flacfetch 0.37.0__tar.gz → 0.37.2__tar.gz
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- {flacfetch-0.37.0/flacfetch.egg-info → flacfetch-0.37.2}/PKG-INFO +1 -1
- {flacfetch-0.37.0 → flacfetch-0.37.2}/flacfetch/core/matching.py +12 -2
- {flacfetch-0.37.0 → flacfetch-0.37.2}/flacfetch/downloaders/spotify.py +39 -4
- {flacfetch-0.37.0 → flacfetch-0.37.2}/flacfetch/providers/gazelle.py +99 -43
- {flacfetch-0.37.0 → flacfetch-0.37.2/flacfetch.egg-info}/PKG-INFO +1 -1
- {flacfetch-0.37.0 → flacfetch-0.37.2}/pyproject.toml +1 -1
- {flacfetch-0.37.0 → flacfetch-0.37.2}/tests/test_gazelle.py +56 -0
- {flacfetch-0.37.0 → flacfetch-0.37.2}/tests/test_matching.py +12 -0
- {flacfetch-0.37.0 → flacfetch-0.37.2}/tests/test_spotify.py +83 -0
- {flacfetch-0.37.0 → flacfetch-0.37.2}/ARCHITECTURE.md +0 -0
- {flacfetch-0.37.0 → flacfetch-0.37.2}/LICENSE +0 -0
- {flacfetch-0.37.0 → flacfetch-0.37.2}/MANIFEST.in +0 -0
- {flacfetch-0.37.0 → flacfetch-0.37.2}/PLAN.md +0 -0
- {flacfetch-0.37.0 → flacfetch-0.37.2}/README.md +0 -0
- {flacfetch-0.37.0 → flacfetch-0.37.2}/flacfetch/__init__.py +0 -0
- {flacfetch-0.37.0 → flacfetch-0.37.2}/flacfetch/api/__init__.py +0 -0
- {flacfetch-0.37.0 → flacfetch-0.37.2}/flacfetch/api/auth.py +0 -0
- {flacfetch-0.37.0 → flacfetch-0.37.2}/flacfetch/api/main.py +0 -0
- {flacfetch-0.37.0 → flacfetch-0.37.2}/flacfetch/api/models.py +0 -0
- {flacfetch-0.37.0 → flacfetch-0.37.2}/flacfetch/api/routes/__init__.py +0 -0
- {flacfetch-0.37.0 → flacfetch-0.37.2}/flacfetch/api/routes/cache.py +0 -0
- {flacfetch-0.37.0 → flacfetch-0.37.2}/flacfetch/api/routes/config.py +0 -0
- {flacfetch-0.37.0 → flacfetch-0.37.2}/flacfetch/api/routes/credentials.py +0 -0
- {flacfetch-0.37.0 → flacfetch-0.37.2}/flacfetch/api/routes/download.py +0 -0
- {flacfetch-0.37.0 → flacfetch-0.37.2}/flacfetch/api/routes/health.py +0 -0
- {flacfetch-0.37.0 → flacfetch-0.37.2}/flacfetch/api/routes/search.py +0 -0
- {flacfetch-0.37.0 → flacfetch-0.37.2}/flacfetch/api/routes/spotify.py +0 -0
- {flacfetch-0.37.0 → flacfetch-0.37.2}/flacfetch/api/routes/torrents.py +0 -0
- {flacfetch-0.37.0 → flacfetch-0.37.2}/flacfetch/api/services/__init__.py +0 -0
- {flacfetch-0.37.0 → flacfetch-0.37.2}/flacfetch/api/services/credential_check.py +0 -0
- {flacfetch-0.37.0 → flacfetch-0.37.2}/flacfetch/api/services/disk_manager.py +0 -0
- {flacfetch-0.37.0 → flacfetch-0.37.2}/flacfetch/api/services/download_manager.py +0 -0
- {flacfetch-0.37.0 → flacfetch-0.37.2}/flacfetch/api/services/health_check.py +0 -0
- {flacfetch-0.37.0 → flacfetch-0.37.2}/flacfetch/api/services/search_cache.py +0 -0
- {flacfetch-0.37.0 → flacfetch-0.37.2}/flacfetch/api/services/secret_manager_utils.py +0 -0
- {flacfetch-0.37.0 → flacfetch-0.37.2}/flacfetch/api/services/spotify_budget.py +0 -0
- {flacfetch-0.37.0 → flacfetch-0.37.2}/flacfetch/core/album_match.py +0 -0
- {flacfetch-0.37.0 → flacfetch-0.37.2}/flacfetch/core/categorize.py +0 -0
- {flacfetch-0.37.0 → flacfetch-0.37.2}/flacfetch/core/config.py +0 -0
- {flacfetch-0.37.0 → flacfetch-0.37.2}/flacfetch/core/interfaces.py +0 -0
- {flacfetch-0.37.0 → flacfetch-0.37.2}/flacfetch/core/log.py +0 -0
- {flacfetch-0.37.0 → flacfetch-0.37.2}/flacfetch/core/manager.py +0 -0
- {flacfetch-0.37.0 → flacfetch-0.37.2}/flacfetch/core/models.py +0 -0
- {flacfetch-0.37.0 → flacfetch-0.37.2}/flacfetch/credential_keeper/__init__.py +0 -0
- {flacfetch-0.37.0 → flacfetch-0.37.2}/flacfetch/credential_keeper/__main__.py +0 -0
- {flacfetch-0.37.0 → flacfetch-0.37.2}/flacfetch/credential_keeper/browser.py +0 -0
- {flacfetch-0.37.0 → flacfetch-0.37.2}/flacfetch/credential_keeper/google_login.py +0 -0
- {flacfetch-0.37.0 → flacfetch-0.37.2}/flacfetch/credential_keeper/keeper.py +0 -0
- {flacfetch-0.37.0 → flacfetch-0.37.2}/flacfetch/credential_keeper/librespot.py +0 -0
- {flacfetch-0.37.0 → flacfetch-0.37.2}/flacfetch/credential_keeper/probe.py +0 -0
- {flacfetch-0.37.0 → flacfetch-0.37.2}/flacfetch/credential_keeper/refresh_request.py +0 -0
- {flacfetch-0.37.0 → flacfetch-0.37.2}/flacfetch/credential_keeper/spotify.py +0 -0
- {flacfetch-0.37.0 → flacfetch-0.37.2}/flacfetch/credential_keeper/youtube.py +0 -0
- {flacfetch-0.37.0 → flacfetch-0.37.2}/flacfetch/downloaders/torrent.py +0 -0
- {flacfetch-0.37.0 → flacfetch-0.37.2}/flacfetch/downloaders/torrent_coordinator.py +0 -0
- {flacfetch-0.37.0 → flacfetch-0.37.2}/flacfetch/downloaders/youtube.py +0 -0
- {flacfetch-0.37.0 → flacfetch-0.37.2}/flacfetch/interface/cli.py +0 -0
- {flacfetch-0.37.0 → flacfetch-0.37.2}/flacfetch/interface/cli_remote.py +0 -0
- {flacfetch-0.37.0 → flacfetch-0.37.2}/flacfetch/providers/ops.py +0 -0
- {flacfetch-0.37.0 → flacfetch-0.37.2}/flacfetch/providers/red.py +0 -0
- {flacfetch-0.37.0 → flacfetch-0.37.2}/flacfetch/providers/spotify.py +0 -0
- {flacfetch-0.37.0 → flacfetch-0.37.2}/flacfetch/providers/youtube.py +0 -0
- {flacfetch-0.37.0 → flacfetch-0.37.2}/flacfetch.egg-info/SOURCES.txt +0 -0
- {flacfetch-0.37.0 → flacfetch-0.37.2}/flacfetch.egg-info/dependency_links.txt +0 -0
- {flacfetch-0.37.0 → flacfetch-0.37.2}/flacfetch.egg-info/entry_points.txt +0 -0
- {flacfetch-0.37.0 → flacfetch-0.37.2}/flacfetch.egg-info/requires.txt +0 -0
- {flacfetch-0.37.0 → flacfetch-0.37.2}/flacfetch.egg-info/top_level.txt +0 -0
- {flacfetch-0.37.0 → flacfetch-0.37.2}/requirements.txt +0 -0
- {flacfetch-0.37.0 → flacfetch-0.37.2}/setup.cfg +0 -0
- {flacfetch-0.37.0 → flacfetch-0.37.2}/setup.py +0 -0
- {flacfetch-0.37.0 → flacfetch-0.37.2}/tests/_route_paths.py +0 -0
- {flacfetch-0.37.0 → flacfetch-0.37.2}/tests/test_album_search.py +0 -0
- {flacfetch-0.37.0 → flacfetch-0.37.2}/tests/test_api_auth.py +0 -0
- {flacfetch-0.37.0 → flacfetch-0.37.2}/tests/test_api_config.py +0 -0
- {flacfetch-0.37.0 → flacfetch-0.37.2}/tests/test_api_deep_health.py +0 -0
- {flacfetch-0.37.0 → flacfetch-0.37.2}/tests/test_api_health_ytdlp.py +0 -0
- {flacfetch-0.37.0 → flacfetch-0.37.2}/tests/test_api_integration.py +0 -0
- {flacfetch-0.37.0 → flacfetch-0.37.2}/tests/test_api_models.py +0 -0
- {flacfetch-0.37.0 → flacfetch-0.37.2}/tests/test_api_routes.py +0 -0
- {flacfetch-0.37.0 → flacfetch-0.37.2}/tests/test_api_services.py +0 -0
- {flacfetch-0.37.0 → flacfetch-0.37.2}/tests/test_cache_api.py +0 -0
- {flacfetch-0.37.0 → flacfetch-0.37.2}/tests/test_cli_cookies.py +0 -0
- {flacfetch-0.37.0 → flacfetch-0.37.2}/tests/test_cli_remote.py +0 -0
- {flacfetch-0.37.0 → flacfetch-0.37.2}/tests/test_cookie_refresh_request.py +0 -0
- {flacfetch-0.37.0 → flacfetch-0.37.2}/tests/test_core.py +0 -0
- {flacfetch-0.37.0 → flacfetch-0.37.2}/tests/test_credential_keeper.py +0 -0
- {flacfetch-0.37.0 → flacfetch-0.37.2}/tests/test_keeper_leaves_youtube.py +0 -0
- {flacfetch-0.37.0 → flacfetch-0.37.2}/tests/test_keeper_probe.py +0 -0
- {flacfetch-0.37.0 → flacfetch-0.37.2}/tests/test_logging.py +0 -0
- {flacfetch-0.37.0 → flacfetch-0.37.2}/tests/test_manager.py +0 -0
- {flacfetch-0.37.0 → flacfetch-0.37.2}/tests/test_models.py +0 -0
- {flacfetch-0.37.0 → flacfetch-0.37.2}/tests/test_ops.py +0 -0
- {flacfetch-0.37.0 → flacfetch-0.37.2}/tests/test_provider_errors.py +0 -0
- {flacfetch-0.37.0 → flacfetch-0.37.2}/tests/test_red.py +0 -0
- {flacfetch-0.37.0 → flacfetch-0.37.2}/tests/test_search_api.py +0 -0
- {flacfetch-0.37.0 → flacfetch-0.37.2}/tests/test_spotify_budget.py +0 -0
- {flacfetch-0.37.0 → flacfetch-0.37.2}/tests/test_torrent_coordinator.py +0 -0
- {flacfetch-0.37.0 → flacfetch-0.37.2}/tests/test_torrent_downloader.py +0 -0
- {flacfetch-0.37.0 → flacfetch-0.37.2}/tests/test_torrent_shared_album.py +0 -0
- {flacfetch-0.37.0 → flacfetch-0.37.2}/tests/test_transcode_hf_noise.py +0 -0
- {flacfetch-0.37.0 → flacfetch-0.37.2}/tests/test_youtube.py +0 -0
- {flacfetch-0.37.0 → flacfetch-0.37.2}/tests/test_youtube_cookies.py +0 -0
- {flacfetch-0.37.0 → flacfetch-0.37.2}/tests/test_youtube_provider.py +0 -0
|
@@ -1,6 +1,14 @@
|
|
|
1
1
|
import re
|
|
2
2
|
from difflib import SequenceMatcher
|
|
3
3
|
|
|
4
|
+
# Straight, curly and backtick apostrophes. People type "lets" / "dont" without
|
|
5
|
+
# them while release files say "Let's" / "Don’t", so matching ignores them.
|
|
6
|
+
_APOSTROPHES = re.compile(r"['\u2018\u2019`]")
|
|
7
|
+
|
|
8
|
+
|
|
9
|
+
def _strip_apostrophes(text: str) -> str:
|
|
10
|
+
return _APOSTROPHES.sub("", text)
|
|
11
|
+
|
|
4
12
|
|
|
5
13
|
def clean_filename(filename: str) -> str:
|
|
6
14
|
# Remove file extension
|
|
@@ -9,6 +17,8 @@ def clean_filename(filename: str) -> str:
|
|
|
9
17
|
# Remove track numbers like "01 - ", "01. ", "1-", "A1 "
|
|
10
18
|
filename = re.sub(r'^(\d{1,3}[ .-]+)+', '', filename)
|
|
11
19
|
filename = re.sub(r'^[A-D]\d{1,2}[ .-]+', '', filename)
|
|
20
|
+
# Scene-style names use underscores for spaces ("lets_get_retarded")
|
|
21
|
+
filename = filename.replace('_', ' ')
|
|
12
22
|
# Remove " - " separators which might remain
|
|
13
23
|
filename = filename.replace(' - ', ' ')
|
|
14
24
|
return filename.strip()
|
|
@@ -18,8 +28,8 @@ def calculate_match_score(track_title: str, filename: str) -> float:
|
|
|
18
28
|
Returns a score 0.0 - 1.0 indicating how well the filename matches the track title.
|
|
19
29
|
1.0 = Exact match (ignoring case/punctuation)
|
|
20
30
|
"""
|
|
21
|
-
cleaned_name = clean_filename(filename).lower()
|
|
22
|
-
target = track_title.lower()
|
|
31
|
+
cleaned_name = _strip_apostrophes(clean_filename(filename).lower())
|
|
32
|
+
target = _strip_apostrophes(track_title.lower())
|
|
23
33
|
|
|
24
34
|
# Exact match check
|
|
25
35
|
if cleaned_name == target:
|
|
@@ -37,6 +37,15 @@ SAMPLE_RATE = 44100
|
|
|
37
37
|
CHANNELS = 2
|
|
38
38
|
BIT_DEPTH = 16
|
|
39
39
|
|
|
40
|
+
# Used when a track's duration can't be resolved (see _resolve_duration_secs).
|
|
41
|
+
DEFAULT_DURATION_SECS = 300
|
|
42
|
+
|
|
43
|
+
# A capture longer than the track by more than this is refused: it means
|
|
44
|
+
# librespot played on past the requested track (e.g. Spotify autoplay queued a
|
|
45
|
+
# "similar" track), so the PCM contains someone else's audio. Durations are
|
|
46
|
+
# floored to whole seconds, so allow a small margin.
|
|
47
|
+
MAX_CAPTURE_OVERRUN_SECS = 5
|
|
48
|
+
|
|
40
49
|
# Serialize all Spotify captures across the process.
|
|
41
50
|
#
|
|
42
51
|
# A Spotify download works by registering a librespot Spotify Connect device on
|
|
@@ -153,6 +162,11 @@ class SpotifyDownloader(Downloader):
|
|
|
153
162
|
def _resolve_duration_secs(
|
|
154
163
|
self, sp: "spotipy.Spotify", track_id: str, release: Release
|
|
155
164
|
) -> int:
|
|
165
|
+
return self._resolve_duration(sp, track_id, release)[0]
|
|
166
|
+
|
|
167
|
+
def _resolve_duration(
|
|
168
|
+
self, sp: "spotipy.Spotify", track_id: str, release: Release
|
|
169
|
+
) -> tuple[int, bool]:
|
|
156
170
|
"""Resolve track duration in seconds.
|
|
157
171
|
|
|
158
172
|
The PCM capture loop compares captured bytes against an expected size
|
|
@@ -164,18 +178,20 @@ class SpotifyDownloader(Downloader):
|
|
|
164
178
|
Prefer the duration already on the release (from search). Otherwise
|
|
165
179
|
look it up via the Web API, which is reliable and cheap. Only fall
|
|
166
180
|
back to the 300s default if both fail.
|
|
181
|
+
|
|
182
|
+
Returns ``(seconds, known)``; ``known`` is False for the default guess.
|
|
167
183
|
"""
|
|
168
184
|
if release.duration_seconds:
|
|
169
|
-
return release.duration_seconds
|
|
185
|
+
return release.duration_seconds, True
|
|
170
186
|
try:
|
|
171
187
|
duration_ms = sp.track(track_id).get("duration_ms")
|
|
172
188
|
if duration_ms:
|
|
173
189
|
secs = duration_ms // 1000
|
|
174
190
|
logger.debug(f"Resolved duration from Spotify API: {secs}s")
|
|
175
|
-
return secs
|
|
191
|
+
return secs, True
|
|
176
192
|
except Exception as e:
|
|
177
193
|
logger.warning(f"Failed to look up track duration, using default: {e}")
|
|
178
|
-
return
|
|
194
|
+
return DEFAULT_DURATION_SECS, False
|
|
179
195
|
|
|
180
196
|
def download(
|
|
181
197
|
self,
|
|
@@ -246,7 +262,7 @@ class SpotifyDownloader(Downloader):
|
|
|
246
262
|
except Exception as e:
|
|
247
263
|
raise SpotifyDownloadError(f"Authentication failed: {e}") from e
|
|
248
264
|
|
|
249
|
-
duration_secs = self.
|
|
265
|
+
duration_secs, duration_known = self._resolve_duration(sp, track_id, release)
|
|
250
266
|
|
|
251
267
|
# Unique device name per capture. A single Spotify account can register
|
|
252
268
|
# several Connect devices at once, so if a previous librespot has not
|
|
@@ -267,6 +283,12 @@ class SpotifyDownloader(Downloader):
|
|
|
267
283
|
"--backend", "pipe",
|
|
268
284
|
"--bitrate", "320",
|
|
269
285
|
"--disable-discovery",
|
|
286
|
+
# Never play past the requested track. librespot decodes faster than
|
|
287
|
+
# real time, so anything autoplay queues after our track would be
|
|
288
|
+
# piped straight into this capture. Since Spotify started sending
|
|
289
|
+
# Web API plays as a "spotify:webapi:<id>" context (2026-11), a
|
|
290
|
+
# librespot that handles that context also autoplays after it.
|
|
291
|
+
"--autoplay", "off",
|
|
270
292
|
# Volume settings for full-quality recording:
|
|
271
293
|
"--initial-volume", "100", # Start at max volume
|
|
272
294
|
"--volume-ctrl", "fixed", # No dynamic volume adjustments
|
|
@@ -383,6 +405,19 @@ class SpotifyDownloader(Downloader):
|
|
|
383
405
|
captured_s = pcm_size / (SAMPLE_RATE * 4)
|
|
384
406
|
logger.debug(f"Captured {pcm_size / (1024*1024):.2f} MB PCM")
|
|
385
407
|
|
|
408
|
+
# Refuse a capture that ran on past the requested track: the extra
|
|
409
|
+
# audio is a different song. Skipped when the duration is only the
|
|
410
|
+
# default guess.
|
|
411
|
+
if (
|
|
412
|
+
duration_known
|
|
413
|
+
and captured_s > duration_secs + MAX_CAPTURE_OVERRUN_SECS
|
|
414
|
+
):
|
|
415
|
+
raise SpotifyDownloadError(
|
|
416
|
+
f"Captured {captured_s:.1f}s of audio but {track_uri} is "
|
|
417
|
+
f"{duration_secs}s long; librespot played past the track, "
|
|
418
|
+
"refusing to save audio that includes another song"
|
|
419
|
+
)
|
|
420
|
+
|
|
386
421
|
# Convert PCM to a per-download temp FLAC, then atomically move it
|
|
387
422
|
# into place. The atomic rename guarantees the returned path is
|
|
388
423
|
# always a complete file even if a concurrent download targets the
|
|
@@ -34,6 +34,27 @@ ALBUM_SEARCH_MAX_GROUPS = 3
|
|
|
34
34
|
# Quotes: ' (apostrophe breaks searches like "I'm With You")
|
|
35
35
|
SPHINX_SPECIAL_CHARS = frozenset(r'()|\\-@~&<>!"/*$^=?:[],.;\'')
|
|
36
36
|
|
|
37
|
+
# Curly apostrophes split words in the tracker index just like straight ones.
|
|
38
|
+
APOSTROPHE_LIKE_CHARS = frozenset('\u2018\u2019`')
|
|
39
|
+
|
|
40
|
+
# Contractions people type without the apostrophe ("lets get it started").
|
|
41
|
+
# The tracker index splits "Let's" into "let" + "s", so the filelist search for
|
|
42
|
+
# "lets" finds nothing; the fallback search spells these out as "let s".
|
|
43
|
+
# Only words that are not also ordinary words (so no "ill", "well", "were").
|
|
44
|
+
APOSTROPHE_LESS_CONTRACTIONS = {
|
|
45
|
+
"aint": "ain t", "arent": "aren t", "cant": "can t", "couldnt": "couldn t",
|
|
46
|
+
"couldve": "could ve", "didnt": "didn t", "doesnt": "doesn t", "dont": "don t",
|
|
47
|
+
"hadnt": "hadn t", "hasnt": "hasn t", "havent": "haven t", "hes": "he s",
|
|
48
|
+
"heres": "here s", "im": "i m", "isnt": "isn t", "its": "it s", "ive": "i ve",
|
|
49
|
+
"lets": "let s", "mustnt": "mustn t", "shes": "she s", "shouldnt": "shouldn t",
|
|
50
|
+
"shouldve": "should ve", "thats": "that s", "theres": "there s", "theyd": "they d",
|
|
51
|
+
"theyll": "they ll", "theyre": "they re", "theyve": "they ve", "wasnt": "wasn t",
|
|
52
|
+
"weve": "we ve", "werent": "weren t", "whats": "what s", "wheres": "where s",
|
|
53
|
+
"whos": "who s", "wont": "won t", "wouldnt": "wouldn t", "wouldve": "would ve",
|
|
54
|
+
"yall": "y all", "youd": "you d", "youll": "you ll", "youre": "you re",
|
|
55
|
+
"youve": "you ve",
|
|
56
|
+
}
|
|
57
|
+
|
|
37
58
|
|
|
38
59
|
class GazelleProvider(Provider):
|
|
39
60
|
"""Base class for Gazelle-based private music trackers.
|
|
@@ -161,7 +182,7 @@ class GazelleProvider(Provider):
|
|
|
161
182
|
are NOT modified - Sphinx handles these via charset_table and ngram_chars.
|
|
162
183
|
"""
|
|
163
184
|
sanitized = query
|
|
164
|
-
for char in SPHINX_SPECIAL_CHARS:
|
|
185
|
+
for char in SPHINX_SPECIAL_CHARS | APOSTROPHE_LIKE_CHARS:
|
|
165
186
|
sanitized = sanitized.replace(char, ' ')
|
|
166
187
|
|
|
167
188
|
# Collapse multiple spaces into one
|
|
@@ -170,6 +191,19 @@ class GazelleProvider(Provider):
|
|
|
170
191
|
logger.debug(f"Sanitized filelist query: '{query}' -> '{sanitized}'")
|
|
171
192
|
return sanitized
|
|
172
193
|
|
|
194
|
+
@staticmethod
|
|
195
|
+
def _expand_contractions(sanitized_title: str) -> Optional[str]:
|
|
196
|
+
"""Spell out apostrophe-less contractions the way the index stores them.
|
|
197
|
+
|
|
198
|
+
"lets get retarded" -> "let s get retarded". Returns None when the title
|
|
199
|
+
has none, so the caller knows there is no fallback search to run.
|
|
200
|
+
"""
|
|
201
|
+
words = sanitized_title.split()
|
|
202
|
+
expanded = [APOSTROPHE_LESS_CONTRACTIONS.get(w.lower(), w) for w in words]
|
|
203
|
+
if expanded == words:
|
|
204
|
+
return None
|
|
205
|
+
return " ".join(expanded)
|
|
206
|
+
|
|
173
207
|
def _find_best_target_file(self, file_list_str: str, track_title: str) -> tuple[Optional[str], Optional[int], float]:
|
|
174
208
|
"""Find the best matching file in a torrent's file list.
|
|
175
209
|
|
|
@@ -1062,57 +1096,79 @@ class GazelleProvider(Provider):
|
|
|
1062
1096
|
# Sanitize the filelist query to remove Sphinx special operators
|
|
1063
1097
|
sanitized_title = self._sanitize_filelist_query(query.title) if query.title else ""
|
|
1064
1098
|
|
|
1099
|
+
try:
|
|
1100
|
+
releases = self._search_browse_filelist(query, sanitized_title)
|
|
1101
|
+
# "lets" only finds files that also drop the apostrophe (often a
|
|
1102
|
+
# stray compilation), never the album's "Let's", so always search
|
|
1103
|
+
# the spelled-out form too and merge.
|
|
1104
|
+
expanded = self._expand_contractions(sanitized_title)
|
|
1105
|
+
if expanded:
|
|
1106
|
+
logger.info(f"{self.name}: also searching contraction form '{expanded}'")
|
|
1107
|
+
seen = {r.source_id for r in releases}
|
|
1108
|
+
try:
|
|
1109
|
+
extra = self._search_browse_filelist(query, expanded)
|
|
1110
|
+
except Exception as e:
|
|
1111
|
+
# Never lose the first search's matches to the extra one.
|
|
1112
|
+
logger.warning(f"{self.name}: contraction search failed, keeping first results: {e}")
|
|
1113
|
+
extra = []
|
|
1114
|
+
for r in extra:
|
|
1115
|
+
if r.source_id not in seen:
|
|
1116
|
+
seen.add(r.source_id)
|
|
1117
|
+
releases.append(r)
|
|
1118
|
+
return releases
|
|
1119
|
+
except requests.RequestException as e:
|
|
1120
|
+
logger.error(f"Connection error to {self.name}: {e}")
|
|
1121
|
+
return []
|
|
1122
|
+
except Exception as e:
|
|
1123
|
+
logger.exception(f"Unexpected error in {self.name}Provider: {e}")
|
|
1124
|
+
return []
|
|
1125
|
+
|
|
1126
|
+
def _search_browse_filelist(self, query: TrackQuery, filelist: str) -> list[Release]:
|
|
1127
|
+
"""One browse by artist + filelist words, then file lists for the top groups."""
|
|
1065
1128
|
params = {
|
|
1066
1129
|
"action": "browse",
|
|
1067
1130
|
"artistname": query.artist,
|
|
1068
|
-
"filelist":
|
|
1131
|
+
"filelist": filelist,
|
|
1069
1132
|
"format": "FLAC"
|
|
1070
1133
|
}
|
|
1071
1134
|
|
|
1072
1135
|
logger.debug(f"Searching {self.name} with params: {params}")
|
|
1073
1136
|
|
|
1074
|
-
|
|
1075
|
-
|
|
1076
|
-
logger.debug(f"Found {len(ordered_group_ids)} groups in {self.name} response")
|
|
1137
|
+
ordered_group_ids, token_eligibility = self._browse(params)
|
|
1138
|
+
logger.debug(f"Found {len(ordered_group_ids)} groups in {self.name} response")
|
|
1077
1139
|
|
|
1078
|
-
|
|
1140
|
+
limited_group_ids = ordered_group_ids[:self.search_limit]
|
|
1079
1141
|
|
|
1080
|
-
|
|
1081
|
-
|
|
1142
|
+
if len(ordered_group_ids) > self.search_limit:
|
|
1143
|
+
logger.info(f"Limiting detailed fetch to top {self.search_limit} groups (out of {len(ordered_group_ids)} found)")
|
|
1082
1144
|
|
|
1083
|
-
|
|
1145
|
+
logger.info(f"Fetching details for up to {len(limited_group_ids)} groups to resolve file lists...")
|
|
1084
1146
|
|
|
1085
|
-
|
|
1086
|
-
|
|
1087
|
-
|
|
1088
|
-
|
|
1089
|
-
|
|
1090
|
-
|
|
1091
|
-
|
|
1092
|
-
|
|
1093
|
-
|
|
1094
|
-
|
|
1095
|
-
|
|
1096
|
-
|
|
1097
|
-
|
|
1098
|
-
|
|
1099
|
-
|
|
1100
|
-
|
|
1101
|
-
|
|
1102
|
-
|
|
1103
|
-
|
|
1104
|
-
|
|
1105
|
-
|
|
1106
|
-
|
|
1107
|
-
|
|
1108
|
-
|
|
1109
|
-
|
|
1110
|
-
|
|
1111
|
-
|
|
1112
|
-
|
|
1113
|
-
except requests.RequestException as e:
|
|
1114
|
-
logger.error(f"Connection error to {self.name}: {e}")
|
|
1115
|
-
return []
|
|
1116
|
-
except Exception as e:
|
|
1117
|
-
logger.exception(f"Unexpected error in {self.name}Provider: {e}")
|
|
1118
|
-
return []
|
|
1147
|
+
releases = []
|
|
1148
|
+
excellent_found = False
|
|
1149
|
+
groups_fetched = 0
|
|
1150
|
+
|
|
1151
|
+
for gid in limited_group_ids:
|
|
1152
|
+
# Rate limit: sleep before each request
|
|
1153
|
+
time.sleep(1.1)
|
|
1154
|
+
|
|
1155
|
+
group_releases = self._fetch_group_details(gid, query.title, token_eligibility)
|
|
1156
|
+
releases.extend(group_releases)
|
|
1157
|
+
groups_fetched += 1
|
|
1158
|
+
|
|
1159
|
+
# Check for early termination
|
|
1160
|
+
if self.early_termination and not excellent_found:
|
|
1161
|
+
for r in group_releases:
|
|
1162
|
+
if (r.seeders and r.seeders >= self.early_termination_seeders and
|
|
1163
|
+
r.release_type in self.early_termination_release_types):
|
|
1164
|
+
excellent_found = True
|
|
1165
|
+
logger.info(f"Found excellent result: {r.artist} - {r.title} ({r.seeders} seeders, {r.release_type})")
|
|
1166
|
+
break
|
|
1167
|
+
|
|
1168
|
+
# If we found an excellent result, fetch a few more groups then stop
|
|
1169
|
+
if excellent_found and groups_fetched >= min(5, len(limited_group_ids)):
|
|
1170
|
+
logger.info(f"Early termination: stopping after {groups_fetched} groups (found excellent result)")
|
|
1171
|
+
break
|
|
1172
|
+
|
|
1173
|
+
logger.info(f"Total matching tracks parsed from {self.name}: {len(releases)}")
|
|
1174
|
+
return releases
|
|
@@ -1080,3 +1080,59 @@ class TestDownloadBudget:
|
|
|
1080
1080
|
p.fetch_artifact_by_id("42")
|
|
1081
1081
|
ledger = p._load_download_ledger("42")
|
|
1082
1082
|
assert [r["kind"] for r in ledger["requests"]] == ["token"]
|
|
1083
|
+
|
|
1084
|
+
|
|
1085
|
+
class TestContractionFallback:
|
|
1086
|
+
"""Apostrophe-less contractions ("lets") miss the tracker index, which stores "let s"."""
|
|
1087
|
+
|
|
1088
|
+
@pytest.fixture
|
|
1089
|
+
def provider(self):
|
|
1090
|
+
return ConcreteGazelleProvider(api_key="test", base_url="https://test.example", cache_subdir="test")
|
|
1091
|
+
|
|
1092
|
+
@pytest.mark.parametrize("title,expected", [
|
|
1093
|
+
("lets get retarded", "let s get retarded"),
|
|
1094
|
+
("Dont Stop Believin", "don t Stop Believin"),
|
|
1095
|
+
("Im Yours", "i m Yours"),
|
|
1096
|
+
("Hello", None),
|
|
1097
|
+
("well ill be there", None),
|
|
1098
|
+
])
|
|
1099
|
+
def test_expand_contractions(self, provider, title, expected):
|
|
1100
|
+
assert provider._expand_contractions(title) == expected
|
|
1101
|
+
|
|
1102
|
+
def test_curly_apostrophe_sanitized(self, provider):
|
|
1103
|
+
assert provider._sanitize_filelist_query("Let’s Get It Started") == "Let s Get It Started"
|
|
1104
|
+
|
|
1105
|
+
def test_also_searches_expanded_title_and_merges(self, provider, monkeypatch):
|
|
1106
|
+
from flacfetch.core.models import TrackQuery
|
|
1107
|
+
comp, album, dup = MagicMock(source_id="1"), MagicMock(source_id="2"), MagicMock(source_id="1")
|
|
1108
|
+
calls = []
|
|
1109
|
+
|
|
1110
|
+
def fake(query, filelist):
|
|
1111
|
+
calls.append(filelist)
|
|
1112
|
+
return [comp] if filelist == "lets get retarded" else [album, dup]
|
|
1113
|
+
|
|
1114
|
+
monkeypatch.setattr(provider, "_search_browse_filelist", fake)
|
|
1115
|
+
result = provider._search_browse(TrackQuery(artist="black eyed peas", title="lets get retarded"))
|
|
1116
|
+
assert calls == ["lets get retarded", "let s get retarded"]
|
|
1117
|
+
assert result == [comp, album]
|
|
1118
|
+
|
|
1119
|
+
def test_expanded_search_failure_keeps_first_results(self, provider, monkeypatch):
|
|
1120
|
+
import requests
|
|
1121
|
+
|
|
1122
|
+
from flacfetch.core.models import TrackQuery
|
|
1123
|
+
comp = MagicMock(source_id="1")
|
|
1124
|
+
|
|
1125
|
+
def fake(query, filelist):
|
|
1126
|
+
if filelist == "let s get retarded":
|
|
1127
|
+
raise requests.Timeout("slow tracker")
|
|
1128
|
+
return [comp]
|
|
1129
|
+
|
|
1130
|
+
monkeypatch.setattr(provider, "_search_browse_filelist", fake)
|
|
1131
|
+
assert provider._search_browse(TrackQuery(artist="x", title="lets get retarded")) == [comp]
|
|
1132
|
+
|
|
1133
|
+
def test_no_retry_without_contractions(self, provider, monkeypatch):
|
|
1134
|
+
from flacfetch.core.models import TrackQuery
|
|
1135
|
+
calls = []
|
|
1136
|
+
monkeypatch.setattr(provider, "_search_browse_filelist", lambda q, f: calls.append(f) or [])
|
|
1137
|
+
assert provider._search_browse(TrackQuery(artist="x", title="Hello")) == []
|
|
1138
|
+
assert calls == ["Hello"]
|
|
@@ -27,3 +27,15 @@ def test_no_match():
|
|
|
27
27
|
score = calculate_match_score("Tonight", "Tomorrow.flac")
|
|
28
28
|
assert score < 0.5
|
|
29
29
|
|
|
30
|
+
|
|
31
|
+
|
|
32
|
+
def test_apostrophe_less_title_matches_contraction():
|
|
33
|
+
# Typed "lets", the file says "Let's" (straight or curly)
|
|
34
|
+
assert calculate_match_score("lets get retarded", "05 - Let's Get Retarded.flac") == 1.0
|
|
35
|
+
assert calculate_match_score("lets get retarded", "05 - Let’s Get Retarded.flac") == 1.0
|
|
36
|
+
assert calculate_match_score("Let's Get Retarded", "05 - Lets Get Retarded.flac") == 1.0
|
|
37
|
+
|
|
38
|
+
|
|
39
|
+
def test_underscored_scene_filename_matches():
|
|
40
|
+
score = calculate_match_score("lets get retarded", "306-black_eyed_peas-lets_get_retarded.flac")
|
|
41
|
+
assert score > 0.6
|
|
@@ -989,3 +989,86 @@ class TestAudioFormatOutput:
|
|
|
989
989
|
vorbis_320 = Quality(format=AudioFormat.VORBIS, bitrate=320)
|
|
990
990
|
|
|
991
991
|
assert vorbis_320 < flac
|
|
992
|
+
|
|
993
|
+
|
|
994
|
+
class TestSpotifyCaptureStopsAtTrackEnd:
|
|
995
|
+
"""A capture must contain only the requested track.
|
|
996
|
+
|
|
997
|
+
2026-11-19: Spotify began sending Web API plays to Connect devices as a
|
|
998
|
+
``spotify:webapi:<id>`` context. librespot 0.8.0 fails those plays (the
|
|
999
|
+
Web API reports "403 Restriction violated"); a librespot that handles them
|
|
1000
|
+
also autoplays a similar track afterwards, and because librespot decodes
|
|
1001
|
+
faster than real time, a test capture of a 191 s track came back 378 s long.
|
|
1002
|
+
"""
|
|
1003
|
+
|
|
1004
|
+
def _run(self, tmp_path, captured_secs, duration_secs=200, api_duration_ms=None):
|
|
1005
|
+
creds_dir = tmp_path / "librespot"
|
|
1006
|
+
creds_dir.mkdir()
|
|
1007
|
+
(creds_dir / "credentials.json").write_text("{}")
|
|
1008
|
+
provider = MagicMock()
|
|
1009
|
+
provider._get_client.return_value.track.return_value = {"duration_ms": api_duration_ms}
|
|
1010
|
+
downloader = SpotifyDownloader(provider=provider)
|
|
1011
|
+
downloader._librespot_path = "/fake/librespot"
|
|
1012
|
+
release = Release(
|
|
1013
|
+
title="Test",
|
|
1014
|
+
artist="Test",
|
|
1015
|
+
quality=Quality(AudioFormat.FLAC),
|
|
1016
|
+
source_name="Spotify",
|
|
1017
|
+
download_url="spotify:track:4PTG3Z6ehGkBFwjybzWkR8",
|
|
1018
|
+
duration_seconds=duration_secs,
|
|
1019
|
+
)
|
|
1020
|
+
|
|
1021
|
+
def fake_wait_for_download(pcm_path, *_args, **_kwargs):
|
|
1022
|
+
pcm_path.write_bytes(b"\x00" * int(captured_secs * 44100 * 4))
|
|
1023
|
+
|
|
1024
|
+
def _convert(pcm_path, flac_tmp_path):
|
|
1025
|
+
flac_tmp_path.write_bytes(b"")
|
|
1026
|
+
return True
|
|
1027
|
+
|
|
1028
|
+
proc = MagicMock()
|
|
1029
|
+
proc.poll.return_value = None
|
|
1030
|
+
with patch(
|
|
1031
|
+
"flacfetch.downloaders.spotify.get_librespot_credentials_dir",
|
|
1032
|
+
return_value=str(creds_dir),
|
|
1033
|
+
), patch(
|
|
1034
|
+
"flacfetch.downloaders.spotify.subprocess.Popen", return_value=proc
|
|
1035
|
+
) as mock_popen, patch.object(
|
|
1036
|
+
downloader, "_wait_for_device", return_value={"id": "dev"}
|
|
1037
|
+
), patch.object(
|
|
1038
|
+
downloader, "_wait_for_track_load", return_value=True
|
|
1039
|
+
), patch.object(
|
|
1040
|
+
downloader, "_wait_for_download", side_effect=fake_wait_for_download
|
|
1041
|
+
), patch.object(downloader, "_stop_librespot"), patch.object(
|
|
1042
|
+
downloader, "_convert_pcm_to_flac", side_effect=_convert
|
|
1043
|
+
):
|
|
1044
|
+
try:
|
|
1045
|
+
result = downloader.download(release, str(tmp_path / "out"))
|
|
1046
|
+
except SpotifyDownloadError as e:
|
|
1047
|
+
return mock_popen.call_args.args[0], e, None
|
|
1048
|
+
return mock_popen.call_args.args[0], None, result
|
|
1049
|
+
|
|
1050
|
+
def test_librespot_runs_with_autoplay_off(self, tmp_path):
|
|
1051
|
+
cmd, _err, _result = self._run(tmp_path, captured_secs=200)
|
|
1052
|
+
i = cmd.index("--autoplay")
|
|
1053
|
+
assert cmd[i + 1] == "off"
|
|
1054
|
+
|
|
1055
|
+
def test_capture_matching_track_length_succeeds(self, tmp_path):
|
|
1056
|
+
_cmd, err, result = self._run(tmp_path, captured_secs=200.9)
|
|
1057
|
+
assert err is None
|
|
1058
|
+
assert result.captured_s == pytest.approx(200.9, abs=0.01)
|
|
1059
|
+
|
|
1060
|
+
def test_capture_running_past_track_is_refused(self, tmp_path):
|
|
1061
|
+
_cmd, err, result = self._run(tmp_path, captured_secs=378)
|
|
1062
|
+
assert result is None
|
|
1063
|
+
assert "played past the track" in str(err)
|
|
1064
|
+
assert not list((tmp_path / "out").glob("*.flac"))
|
|
1065
|
+
|
|
1066
|
+
def test_overrun_check_skipped_when_duration_is_only_the_default(self, tmp_path):
|
|
1067
|
+
# No duration on the release and none from the API -> 300 s guess.
|
|
1068
|
+
_cmd, err, _result = self._run(tmp_path, captured_secs=420, duration_secs=None)
|
|
1069
|
+
assert err is None
|
|
1070
|
+
|
|
1071
|
+
def test_overrun_check_applies_to_a_real_300s_track(self, tmp_path):
|
|
1072
|
+
_cmd, err, result = self._run(tmp_path, captured_secs=420, duration_secs=300)
|
|
1073
|
+
assert result is None
|
|
1074
|
+
assert "played past the track" in str(err)
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|