Files
UltraSinger/src/modules/musicbrainz_client.py
T
2024-12-21 23:54:01 +01:00

205 lines
7.0 KiB
Python

import musicbrainzngs
import string
from Levenshtein import ratio
from dataclasses import dataclass
from typing import Optional
from src.Settings import Settings
from modules.console_colors import ULTRASINGER_HEAD, blue_highlighted, red_highlighted
@dataclass
class SongInfo:
title: str
artist: str
year: Optional[str] = None
genres: Optional[str] = None
cover_image_data: Optional[bytes] = None
cover_url: Optional[str] = None
title_filter = [
"official video",
"official music video",
"Offizielles Musikvideo",
]
def __clean_string(s: str) -> str:
return s.translate(str.maketrans('', '', string.punctuation)).lower().strip()
def search_musicbrainz(title: str, artist) -> SongInfo:
# Musicbrainz API documentation
# https://python-musicbrainzngs.readthedocs.io/en/latest/api/
musicbrainzngs.set_useragent("UltraSinger", Settings.APP_VERSION, "https://github.com/rakuri255/UltraSinger")
# remove from search_string "official video"
# todo: do we need filter?
origin_title = title
for filter in title_filter:
title = title.lower().replace(filter.lower(), "").strip()
if artist is not None:
artist = artist.lower().replace(filter.lower(), "").strip()
if artist is None:
recording = __single_line_search(title)
else:
recording = __multi_line_search(artist, title)
if recording is None:
print(f"{ULTRASINGER_HEAD} {red_highlighted('No match found')}")
return SongInfo(title=origin_title, artist="Unknown Artist")
artist = recording['artist-credit-phrase']
title = recording['title']
print(
f"{ULTRASINGER_HEAD} Found data on Musicbrainz: Artist={blue_highlighted(artist)} Title={blue_highlighted(title)}")
year = __get_year(recording)
genres = __get_genres(recording)
image_data, image_url = __get_image(recording)
return SongInfo(title=title, artist=artist, year=year, genres=genres, cover_image_data=image_data, cover_url=image_url)
def __single_line_search(search_string):
search_string = __clean_string(search_string)
search_string = __filter_words(search_string)
artists = musicbrainzngs.search_artists(search_string, limit=10, artist=search_string)
recordings = musicbrainzngs.search_recordings(search_string, limit=100, artistname=search_string)
found_artist = None
for record in recordings['recording-list']:
if found_artist is not None:
break
for artist_credit in record['artist-credit']:
if found_artist is not None:
break
if isinstance(artist_credit, str):
continue
# todo: there is also an "alias-list". Maybe search also there?
for artist in artists['artist-list']:
if artist_credit['artist'] and artist_credit['artist']['id'] == artist['id']:
found_artist = record['artist-credit-phrase']
break
if found_artist is None:
return None
recordings = [x for x in recordings['recording-list'] if
__clean_string(x['artist-credit-phrase']) == __clean_string(found_artist)]
recording = None
for record in recordings:
if __clean_string(record['title']) in __clean_string(search_string):
recording = record
return recording
def __filter_words(search_string):
for filter in title_filter:
search_string = search_string.lower().replace(filter.lower(), "").strip()
return search_string
def __multi_line_search(artist: str, title: str):
# Try both combinations since we don't know which one is the artist and which one is the title
artist1, title1 = artist, title
artist2, title2 = title, artist
result1 = musicbrainzngs.search_recordings(recording=title1, limit=10, artist=artist1, artistname=artist1)
result2 = musicbrainzngs.search_recordings(recording=title2, limit=10, artist=artist2, artistname=artist2)
# Filter result to ['artist-credit-phrase'] == artist
record1 = [x for x in result1['recording-list'] if
__clean_string(x['artist-credit-phrase']) == __clean_string(artist1)]
record2 = [x for x in result2['recording-list'] if
__clean_string(x['artist-credit-phrase']) == __clean_string(artist2)]
if len(record1) > 0 and len(record2) > 0:
best_match1 = max(record1, key=lambda x: ratio(__clean_string(x['title']), __clean_string(title1)))
best_match2 = max(record2, key=lambda x: ratio(__clean_string(x['title']), __clean_string(title2)))
is_match1 = ratio(__clean_string(title1), __clean_string(best_match1['title'])) > ratio(__clean_string(title2),
__clean_string(
best_match2[
'title']))
if is_match1:
recording = record1[0]
else:
recording = record2[0]
elif len(record1) > 0:
recording = record1[0]
elif len(record2) > 0:
recording = record2[0]
elif result1['recording-count'] > 0: # Artist = Title
recording = result1['recording-list'][0]
elif result2['recording-count'] > 0: # Artist = Title
recording = result2['recording-list'][0]
else:
recording = None
return recording
def __get_image(recording) -> (bytes, str):
image_data = None
image_url = None
if 'release-list' in recording:
for release in recording['release-list']:
try:
image_data = musicbrainzngs.get_image_front(release['id'])
image_list = musicbrainzngs.get_image_list(release['id'])
for image in image_list['images']:
if image['front']:
image_url = image['image']
break
break
except musicbrainzngs.ResponseError:
continue
if image_data is not None:
print(f"{ULTRASINGER_HEAD} Found cover image")
return image_data, image_url
def __get_year(recording):
year = None
if 'release-list' not in recording:
return year
release_group_id = recording['release-list'][0]['release-group']['id']
release_group = musicbrainzngs.get_release_group_by_id(release_group_id)
if 'first-release-date' not in release_group['release-group']:
return year
year = release_group['release-group']['first-release-date'].strip()
year = year.split('-')[0]
if year is not None:
print(f"{ULTRASINGER_HEAD} Found year: {blue_highlighted(year)}")
return year
def __get_genres(recording) -> str:
# todo secondary-type-list ??
genres = None
if 'tag-list' in recording:
genres = ""
for tag in recording['tag-list']:
genres += f"{tag['name'].strip()},"
if genres is not None:
print(f"{ULTRASINGER_HEAD} Found genres: {blue_highlighted(genres)}")
return genres