music-kraken-core/src/music_kraken/database/database.py

338 lines
12 KiB
Python
Raw Normal View History

2022-11-22 08:21:28 +00:00
from typing import List
2022-10-27 12:15:18 +00:00
import sqlite3
import os
import logging
import json
2022-11-29 10:47:46 +00:00
from pkg_resources import resource_string
2022-10-27 12:15:18 +00:00
2022-12-01 12:15:30 +00:00
from .song import (
Song,
2022-12-01 15:14:59 +00:00
Lyrics,
Metadata,
Target,
Artist,
Source
2022-12-01 12:15:30 +00:00
)
from .get_song import get_song_from_response
2022-11-27 14:07:02 +00:00
from ..utils.shared import (
DATABASE_LOGGER
)
2022-10-27 17:53:12 +00:00
2022-11-27 14:07:02 +00:00
logger = DATABASE_LOGGER
2022-11-25 17:27:48 +00:00
2022-11-06 17:10:00 +00:00
class Database:
2022-11-29 10:47:46 +00:00
def __init__(self, path_to_db: str, reset_anyways: bool = False):
2022-11-06 17:10:00 +00:00
self.path_to_db = path_to_db
self.connection = sqlite3.connect(self.path_to_db)
self.cursor = self.connection.cursor()
# init database
2022-11-29 10:47:46 +00:00
self.init_db(reset_anyways=reset_anyways)
2022-11-06 17:10:00 +00:00
2022-11-29 10:47:46 +00:00
def init_db(self, reset_anyways: bool = False):
2022-11-06 17:10:00 +00:00
# check if db exists
exists = True
try:
query = 'SELECT * FROM track;'
self.cursor.execute(query)
_ = self.cursor.fetchall()
except sqlite3.OperationalError:
exists = False
if not exists:
2022-11-25 17:27:48 +00:00
logger.info("Database does not exist yet.")
2022-11-06 17:10:00 +00:00
if reset_anyways or not exists:
# reset the database if reset_anyways is true or if an error has been thrown previously.
2022-11-29 10:47:46 +00:00
logger.info(f"Reseting the database.")
2022-11-06 17:10:00 +00:00
2022-11-29 10:47:46 +00:00
query = resource_string("music_kraken", "static_files/temp_database_structure.sql").decode('utf-8')
self.cursor.executescript(query)
self.connection.commit()
2022-11-06 17:10:00 +00:00
def add_artist(
self,
musicbrainz_artistid: str,
artist: str = None
):
query = "INSERT OR REPLACE INTO artist (id, name) VALUES (?, ?);"
values = musicbrainz_artistid, artist
self.cursor.execute(query, values)
self.connection.commit()
def add_release_group(
self,
musicbrainz_releasegroupid: str,
artist_ids: list,
albumartist: str = None,
albumsort: int = None,
musicbrainz_albumtype: str = None,
compilation: str = None,
album_artist_id: str = None
):
# add adjacency
adjacency_list = []
for artist_id in artist_ids:
adjacency_list.append((artist_id, musicbrainz_releasegroupid))
adjacency_values = tuple(adjacency_list)
adjacency_query = "INSERT OR REPLACE INTO artist_release_group (artist_id, release_group_id) VALUES (?, ?);"
self.cursor.executemany(adjacency_query, adjacency_values)
self.connection.commit()
# add release group
query = "INSERT OR REPLACE INTO release_group (id, albumartist, albumsort, musicbrainz_albumtype, compilation, album_artist_id) VALUES (?, ?, ?, ?, ?, ?);"
values = musicbrainz_releasegroupid, albumartist, albumsort, musicbrainz_albumtype, compilation, album_artist_id
self.cursor.execute(query, values)
self.connection.commit()
def add_release(
self,
musicbrainz_albumid: str,
release_group_id: str,
title: str = None,
copyright_: str = None,
album_status: str = None,
language: str = None,
year: str = None,
date: str = None,
country: str = None,
barcode: str = None
):
query = "INSERT OR REPLACE INTO release_ (id, release_group_id, title, copyright, album_status, language, year, date, country, barcode) VALUES (?, ?, ?, ?, ?, ?, ?, ?, ?, ?);"
values = musicbrainz_albumid, release_group_id, title, copyright_, album_status, language, year, date, country, barcode
self.cursor.execute(query, values)
self.connection.commit()
def add_track(
self,
musicbrainz_releasetrackid: str,
musicbrainz_albumid: str,
feature_aritsts: list,
2022-11-06 22:01:03 +00:00
tracknumber: str = None,
2022-11-06 17:10:00 +00:00
track: str = None,
2022-11-29 13:16:36 +00:00
isrc: str = None,
length: int = None
2022-11-06 17:10:00 +00:00
):
# add adjacency
adjacency_list = []
for artist_id in feature_aritsts:
adjacency_list.append((artist_id, musicbrainz_releasetrackid))
adjacency_values = tuple(adjacency_list)
adjacency_query = "INSERT OR REPLACE INTO artist_track (artist_id, track_id) VALUES (?, ?);"
self.cursor.executemany(adjacency_query, adjacency_values)
self.connection.commit()
# add track
2022-11-29 13:16:36 +00:00
query = "INSERT OR REPLACE INTO track (id, release_id, track, isrc, tracknumber, length) VALUES (?, ?, ?, ?, ?, ?);"
values = musicbrainz_releasetrackid, musicbrainz_albumid, track, isrc, tracknumber, length
2022-11-06 17:10:00 +00:00
self.cursor.execute(query, values)
self.connection.commit()
@staticmethod
def get_custom_track_query(custom_where: list) -> str:
where_args = [
2022-11-18 07:41:49 +00:00
"1 = 1"
2022-11-06 17:10:00 +00:00
]
where_args.extend(custom_where)
where_arg = " AND ".join(where_args)
query = f"""
SELECT DISTINCT
json_object(
'artists', json_group_array(
(
SELECT DISTINCT json_object(
'id', artist.id,
'name', artist.name
)
)
),
'source', json_group_array(
(
2022-11-18 09:16:06 +00:00
SELECT DISTINCT json_object(
2022-11-21 11:25:40 +00:00
'src', src_table.src,
'url', src_table.url,
'valid', src_table.valid
)
)
),
2022-12-01 12:15:30 +00:00
'lyrics', json_group_array(
(
SELECT DISTINCT json_object(
'text', lyrics_table.text
'language', lyrics_table.language
)
)
),
'target', json_group_array(
(
SELECT DISTINCT json_object(
'file', target.file
'path', target.path
)
)
),
'id', track.id,
2022-12-01 12:15:30 +00:00
'mb_id', track.mb_id,
2022-11-06 22:01:03 +00:00
'tracknumber', track.tracknumber,
2022-11-21 11:25:40 +00:00
'titlesort', track.tracknumber,
'musicbrainz_releasetrackid', track.id,
'musicbrainz_albumid', release_.id,
'title', track.track,
'isrc', track.isrc,
'album', release_.title,
'copyright', release_.copyright,
'album_status', release_.album_status,
'language', release_.language,
'year', release_.year,
'date', release_.date,
'country', release_.country,
'barcode', release_.barcode,
'albumartist', release_group.albumartist,
'albumsort', release_group.albumsort,
'musicbrainz_albumtype', release_group.musicbrainz_albumtype,
'compilation', release_group.compilation,
'album_artist_id', release_group.album_artist_id,
2022-11-29 13:16:36 +00:00
'length', track.length,
'path', track.path,
'file', track.file,
'genre', track.genre,
'url', track.url,
2022-11-14 13:57:44 +00:00
'src', track.src,
'lyrics', track.lyrics
)
2022-11-21 11:25:40 +00:00
FROM track
LEFT JOIN release_ ON track.release_id = release_.id
LEFT JOIN release_group ON release_.id = release_group.id
LEFT JOIN artist_track ON track.id = artist_track.track_id
LEFT JOIN artist ON artist_track.artist_id = artist.id
LEFT JOIN source src_table ON track.id = src_table.track_id
2022-12-01 12:15:30 +00:00
LEFT JOIN lyrics lyrics_table ON track.id = lyrics_table.track_id
LEFT JOIN target ON track.id = target.track_id
2022-10-27 21:09:37 +00:00
WHERE
{where_arg}
GROUP BY track.id;
2022-11-06 17:10:00 +00:00
"""
return query
2022-12-01 12:15:30 +00:00
def get_custom_track(self, custom_where: list) -> List[Song]:
2022-11-06 17:10:00 +00:00
query = Database.get_custom_track_query(custom_where=custom_where)
return [get_song_from_response(json.loads(i[0])) for i in self.cursor.execute(query)]
2022-11-06 17:10:00 +00:00
def get_track_metadata(self, musicbrainz_releasetrackid: str):
# this would be vulnerable if musicbrainz_releasetrackid would be user input
resulting_tracks = self.get_custom_track([f'track.id == "{musicbrainz_releasetrackid}"'])
if len(resulting_tracks) != 1:
return -1
2022-11-06 17:10:00 +00:00
return resulting_tracks[0]
2022-10-27 21:09:37 +00:00
2022-12-01 12:15:30 +00:00
def get_tracks_to_download(self) -> List[Song]:
2022-11-06 17:10:00 +00:00
return self.get_custom_track(['track.downloaded == 0'])
2022-12-01 12:15:30 +00:00
def get_tracks_without_src(self) -> List[Song]:
2022-11-06 17:10:00 +00:00
return self.get_custom_track(["(track.url IS NULL OR track.src IS NULL)"])
2022-12-01 12:15:30 +00:00
def get_tracks_without_isrc(self) -> List[Song]:
2022-11-06 17:10:00 +00:00
return self.get_custom_track(["track.isrc IS NULL"])
2022-12-01 12:15:30 +00:00
def get_tracks_without_filepath(self) -> List[Song]:
2022-11-06 17:10:00 +00:00
return self.get_custom_track(["(track.file IS NULL OR track.path IS NULL OR track.genre IS NULL)"])
2022-12-01 12:15:30 +00:00
def get_tracks_for_lyrics(self) -> List[Song]:
2022-11-14 14:44:32 +00:00
return self.get_custom_track(["track.lyrics IS NULL"])
2022-12-01 12:15:30 +00:00
def add_lyrics(self, song: Song, lyrics: Lyrics):
2022-11-14 13:57:44 +00:00
query = f"""
UPDATE track
SET lyrics = ?
2022-11-25 17:27:48 +00:00
WHERE '{song.id}' == id;
2022-11-14 13:57:44 +00:00
"""
2022-11-25 17:27:48 +00:00
self.cursor.execute(query, (str(lyrics.text),))
2022-11-14 13:57:44 +00:00
self.connection.commit()
2022-11-06 17:10:00 +00:00
def update_download_status(self, track_id: str):
2022-11-09 21:52:32 +00:00
query = f"UPDATE track SET downloaded = 1, WHERE '{track_id}' == id;"
self.cursor.execute(query)
self.connection.commit()
2022-11-25 07:29:18 +00:00
def set_field_of_song(self, track_id: str, key: str, value: str):
query = f"UPDATE track SET {key} = ? WHERE '{track_id}' == id;"
2022-11-25 17:27:48 +00:00
self.cursor.execute(query, (value,))
2022-11-25 07:29:18 +00:00
self.connection.commit()
2022-11-06 17:10:00 +00:00
def set_download_data(self, track_id: str, url: str, src: str):
query = f"""
UPDATE track
SET url = ?,
src = ?
WHERE '{track_id}' == id;
2022-11-06 17:10:00 +00:00
"""
self.cursor.execute(query, (url, src))
self.connection.commit()
2022-11-25 17:27:48 +00:00
2022-11-21 11:25:40 +00:00
query = "INSERT OR REPLACE INTO source (track_id, src, url) VALUES (?, ?, ?);"
self.cursor.execute(query, (track_id, src, url))
self.connection.commit()
2022-11-06 17:10:00 +00:00
def set_filepath(self, track_id: str, file: str, path: str, genre: str):
query = f"""
UPDATE track
SET file = ?,
path = ?,
genre = ?
WHERE '{track_id}' == id;
2022-11-06 17:10:00 +00:00
"""
self.cursor.execute(query, (file, path, genre))
self.connection.commit()
2022-12-01 15:14:59 +00:00
def write_target(self, song_id: str, target: Target):
query = f"UPDATE track SET file = ?, path = ? WHERE '{song_id}' == id;"
self.cursor.execute(query, (target.file, target.path))
self.connection.commit()
2022-12-01 12:15:30 +00:00
2022-12-01 15:14:59 +00:00
def write_artist(self, artist: Artist, song_id: str = None, release_group_id: str = None):
artist_id = artist.id
2022-10-27 13:55:16 +00:00
2022-12-01 15:14:59 +00:00
query = "INSERT OR REPLACE INTO artist (id, mb_id, name) VALUES (?, ?, ?);"
self.cursor.execute(query, (artist_id, artist.mb_id, artist.name))
self.connection.commit()
2022-11-06 17:10:00 +00:00
2022-12-01 15:14:59 +00:00
if song_id is not None:
adjacency_query = "INSERT OR REPLACE INTO artist_track (artist_id, track_id) VALUES (?, ?);"
self.cursor.execute(adjacency_query, (artist_id, song_id))
self.connection.commit()
2022-11-06 17:10:00 +00:00
2022-12-01 15:14:59 +00:00
if release_group_id is not None:
adjacency_query = "INSERT OR REPLACE INTO artist_release_group (artist_id, release_group_id) VALUES (?, ?);"
self.cursor.execute(adjacency_query, (artist_id, release_group_id))
self.connection.commit()
2022-11-06 17:10:00 +00:00
2022-12-01 15:14:59 +00:00
def write_many_artists(self, song_id: str, artist_list: List[Artist]):
for artist in artist_list:
self.write_artist(song_id=song_id, artist=artist)
2022-11-06 17:10:00 +00:00
2022-12-01 15:43:25 +00:00
def write_source(self, song_id: str, source: Source):
pass
def write_many_sources(self, song_id: str, source_list: List[Source]):
for source in source_list:
self.write_source(song_id=song_id, source=source)
2022-12-01 15:14:59 +00:00
def write_song(self, song: Song):
song_id = song.id
# write artists
self.write_many_artists(song_id=song_id, artist_list=song.artists)
2022-12-01 15:43:25 +00:00
# write sources
self.write_many_sources(song_id=song_id, source_list=song.sources)
2022-12-01 15:14:59 +00:00
# write target
self.write_target(song_id=song_id, target=song.target)
2022-10-27 21:09:37 +00:00
2022-12-01 15:14:59 +00:00
def write_many_song(self, songs: List[Song]):
for song in songs:
self.write_song(song=song)