Pierre-Olivier Mercier
7e0f08b863
Give tracks a source-agnostic identity so the same song from different sources no longer replays in a loop. - Canonicalizer resolves (artist, title) to a MusicBrainz recording MBID (no API key; ~1 req/s, descriptive User-Agent, best-effort). Hits and confirmed misses are cached in SQLite; transient errors are not. - Track.key becomes mbid:<id> when resolved, else a normalized name:<artist>|<title> fallback — still source-agnostic. - Scheduler now owns the authoritative anti-repeat on the canonical key, canonicalizing the drawn track with a bounded retry; providers keep a cheap recent-locator filter to limit retries. - db: canonical_cache table, history.locator column with migration for existing databases, recent_locators(). - Canonicalization can be turned off via RADIEO_CANONICAL_ENABLED=0. Verified: MBID hit/cache/miss, cross-source key collapse, scheduler dodging a recent play, schema migration, and full stack (Navidrome + yt-dlp) with zero Python tracebacks and a valid 192 kbps MP3 stream. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
47 lines
1.7 KiB
Python
47 lines
1.7 KiB
Python
"""Shared data model.
|
|
|
|
A ``Track`` is the uniform object every provider emits: a *resolved* reference
|
|
(which backend can download it, and where) plus display metadata. Fetchers turn
|
|
it into a local file; the queue and the state database use ``key`` for
|
|
de-duplication and anti-repeat.
|
|
"""
|
|
|
|
import re
|
|
from dataclasses import dataclass
|
|
|
|
_WS = re.compile(r"\s+")
|
|
|
|
|
|
def norm_name(value: str) -> str:
|
|
"""Normalize an artist/title for stable keying and cache lookups:
|
|
case-fold and collapse whitespace. Kept deliberately light (no accent
|
|
stripping) so distinct titles never collapse together."""
|
|
return _WS.sub(" ", value.strip()).casefold()
|
|
|
|
|
|
@dataclass(frozen=True)
|
|
class Track:
|
|
backend: str # which fetcher handles it: "subsonic" | "ytdlp"
|
|
locator: str # backend-specific: Subsonic song id, or a media URL
|
|
artist: str
|
|
title: str
|
|
origin: str # provider that produced it, e.g. "navidrome"
|
|
mbid: str | None = None # filled by the Canonicalizer (milestone 5)
|
|
source_ext: str | None = None # filename hint, e.g. "mp3", "flac"
|
|
source_url: str | None = None # container URL a track was picked from
|
|
|
|
@property
|
|
def key(self) -> str:
|
|
"""Stable, source-agnostic identity for de-duplication and anti-repeat.
|
|
|
|
The Canonicalizer fills ``mbid`` when it can, giving a truly
|
|
cross-source identity. When it can't, we fall back to the normalized
|
|
``(artist, title)`` — still source-agnostic, so the same track fetched
|
|
from two backends collapses to one key.
|
|
"""
|
|
if self.mbid:
|
|
return f"mbid:{self.mbid}"
|
|
return f"name:{norm_name(self.artist)}|{norm_name(self.title)}"
|
|
|
|
def __str__(self) -> str:
|
|
return f"{self.artist} — {self.title} [{self.origin}]"
|