mirror of
https://github.com/Horizontal42/Sort_YM.git
synced 2026-11-10 15:42:02 +03:00
942c2d19e4
Медленная/нестабильная сеть до api.music.yandex.net роняла весь процесс при первом же таймауте (5с по умолчанию у библиотеки). Добавлен with_retries с экспоненциальной паузой на всех сетевых точках входа (fetch, language, genres, apply, client.init), таймаут запроса поднят до 20с (настраивается в config.toml). Заодно поправлен скрытый баг: language.py раньше кэшировал таймаут как "текста нет" навсегда - теперь только настоящий NotFoundError трактуется так.
81 lines
2.4 KiB
Python
81 lines
2.4 KiB
Python
from __future__ import annotations
|
||
|
||
import json
|
||
import time
|
||
from pathlib import Path
|
||
|
||
from yandex_music import Client, Track
|
||
|
||
from .ymclient import chunked, with_retries
|
||
|
||
TRACKS_CACHE_FILE = "tracks.json"
|
||
|
||
|
||
def _track_genre(track: Track) -> str | None:
|
||
for album in track.albums:
|
||
if album.genre:
|
||
return album.genre
|
||
return None
|
||
|
||
|
||
def _serialize(track: Track) -> dict:
|
||
album_id = track.albums[0].id if track.albums else None
|
||
return {
|
||
"id": track.id,
|
||
"album_id": album_id,
|
||
"title": track.title or "",
|
||
"artists": [a.name for a in track.artists if a.name],
|
||
"genre_raw": _track_genre(track),
|
||
"lyrics_available": bool(track.lyrics_available),
|
||
}
|
||
|
||
|
||
def _atomic_write_json(path: Path, data: dict) -> None:
|
||
tmp = path.with_suffix(".tmp")
|
||
tmp.write_text(json.dumps(data, ensure_ascii=False, indent=2), encoding="utf-8")
|
||
tmp.replace(path)
|
||
|
||
|
||
def load_tracks_cache(cache_dir: Path) -> dict[str, dict]:
|
||
cache_file = cache_dir / TRACKS_CACHE_FILE
|
||
if cache_file.exists():
|
||
return json.loads(cache_file.read_text(encoding="utf-8"))
|
||
return {}
|
||
|
||
|
||
def fetch_liked_tracks(
|
||
client: Client,
|
||
cache_dir: Path,
|
||
batch_size: int,
|
||
batch_delay: float,
|
||
) -> dict[str, dict]:
|
||
cache_dir.mkdir(parents=True, exist_ok=True)
|
||
cache_file = cache_dir / TRACKS_CACHE_FILE
|
||
cache = load_tracks_cache(cache_dir)
|
||
|
||
likes = with_retries(lambda: client.users_likes_tracks())
|
||
if likes is None:
|
||
print("Не удалось получить список лайкнутых треков.")
|
||
return cache
|
||
|
||
all_ids = likes.tracks_ids
|
||
missing_ids = [tid for tid in all_ids if tid not in cache]
|
||
|
||
print(f"Лайкнутых треков: {len(all_ids)}, новых для загрузки: {len(missing_ids)}")
|
||
|
||
for batch in chunked(missing_ids, batch_size):
|
||
tracks = with_retries(lambda: client.tracks(batch))
|
||
for track in tracks:
|
||
cache[track.track_id] = _serialize(track)
|
||
_atomic_write_json(cache_file, cache)
|
||
print(f" загружено {len(cache)}/{len(all_ids)}")
|
||
time.sleep(batch_delay)
|
||
|
||
stale = set(cache) - set(all_ids)
|
||
for tid in stale:
|
||
del cache[tid]
|
||
if stale:
|
||
_atomic_write_json(cache_file, cache)
|
||
print(f" убрано из кэша (больше не лайкнуто): {len(stale)}")
|
||
|
||
return cache
|