restructured slightly and addet skript to upload to PyPI

2022-11-15 00:27:03 +01:00
parent 5104fe3d1d
commit c6d0acff9b
36 changed files with 291 additions and 2 deletions
--- a/src/music_kraken/lyrics/init.py
+++ b/src/music_kraken/lyrics/init.py
--- a/src/music_kraken/lyrics/genius.py
+++ b/src/music_kraken/lyrics/genius.py
@@ -0,0 +1,171 @@
+import requests
+from typing import List
+from bs4 import BeautifulSoup
+import pycountry
+
+from ..utils.shared import *
+from ..utils import phonetic_compares
+from ..utils.object_handeling import get_elem_from_obj
+
+# search doesn't support isrc
+# https://genius.com/api/search/multi?q=I Prevail - Breaking Down
+# https://genius.com/api/songs/6192944
+# https://docs.genius.com/
+
+session = requests.Session()
+session.headers = {
+    "Connection": "keep-alive",
+    "Referer": "https://genius.com/search/embed"
+}
+session.proxies = proxies
+
+logger = GENIUS_LOGGER
+
+
+class Song:
+    def __init__(self, raw_data: dict, desirered_data: dict):
+        self.raw_data = raw_data
+        self.desired_data = desirered_data
+
+        song_data = get_elem_from_obj(self.raw_data, ['result'], return_if_none={})
+        self.id = get_elem_from_obj(song_data, ['id'])
+        self.artist = get_elem_from_obj(song_data, ['primary_artist', 'name'])
+        self.title = get_elem_from_obj(song_data, ['title'])
+
+        lang_code = get_elem_from_obj(song_data, ['language']) or "en"
+        self.language = pycountry.languages.get(alpha_2=lang_code)
+        self.lang = self.language.alpha_3
+        self.url = get_elem_from_obj(song_data, ['url'])
+
+        # maybe could be implemented
+        self.lyricist: str
+
+        if get_elem_from_obj(song_data, ['lyrics_state']) != "complete":
+            logger.warning(
+                f"lyrics state of {self.title} by {self.artist} is not complete but {get_elem_from_obj(song_data, ['lyrics_state'])}")
+
+        self.valid = self.is_valid()
+        if self.valid:
+            logger.info(f"found lyrics for \"{self.__repr__()}\"")
+        else:
+            return
+
+        self.lyrics = self.fetch_lyrics()
+        if self.lyrics is None:
+            self.valid = False
+
+    def is_valid(self) -> bool:
+        title_match, title_distance = phonetic_compares.match_titles(self.title, self.desired_data['track'])
+        artist_match, artist_distance = phonetic_compares.match_artists(self.desired_data['artist'], self.artist)
+
+        return not title_match and not artist_match
+
+    def __repr__(self) -> str:
+        return f"{self.title} by {self.artist} ({self.url})"
+
+    def fetch_lyrics(self) -> str | None:
+        if not self.valid:
+            logger.warning(f"{self.__repr__()} is invalid but the lyrics still get fetched. Something could be wrong.")
+
+        r = session.get(self.url)
+        if r.status_code != 200:
+            logging.warning(f"{r.url} returned {r.status_code}:\n{r.content}")
+            return None
+
+        soup = BeautifulSoup(r.content, "html.parser")
+        lyrics_soups = soup.find_all('div', {'data-lyrics-container': "true"})
+        if len(lyrics_soups) == 0:
+            logger.warning(f"didn't found lyrics on {self.url}")
+            return None
+        if len(lyrics_soups) != 1:
+            logger.warning(f"number of lyrics_soups doesn't equals 1, but {len(lyrics_soups)} on {self.url}")
+
+        lyrics = "\n".join([lyrics_soup.getText(separator="\n", strip=True) for lyrics_soup in lyrics_soups])
+        print(lyrics)
+
+        # <div data-lyrics-container="true" class="Lyrics__Container-sc-1ynbvzw-6 YYrds">With the soundle
+        self.lyrics = lyrics
+        return lyrics
+
+
+def process_multiple_songs(song_datas: list, desired_data: dict) -> List[Song]:
+    all_songs = [Song(song_data, desired_data) for song_data in song_datas]
+    return all_songs
+
+
+def search_song_list(artist: str, track: str) -> List[Song]:
+    endpoint = "https://genius.com/api/search/multi?q="
+    url = f"{endpoint}{artist} - {track}"
+    logging.info(f"requesting {url}")
+
+    desired_data = {
+        'artist': artist,
+        'track': track
+    }
+
+    r = session.get(url)
+    if r.status_code != 200:
+        logging.warning(f"{r.url} returned {r.status_code}:\n{r.content}")
+        return []
+    content = r.json()
+    if get_elem_from_obj(content, ['meta', 'status']) != 200:
+        logging.warning(f"{r.url} returned {get_elem_from_obj(content, ['meta', 'status'])}:\n{content}")
+        return []
+
+    sections = get_elem_from_obj(content, ['response', 'sections'])
+    for section in sections:
+        section_type = get_elem_from_obj(section, ['type'])
+        if section_type == "song":
+            return process_multiple_songs(get_elem_from_obj(section, ['hits'], return_if_none=[]), desired_data)
+
+    return []
+
+
+def search(artist: str, track: str) -> list:
+    results = []
+    r = search_song_list(artist, track)
+    for r_ in r:
+        if r_.valid:
+            results.append(r_)
+    return results
+
+
+if __name__ == "__main__":
+    logging.basicConfig(level=logging.DEBUG)
+
+    """
+    song = Song(
+        {'highlights': [], 'index': 'song', 'type': 'song',
+         'result': {'_type': 'song', 'annotation_count': 0, 'api_path': '/songs/6142483',
+                    'artist_names': 'Psychonaut 4',
+                    'full_title': 'Sana Sana Sana, Cura Cura Cura by\xa0Psychonaut\xa04',
+                    'header_image_thumbnail_url': 'https://images.genius.com/f9f67a3f9c801f697fbaf68c7efd3599.300x300x1.jpg',
+                    'header_image_url': 'https://images.genius.com/f9f67a3f9c801f697fbaf68c7efd3599.651x651x1.jpg',
+                    'id': 6142483, 'instrumental': False, 'language': 'en', 'lyrics_owner_id': 4443216,
+                    'lyrics_state': 'complete', 'lyrics_updated_at': 1604698709,
+                    'path': '/Psychonaut-4-sana-sana-sana-cura-cura-cura-lyrics', 'pyongs_count': None,
+                    'relationships_index_url': 'https://genius.com/Psychonaut-4-sana-sana-sana-cura-cura-cura-sample',
+                    'release_date_components': {'year': 2020, 'month': 7, 'day': 1},
+                    'release_date_for_display': 'July 1, 2020',
+                    'release_date_with_abbreviated_month_for_display': 'Jul. 1, 2020',
+                    'song_art_image_thumbnail_url': 'https://images.genius.com/f9f67a3f9c801f697fbaf68c7efd3599.300x300x1.jpg',
+                    'song_art_image_url': 'https://images.genius.com/f9f67a3f9c801f697fbaf68c7efd3599.651x651x1.jpg',
+                    'stats': {'unreviewed_annotations': 0, 'hot': False}, 'title': 'Sana Sana Sana, Cura Cura Cura',
+                    'title_with_featured': 'Sana Sana Sana, Cura Cura Cura', 'updated_by_human_at': 1647353214,
+                    'url': 'https://genius.com/Psychonaut-4-sana-sana-sana-cura-cura-cura-lyrics',
+                    'featured_artists': [], 'primary_artist': {'_type': 'artist', 'api_path': '/artists/1108956',
+                                                               'header_image_url': 'https://images.genius.com/ff13efc74a043237cfca3fc0a6cb12dd.1000x563x1.jpg',
+                                                               'id': 1108956,
+                                                               'image_url': 'https://images.genius.com/25ff7cfdcb6d92a9f19ebe394a895736.640x640x1.jpg',
+                                                               'index_character': 'p', 'is_meme_verified': False,
+                                                               'is_verified': False, 'name': 'Psychonaut 4',
+                                                               'slug': 'Psychonaut-4',
+                                                               'url': 'https://genius.com/artists/Psychonaut-4'}}},
+        {'artist': 'Psychonaut 4', 'track': 'Sana Sana Sana, Cura Cura Cura'}
+    )
+    print(song.fetch_lyrics())
+    """
+
+    songs = search("Zombiez", "WALL OF Z")
+    for song in songs:
+        print(song)
--- a/src/music_kraken/lyrics/lyrics.py
+++ b/src/music_kraken/lyrics/lyrics.py
@@ -0,0 +1,94 @@
+import mutagen
+from mutagen.id3 import ID3, USLT
+
+
+from ..metadata import database as db
+from ..utils.shared import *
+from . import genius
+
+logger = LYRICS_LOGGER
+
+"""
+This whole Part is bodgy as hell and I need to rewrite this little file urgently. genius.py is really clean though :3
+Just wanted to get it to work.
+ - lyrics need to be put in the database and everything should continue from there then
+"""
+
+"""
+https://cweiske.de/tagebuch/rhythmbox-lyrics.htm
+Rythmbox, my music player doesn't support ID3 lyrics (USLT) yet, so I have to find something else
+Lyrics in MP3 ID3 tags (SYLT/USLT) is still missing, because GStreamer does not support that yet.
+
+One possible sollution would be to use ogg/vorbis files. Those lyrics are supported in rythmbox
+'So, the next Rhythmbox release (3.5.0 or 3.4.2) will read lyrics directly from ogg/vorbis files, using the LYRICS and SYNCLYRICS tags.'
+Another possible sollution (probaply the better one cuz I dont need to refactor whole metadata AGAIN)
+would be to write a Rhythmbox plugin that fetches lyrics from ID3 USLT
+
+I have written that Rhythmbox plugin: https://github.com/HeIIow2/rythmbox-id3-lyrics-support
+"""
+
+
+# https://www.programcreek.com/python/example/63462/mutagen.mp3.EasyMP3
+# https://code.activestate.com/recipes/577138-embed-lyrics-into-mp3-files-using-mutagen-uslt-tag/
+
+
+def add_lyrics(file_name, lyrics):
+    file_path = os.path.join(MUSIC_DIR, file_name)
+    if not os.path.exists(file_path):
+        return
+
+    try:
+        tags = ID3(file_path)
+    except mutagen.id3.ID3NoHeaderError:
+        return
+
+    logger.info(f"adding lyrics to the file {file_path}")
+
+    uslt_output = USLT(encoding=3, lang=lyrics.lang, desc=u'desc', text=lyrics.lyrics)
+    tags["USLT::'eng'"] = uslt_output
+    tags.save(file_path)
+
+
+def fetch_single_lyrics(row: dict):
+    artists = [artist['name'] for artist in row['artists']]
+    track = row['title']
+    id_ = row['id']
+
+    logger.info(f"try fetching lyrics for \"{track}\" by \"{', '.join(artists)}")
+
+    lyrics = []
+    for artist in artists:
+        lyrics.extend(genius.search(artist, track))
+    if len(lyrics) == 0:
+        return
+
+    logger.info("found lyrics")
+    database.add_lyrics(id_, lyrics=lyrics[0])
+    add_lyrics(row['file'], lyrics[0])
+
+
+def fetch_lyrics():
+    for row in database.get_tracks_for_lyrics():
+        fetch_single_lyrics(row)
+
+
+if __name__ == "__main__":
+    import tempfile
+    import os
+
+    temp_folder = "music-downloader"
+    temp_dir = os.path.join(tempfile.gettempdir(), temp_folder)
+    if not os.path.exists(temp_dir):
+        os.mkdir(temp_dir)
+
+    logging.basicConfig(level=logging.DEBUG)
+    db_logger = logging.getLogger("database")
+    db_logger.setLevel(logging.DEBUG)
+
+    database = db.Database(os.path.join(temp_dir, "metadata.db"),
+                           os.path.join(temp_dir, "database_structure.sql"),
+                           "https://raw.githubusercontent.com/HeIIow2/music-downloader/new_metadata/assets/database_structure.sql",
+                           db_logger,
+                           reset_anyways=False)
+
+    fetch_lyrics()