dist("A Song Title", "Song Title, A")
self.assertEqual(dist, 0.0)
def test_solo_pattern(self):
# Just make sure these don't crash.
match.string_dist(u'The ', u'')
match.string_dist(u'(EP)', u'(EP)')
match.string_dist(u', An', u'')
def test_empty_strings(self):
dist = match.string_dist("", "")
self.assertEqual(dist, 0.0)
def test_solo_pattern(self):
# Just make sure these don't crash.
match.string_dist('The ', '')
match.string_dist('(EP)', '(EP)')
match.string_dist(', An', '')
def test_heuristic_does_not_harm_distance(self):
dist = match.string_dist("Untitled", "[Untitled]")
self.assertEqual(dist, 0.0)
def test_postfix_an(self):
dist = match.string_dist("An Album Title", "Album Title, An")
self.assertEqual(dist, 0.0)
def test_ampersand_expansion(self):
dist = match.string_dist('And', '&')
self.assertEqual(dist, 0.0)
def test_solo_pattern(self):
# Just make sure these don't crash.
match.string_dist("The ", "")
match.string_dist("(EP)", "(EP)")
match.string_dist(", An", "")
def test_accented_characters(self):
dist = match.string_dist(u'\xe9\xe1\xf1', u'ean')
self.assertEqual(dist, 0.0)
def test_ampersand_expansion(self):
dist = match.string_dist("And", "&")
self.assertEqual(dist, 0.0)
def test_equal_strings(self):
dist = match.string_dist("Some String", "Some String")
self.assertEqual(dist, 0.0)
def test_equal_strings(self):
dist = match.string_dist(u'Some String', u'Some String')
self.assertEqual(dist, 0.0)
def test_different_strings(self):
dist = match.string_dist("Some String", "Totally Different")
self.assertNotEqual(dist, 0.0)
def test_punctuation_ignored(self):
dist = match.string_dist(u'Some String', u'Some.String!')
self.assertEqual(dist, 0.0)
def test_punctuation_ignored(self):
dist = match.string_dist("Some String", "Some.String!")
self.assertEqual(dist, 0.0)
def test_leading_the_has_lower_weight(self):
dist1 = match.string_dist(u'XXX Band Name', u'Band Name')
dist2 = match.string_dist(u'The Band Name', u'Band Name')
self.assert_(dist2 < dist1)
def test_case_ignored(self):
dist = match.string_dist("Some String", "sOME sTring")
self.assertEqual(dist, 0.0)
def test_brackets_have_lower_weight(self):
dist1 = match.string_dist(u'One .Two.', u'One')
dist2 = match.string_dist(u'One [Two]', u'One')
self.assert_(dist2 < dist1)
def test_leading_the_has_lower_weight(self):
dist1 = match.string_dist("XXX Band Name", "Band Name")
dist2 = match.string_dist("The Band Name", "Band Name")
self.assert_(dist2 < dist1)
def test_featured_has_lower_weight(self):
dist1 = match.string_dist(u'My Song blah Someone', u'My Song')
dist2 = match.string_dist(u'My Song feat Someone', u'My Song')
self.assert_(dist2 < dist1)
def test_parens_have_lower_weight(self):
dist1 = match.string_dist("One .Two.", "One")
dist2 = match.string_dist("One (Two)", "One")
self.assert_(dist2 < dist1)
def test_postfix_a(self):
dist = match.string_dist(u'A Song Title', u'Song Title, A')
self.assertEqual(dist, 0.0)
def test_brackets_have_lower_weight(self):
dist1 = match.string_dist("One .Two.", "One")
dist2 = match.string_dist("One [Two]", "One")
self.assert_(dist2 < dist1)
def test_empty_strings(self):
dist = match.string_dist(u'', u'')
self.assertEqual(dist, 0.0)
def test_ep_label_has_zero_weight(self):
dist = match.string_dist("My Song (EP)", "My Song")
self.assertEqual(dist, 0.0)
def test_heuristic_does_not_harm_distance(self):
dist = match.string_dist(u'Untitled', u'[Untitled]')
self.assertEqual(dist, 0.0)
def test_featured_has_lower_weight(self):
dist1 = match.string_dist("My Song blah Someone", "My Song")
dist2 = match.string_dist("My Song feat Someone", "My Song")
self.assert_(dist2 < dist1)
def test_accented_characters(self):
dist = match.string_dist(u'\xe9\xe1\xf1', u'ean')
self.assertEqual(dist, 0.0)
def test_postfix_the(self):
dist = match.string_dist("The Song Title", "Song Title, The")
self.assertEqual(dist, 0.0)
