4af7600fd5
radoslav-orlov: "create lossy copies of lossless tracks" only recognized FLAC, even though ALAC/ WAV/AIFF/DSD are now quality-profile formats. the FLAC knowledge was hardcoded in 3 separate places (the import path, the Lossy Converter scan, and the fix executor) — exactly how a format gets added in one spot but not another. kettui-style fix — one canonical seam both sites route through, instead of 3 more string edits: - new core/quality/lossless.py: is_lossless_format / is_lossless_audio_path (pure; injects a codec probe for the ambiguous .m4a/.mp4 — ALAC vs AAC — so the decision stays testable with no I/O), LOSSLESS_FORMATS (single source of truth, derived-consistent with model.tier_score), and the lossy_output_would_overwrite_source safety invariant. - create_lossy_copy + the Lossy Converter scan + repair_worker._fix_missing_lossy_copy all route through it. SQL pre-filters by candidate extensions, then each file is confirmed (probing .m4a). - SAFETY: a lossy copy must never be written over its own source — an .m4a ALAC source + AAC target lands on the same .m4a path, and ffmpeg runs with -y. all three sites now bail on the overwrite case BEFORE ffmpeg (the existing delete-original guard was too late — the source was already clobbered). dropped a vestigial mutagen FLAC import; updated FLAC-only UI strings. 19 tests: full seam coverage (formats, the .m4a ALAC/AAC probe branch, candidate extensions, the overwrite guard), a tier-model consistency test that fails if the lossless set drifts, and import- site wiring tests — WAV now converts (was rejected), and the .m4a-ALAC+AAC overwrite case proves ffmpeg NEVER runs. 286 quality/import/repair tests green, ruff clean.
121 lines
5 KiB
Python
121 lines
5 KiB
Python
"""The canonical "is this lossless?" seam + the lossy-copy overwrite invariant
|
|
(#941). All pure — no files, no ffmpeg — so the decision and the safety guard are
|
|
unit-testable in isolation. Both the import path (create_lossy_copy) and the Lossy
|
|
Converter repair job route through these, so the same rules drive both."""
|
|
|
|
from core.quality.lossless import (
|
|
LOSSLESS_FORMATS,
|
|
LOSSLESS_CANDIDATE_EXTENSIONS,
|
|
is_lossless_format,
|
|
is_lossless_audio_path,
|
|
lossy_output_would_overwrite_source,
|
|
)
|
|
from core.quality.model import AudioQuality
|
|
from core.repair_jobs.lossy_converter import _lossless_ext_where
|
|
|
|
|
|
# ── is_lossless_format ──
|
|
|
|
def test_lossless_formats_recognized():
|
|
for fmt in ('flac', 'alac', 'wav', 'dsf', 'FLAC', 'Dsf'):
|
|
assert is_lossless_format(fmt) is True
|
|
|
|
|
|
def test_lossy_formats_not_lossless():
|
|
for fmt in ('mp3', 'aac', 'ogg', 'opus', 'wma', 'unknown', '', None):
|
|
assert is_lossless_format(fmt) is False
|
|
|
|
|
|
# ── is_lossless_audio_path (the ambiguity is the whole point) ──
|
|
|
|
def test_unambiguous_extensions_decided_by_extension():
|
|
for path in ('/m/a.flac', '/m/a.wav', '/m/a.wave', '/m/a.aiff', '/m/a.aif',
|
|
'/m/a.dsf', '/m/a.dff', '/m/a.alac', '/m/A.FLAC'):
|
|
assert is_lossless_audio_path(path) is True
|
|
|
|
|
|
def test_lossy_extensions_are_not_lossless():
|
|
for path in ('/m/a.mp3', '/m/a.ogg', '/m/a.opus', '/m/a.wma', '/m/a.aac'):
|
|
assert is_lossless_audio_path(path) is False
|
|
|
|
|
|
def test_m4a_without_probe_is_not_lossless():
|
|
# The safe default: with no codec probe, an .m4a can't be proven lossless, so
|
|
# an AAC file is never misclassified as lossless and converted/deleted.
|
|
assert is_lossless_audio_path('/m/a.m4a') is False
|
|
assert is_lossless_audio_path('/m/a.mp4') is False
|
|
|
|
|
|
def test_m4a_alac_is_lossless_via_probe():
|
|
assert is_lossless_audio_path('/m/a.m4a', probe_codec=lambda _p: 'alac') is True
|
|
assert is_lossless_audio_path('/m/a.mp4', probe_codec=lambda _p: 'ALAC') is True
|
|
|
|
|
|
def test_m4a_aac_is_not_lossless_via_probe():
|
|
assert is_lossless_audio_path('/m/a.m4a', probe_codec=lambda _p: 'mp4a.40.2') is False
|
|
|
|
|
|
def test_probe_exception_is_not_lossless():
|
|
def _boom(_p):
|
|
raise RuntimeError("probe failed")
|
|
assert is_lossless_audio_path('/m/a.m4a', probe_codec=_boom) is False
|
|
|
|
|
|
# ── LOSSLESS_CANDIDATE_EXTENSIONS (the SQL pre-filter set) ──
|
|
|
|
def test_candidate_extensions_cover_lossless_plus_ambiguous():
|
|
for e in ('.flac', '.wav', '.aiff', '.dsf', '.dff', '.alac', '.m4a', '.mp4'):
|
|
assert e in LOSSLESS_CANDIDATE_EXTENSIONS
|
|
# raw lossy extensions must NOT be candidates
|
|
for e in ('.mp3', '.aac', '.ogg', '.opus', '.wma'):
|
|
assert e not in LOSSLESS_CANDIDATE_EXTENSIONS
|
|
|
|
|
|
def test_sql_where_clause_matches_candidates_only():
|
|
where = _lossless_ext_where('t.file_path')
|
|
assert "LIKE '%.flac'" in where and "LIKE '%.dsf'" in where and "LIKE '%.m4a'" in where
|
|
assert "LIKE '%.mp3'" not in where and "LIKE '%.aac'" not in where
|
|
|
|
|
|
# ── lossy_output_would_overwrite_source (the safety invariant) ──
|
|
|
|
def test_overwrite_detected_when_paths_equal():
|
|
assert lossy_output_would_overwrite_source('/m/Album/01.m4a', '/m/Album/01.m4a') is True
|
|
|
|
|
|
def test_overwrite_detected_after_normalization():
|
|
assert lossy_output_would_overwrite_source('/m/Album/../Album/01.m4a', '/m/Album/01.m4a') is True
|
|
|
|
|
|
def test_no_overwrite_for_different_extension():
|
|
assert lossy_output_would_overwrite_source('/m/Album/01.flac', '/m/Album/01.mp3') is False
|
|
assert lossy_output_would_overwrite_source('/m/Album/01.m4a', '/m/Album/01.mp3') is False
|
|
|
|
|
|
def test_overwrite_guard_handles_empty():
|
|
assert lossy_output_would_overwrite_source('', '/x.mp3') is False
|
|
assert lossy_output_would_overwrite_source('/x.flac', '') is False
|
|
|
|
|
|
# ── anti-drift: the seam must agree with the quality tier model ──
|
|
|
|
def test_lossless_set_consistent_with_tier_model():
|
|
"""If a format is in LOSSLESS_FORMATS it must out-rank every lossy format in
|
|
tier_score — guards against the two lists drifting apart (the whole reason
|
|
this seam exists)."""
|
|
lossy = ('mp3', 'aac', 'ogg', 'opus', 'wma')
|
|
worst_lossless = min(AudioQuality(f, bitrate=11290, sample_rate=44100, bit_depth=16).tier_score()
|
|
for f in LOSSLESS_FORMATS)
|
|
best_lossy = max(AudioQuality(f, bitrate=320).tier_score() for f in lossy)
|
|
assert worst_lossless > best_lossy
|
|
|
|
|
|
# ── regression: the exact bug class this guards (overwrite the original) ──
|
|
|
|
def test_regression_m4a_alac_to_aac_would_overwrite_and_is_blocked():
|
|
"""An .m4a ALAC source converted with the AAC codec lands on the SAME .m4a
|
|
path. The guard must catch it so ffmpeg -y never destroys the original."""
|
|
src = '/library/Sade/Diamond Life/01. Smooth Operator.m4a' # ALAC
|
|
out = src # AAC target → .m4a
|
|
assert is_lossless_audio_path(src, probe_codec=lambda _p: 'alac') is True
|
|
assert lossy_output_would_overwrite_source(src, out) is True # → callers skip
|