izebra_projects/musicfetch
1
0
Fork 0
Code Issues Pull Requests Activity

fix: harden --repair against music videos; first-artist folder for single URLs

Browse Source 
--repair was clobbering good tags and erroring on real libraries:
- Validate the parsed id per source (YouTube 11-char, SoundCloud numeric) so
  junk ids from bracketed descriptors ([Official Video]) are skipped, not queried.
- Skip files whose source returns no real music metadata (no album/year, e.g.
  music videos) instead of overwriting clean tags with channel/decorated titles.
- Year from release info only (sane 1000-2100), never upload_date (which gave
  wrong years for old songs and bogus values like 6577).
- album/year are authoritative; artist/title are fill-missing-only (no clobber).

Also: download_single now uses the first artist for the folder (matching the
search/playlist paths) so single-URL downloads stop creating multi-artist dirs.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
This commit is contained in:
Zeb Hering
2026-06-10 18:53:27 -07:00
parent 1a81f64cc3
commit c6e28a4f75
3 changed files with 140 additions and 58 deletions
Download Patch File Download Diff File Expand all files Collapse all files

117
tests/test_repair.py
View File

@@ -1,13 +1,15 @@
import server.mf # noqa: F401 — loads musicfetch, registers musicfetch_core
import musicfetch_core as mf
YT_ID = "dQw4w9WgXcQ" # valid 11-char YouTube id
# ---- _is_source_dir ----
def test_is_source_dir():
assert mf._is_source_dir("youtube") is True
assert mf._is_source_dir("soundcloud") is True
assert mf._is_source_dir("downloads") is True
assert mf._is_source_dir("Discovery") is False # Lidarr album folder
assert mf._is_source_dir("Discovery") is False # Lidarr album folder
assert mf._is_source_dir("Random Access Memories") is False
assert mf._is_source_dir("") is False
@@ -16,38 +18,39 @@ def test_is_source_dir():
def test_parse_track_file():
assert mf._parse_track_file("Under My Skin [nGSNF2l44Zc].opus") == ("Under My Skin", "nGSNF2l44Zc")
assert mf._parse_track_file("Ignomon [2202690443].m4a") == ("Ignomon", "2202690443")
# greedy title: real id is the LAST bracket
assert mf._parse_track_file("WHO GON' SLIDE [Official Music Video] [AxjP9s6J3uY].opus") \
== ("WHO GON' SLIDE [Official Music Video]", "AxjP9s6J3uY")
assert mf._parse_track_file("no-id-here.opus") is None
assert mf._parse_track_file("cover.jpg") is None
# ---- _repair_id_ok ----
def test_repair_id_ok():
assert mf._repair_id_ok("youtube", YT_ID) is True
assert mf._repair_id_ok("youtube", "Official Video") is False # space, wrong length
assert mf._repair_id_ok("youtube", "Cover") is False
assert mf._repair_id_ok("soundcloud", "2202690443") is True
assert mf._repair_id_ok("soundcloud", "abc") is False
assert mf._repair_id_ok("bandcamp", "x") is False
# ---- _valid_year ----
def test_valid_year():
assert mf._valid_year({"release_year": 2001}) == "2001"
assert mf._valid_year({"release_date": "1976-09-10"}) == "1976"
assert mf._valid_year({"upload_date": "20110101"}) == "" # upload date ignored
assert mf._valid_year({"release_year": 6577}) == "" # out of range
assert mf._valid_year({}) == ""
# ---- _repair_probe_url ----
def test_repair_probe_url():
assert mf._repair_probe_url("youtube", "vid") == "https://music.youtube.com/watch?v=vid"
assert mf._repair_probe_url("youtube", YT_ID) == f"https://music.youtube.com/watch?v={YT_ID}"
assert mf._repair_probe_url("soundcloud", "123") == "https://api.soundcloud.com/tracks/123"
assert mf._repair_probe_url("bandcamp", "x") is None
# ---- _desired_tags ----
def test_desired_tags_full():
meta = {"artist": "Daft Punk", "title": "Aerodynamic", "album": "Discovery", "release_year": 2001}
assert mf._desired_tags(meta) == {"artist": "Daft Punk", "title": "Aerodynamic",
"album": "Discovery", "date": "2001"}
def test_desired_tags_year_fallbacks_and_omits_empty():
meta = {"title": "T", "uploader": "Chan", "upload_date": "20230102"} # no album, no release_year
out = mf._desired_tags(meta)
assert out["date"] == "2023"
assert out["title"] == "T"
assert out["artist"] == "Chan"
assert "album" not in out # omitted when empty
def test_desired_tags_drops_unknown_artist():
meta = {"title": "T"} # get_artist_from_metadata -> "Unknown Artist"
assert "artist" not in mf._desired_tags(meta)
# ---- repair_file (fake audio + mocked metadata) ----
class _FakeAudio(dict):
def __init__(self, initial):
@@ -58,13 +61,13 @@ class _FakeAudio(dict):
self.saved = True
def test_repair_file_writes_changed_fields(monkeypatch):
def test_repair_file_fixes_album_and_year(monkeypatch):
monkeypatch.setattr(mf, "run_yt_dlp_get_metadata",
lambda url: {"artist": "Daft Punk", "title": "Aerodynamic",
"album": "Discovery", "release_year": 2001})
audio = _FakeAudio({"artist": ["Daft Punk"], "title": ["Aerodynamic"]}) # album/date missing
monkeypatch.setattr(mf, "_open_audio", lambda path: (audio, None))
changed = mf.repair_file("X/youtube/Aerodynamic [vid].opus", "youtube", dry_run=False)
changed = mf.repair_file(f"X/youtube/Aerodynamic [{YT_ID}].opus", "youtube", dry_run=False)
assert set(changed) == {"album=Discovery", "date=2001"}
assert audio["album"] == ["Discovery"]
assert audio["date"] == ["2001"]
@@ -76,30 +79,69 @@ def test_repair_file_dry_run_writes_nothing(monkeypatch):
lambda url: {"artist": "A", "title": "T", "album": "Alb", "release_year": 2020})
audio = _FakeAudio({})
monkeypatch.setattr(mf, "_open_audio", lambda path: (audio, None))
changed = mf.repair_file("X/youtube/T [vid].opus", "youtube", dry_run=True)
assert changed # reports would-change
assert audio == {} # nothing written
changed = mf.repair_file(f"X/youtube/T [{YT_ID}].opus", "youtube", dry_run=True)
assert changed
assert audio == {}
assert audio.saved is False
def test_repair_file_skips_music_video(monkeypatch):
# No album AND no valid release year -> treat as a video, leave tags alone.
monkeypatch.setattr(mf, "run_yt_dlp_get_metadata",
lambda url: {"title": "Artist - Song (Official Music Video)",
"uploader": "SomeVEVO", "upload_date": "20110101"})
audio = _FakeAudio({"artist": ["Real Artist"], "title": ["Song"]})
monkeypatch.setattr(mf, "_open_audio", lambda path: (audio, None))
changed = mf.repair_file(f"X/youtube/Song [{YT_ID}].opus", "youtube", dry_run=False)
assert changed == []
assert audio == {"artist": ["Real Artist"], "title": ["Song"]} # untouched
def test_repair_file_fills_missing_but_never_clobbers(monkeypatch):
# Source artist is a channel name; existing artist must be kept.
monkeypatch.setattr(mf, "run_yt_dlp_get_metadata",
lambda url: {"artist": "SomeChannelVEVO", "title": "Channel Decorated Title",
"album": "Real Album", "release_year": 2019})
audio = _FakeAudio({"artist": ["Correct Artist"], "title": ["Clean Title"]})
monkeypatch.setattr(mf, "_open_audio", lambda path: (audio, None))
changed = mf.repair_file(f"X/youtube/x [{YT_ID}].opus", "youtube", dry_run=False)
assert set(changed) == {"album=Real Album", "date=2019"}
assert audio["artist"] == ["Correct Artist"] # NOT overwritten with channel
assert audio["title"] == ["Clean Title"] # NOT overwritten with decorated title
def test_repair_file_fills_missing_artist_when_absent(monkeypatch):
monkeypatch.setattr(mf, "run_yt_dlp_get_metadata",
lambda url: {"artist": "Real Artist", "title": "T",
"album": "Alb", "release_year": 2020})
audio = _FakeAudio({}) # nothing present -> fill artist + title too
monkeypatch.setattr(mf, "_open_audio", lambda path: (audio, None))
changed = mf.repair_file(f"X/youtube/T [{YT_ID}].opus", "youtube", dry_run=False)
assert set(changed) == {"album=Alb", "date=2020", "artist=Real Artist", "title=T"}
def test_repair_file_skips_bad_id(monkeypatch):
called = {"meta": False}
monkeypatch.setattr(mf, "run_yt_dlp_get_metadata",
lambda url: called.update(meta=True) or {})
# last bracket is a descriptor, not a real id
assert mf.repair_file("X/youtube/Song [Official Video].opus", "youtube", dry_run=False) == []
assert called["meta"] is False # never hit the network
def test_repair_file_skips_unparseable(monkeypatch):
called = {"meta": False}
monkeypatch.setattr(mf, "run_yt_dlp_get_metadata",
lambda url: called.update(meta=True) or {})
assert mf.repair_file("X/youtube/no-id.opus", "youtube", dry_run=False) == []
assert called["meta"] is False # never hit the network
def test_repair_file_skips_unqueryable_source(monkeypatch):
monkeypatch.setattr(mf, "run_yt_dlp_get_metadata", lambda url: {"title": "x"})
assert mf.repair_file("X/bandcamp/T [id].m4a", "bandcamp", dry_run=False) == []
assert called["meta"] is False
# ---- repair_library (real temp tree, repair_file mocked) ----
def test_repair_library_scans_only_source_dirs(tmp_path, monkeypatch):
root = tmp_path
(root / "Daft Punk" / "youtube").mkdir(parents=True)
(root / "Daft Punk" / "youtube" / "Aerodynamic [vid].opus").write_text("x")
(root / "Daft Punk" / "youtube" / f"Aerodynamic [{YT_ID}].opus").write_text("x")
(root / "Daft Punk" / "Discovery").mkdir(parents=True) # Lidarr album -> skip
(root / "Daft Punk" / "Discovery" / "Aerodynamic.flac").write_text("x")
(root / "Ephixa" / "soundcloud").mkdir(parents=True)
@@ -110,9 +152,8 @@ def test_repair_library_scans_only_source_dirs(tmp_path, monkeypatch):
lambda path, source, dry_run: visited.append((source, path)) or ["album=X"])
scanned, changed = mf.repair_library(str(root), dry_run=False)
assert scanned == 2 and changed == 2
sources = sorted(s for s, _ in visited)
assert sources == ["soundcloud", "youtube"] # Discovery album folder skipped
assert sorted(s for s, _ in visited) == ["soundcloud", "youtube"] # album folder skipped
def test_repair_library_missing_root(monkeypatch):
def test_repair_library_missing_root():
assert mf.repair_library("/no/such/dir", dry_run=False) == (0, 0)