feat: write playlists into the mirror from the listening history

Six rules over the scrobble history and the library index, written as extended
M3U into <mirror>/_playlists/ and rebuilt every pass. Three look at what gets
played -- heavy rotation, all-time, and things played heavily once and silent
for a year. Three look at what does not: album tracks skipped on records
otherwise played constantly, unplayed tracks by the artists played most, and
unplayed tracks from anywhere. Four fifths of the library has never been played,
so both halves are needed.

Ninety days was the obvious window for "recent" and is the wrong one. On a real
history it holds a few hundred plays spread across a twenty-thousand track
rotation, and nothing ranks meaningfully. Twelve months does.

The two rotating playlists are shuffled by week rather than by pass. A pass runs
every few hours; a playlist that reorders itself each time is one that has to be
re-imported each time, because the Music app imports a snapshot of a file rather
than tracking it.

Lidarr knows where the lossless source is, and the playlists have to point at
the MP3s music-mirror produced from it. The library root is derived from the
common parent of the indexed artist folders, so it agrees with Lidarr by
construction instead of being kept in step by hand, and can be overridden.
Entries are relative to the playlist file, so one file works from the NAS, from
a Mac over SMB and from Linux.

A track is listed only once its mirror file has been confirmed to exist: Lidarr
holding the FLAC says nothing about whether the MP3 has been encoded. When many
are missing the run says so, because that is what a misconfigured root looks
like -- every path mapping to nothing -- and silence there would present an
empty playlist as a correct one.

README.md

@@ -7,9 +7,10 @@ which keeps an MP3 copy of a lossless library for an iPod. This one answers the
 question that mirror cannot: which of it is worth carrying, and which of it has
 not been played in years.
 
-**This is stage two.** It ingests the scrobble history, indexes the library from
-Lidarr, and matches one to the other. There are no playlists yet, and it writes
-nothing back — every Lidarr call is a `GET`. See "Where this is going" below.
+**This is stage three.** It ingests the scrobble history, indexes the library
+from Lidarr, matches one to the other, and writes playlists into the mirror.
+Nothing is written back to Lidarr — every call there is a `GET`. See "Where this
+is going" below.
 
 ## What it does today
 
@@ -18,6 +19,7 @@ nothing back — every Lidarr call is a `GET`. See "Where this is going" below.
 - Indexes every artist, album and track Lidarr knows about, with file paths and
   the date each file landed.
 - Ties the two together and reports how well it managed.
+- Writes M3U playlists into the mirror, from the listening history.
 
 ## Matching
 
@@ -126,6 +128,50 @@ and a shared title is a weak proxy for a shared song:
 Only the first is worth chasing. Counting all three as matcher failures
 overstates the problem and would over-block the cull.
 
+## Playlists
+
+Written into `<mirror>/_playlists/` as extended M3U, rebuilt every pass. Six
+rules, capped at `--playlist-limit` tracks each:
+
+| Playlist | Rule |
+| -------------------- | --------------------------------------------------------- |
+| `heavy-rotation` | Most played over the last twelve months |
+| `all-time` | Most played ever |
+| `neglected` | Played heavily once, silent for twelve months |
+| `deep-cuts` | Never played, from albums whose other tracks you play constantly |
+| `unheard-favourites` | Never played, by the artists you play most |
+| `unheard` | Never played, anywhere in the library |
+
+Ninety days was the obvious window for "recent" and is the wrong one: on a real
+history it holds a few hundred plays spread thinly across a twenty-thousand
+track rotation, so nothing ranks meaningfully. Twelve months does.
+
+The two `unheard` playlists rotate **weekly**, not per pass. A pass runs every
+few hours, and a playlist that reorders itself each time is one that has to be
+re-imported each time — the Music app imports a snapshot of a file, it does not
+track it.
+
+### Paths
+
+Lidarr knows where the lossless source is; the playlists have to point at the
+MP3s music-mirror made from it. The mapping strips a library root from Lidarr's
+track paths and re-roots them under the mirror, with the suffix changed.
+
+`--library-root` is derived from the common parent of the indexed artist folders
+when unset, so it agrees with Lidarr by construction rather than by being kept
+in step by hand. Override it if that guess is wrong.
+
+Entries are written **relative to the playlist file**, so one playlist works
+from the NAS, from a Mac over SMB, and from Linux, without rewriting.
+
+A track is only listed once its mirror file has been confirmed to exist. Lidarr
+holding the FLAC says nothing about whether the MP3 has been encoded yet. If a
+large number are missing, the run says so — that is what a wrong `--library-root`
+or `--mirror` looks like, since the paths then map to nothing at all.
+
+`_playlists/` survives music-mirror's prune: it only deletes `*.mp3`, and its
+empty-directory sweep skips a directory holding M3Us.
+
 ## How the ingest works
 
 Two halves, both taking their bounds from the database rather than from a saved
@@ -180,6 +226,9 @@ music-curator --report-only   # report on the store, fetch nothing
 | `--backfill-limit` | `MUSIC_CURATOR_BACKFILL_LIMIT` | `0` | Cap backfill requests per pass; 0 for no cap |
 | `--lidarr-url` | `MUSIC_CURATOR_LIDARR_URL` | unset | Lidarr base URL, e.g. `http://lidarr:8686` |
 | `--lidarr-api-key` | `MUSIC_CURATOR_LIDARR_API_KEY` | unset | Lidarr API key |
+| `--mirror` | `MUSIC_CURATOR_MIRROR` | unset | Root of the MP3 mirror; playlists go here |
+| `--library-root` | `MUSIC_CURATOR_LIBRARY_ROOT` | derived | Prefix to strip from Lidarr's paths |
+| `--playlist-limit` | `MUSIC_CURATOR_PLAYLIST_LIMIT` | `100` | Most tracks in any one playlist |
 | `--skip-index` | — | off | Match against the index already held |
 | `--report-only` | — | off | Report without fetching |
 
@@ -228,7 +277,8 @@ nix shell nixpkgs#python3Packages.pytest -c pytest
 | ------------------------------------------------ | ------------ |
 | Last.fm ingest and store | done |
 | Lidarr index and the scrobble-to-track matcher | done |
-| M3U playlists written into the mirror | next |
+| M3U playlists from the listening history | done |
+| Genre and mood playlists from Last.fm tags | next |
 | Cold-music report, unmonitoring what is not played | last |
 
 The cull will unmonitor cold albums in Lidarr and tag their artists. It will

compose.yaml

@@ -30,5 +30,11 @@ services:
       # Cap the backfill at this many requests per pass. Unlimited by default,
       # which finishes a long history in one go.
       # MUSIC_CURATOR_BACKFILL_LIMIT: "0"
+      # The MP3 mirror music-mirror maintains. Playlists are written into
+      # _playlists/ inside it; leave unset to skip them.
+      MUSIC_CURATOR_MIRROR: /mirror
+      # Most tracks in any one playlist.
+      # MUSIC_CURATOR_PLAYLIST_LIMIT: "100"
     volumes:
       - /mnt/tank/apps/music-curator:/data
+      - /mnt/tank/media/music-mp3:/mirror

music_curator.py

@@ -26,6 +26,7 @@ import re
 import signal
 import sqlite3
 import sys
+import tempfile
 import time
 import unicodedata
 import urllib.error
@@ -62,6 +63,16 @@ REQUEST_DELAY_SECONDS = 0.25
 BACKOFF_SECONDS = 2.0
 BACKOFF_CEILING_SECONDS = 60.0
 
+# What music-mirror names its output, and where playlists are written inside the
+# mirror. music-mirror's prune only deletes `*.mp3` and only removes directories
+# it finds empty, so a directory of M3Us survives it untouched.
+MIRROR_SUFFIX = ".mp3"
+PLAYLIST_DIRECTORY = "_playlists"
+
+# Group-readable, for the same reason music-mirror sets it: the mirror is read
+# back by whatever serves it, and a playlist nobody can read is not a playlist.
+GROUP_READ = 0o040
+
 SCHEMA_VERSION = "2"
 
 # Versions this build upgrades in place. Everything added since version 1 is a
@@ -1046,6 +1057,225 @@ def format_time(uts):
     return datetime.fromtimestamp(uts, tz=timezone.utc).strftime("%Y-%m-%d %H:%M")
 
 
+# Playlists written into the mirror. Each is a rule over the listening history
+# and the library index; none of them look at the audio.
+#
+# The rotating ones are shuffled by week rather than by pass. A pass runs every
+# few hours, and a playlist that reorders itself every time is one that has to
+# be re-imported every time -- the Music app does not track a file, it imports a
+# snapshot of one.
+ROTATION_PERIOD_SECONDS = 7 * 86400
+SHUFFLE_MULTIPLIER = 2654435761
+SHUFFLE_MODULUS = 104729
+
+# Common table expressions the rules share. `played` is every library track that
+# has ever been matched to a scrobble, with its count and the last time it was
+# heard; `recent` is the same restricted to a window.
+PLAYED_CTE = """
+WITH played AS (
+    SELECT sk.track_id AS track_id, COUNT(*) AS plays, MAX(s.uts) AS last_uts
+      FROM scrobble s
+      JOIN scrobble_key sk ON sk.artist = s.artist AND sk.track = s.track
+     WHERE sk.track_id IS NOT NULL
+     GROUP BY sk.track_id
+)
+"""
+RECENT_CTE = """
+WITH recent AS (
+    SELECT sk.track_id AS track_id, COUNT(*) AS plays
+      FROM scrobble s
+      JOIN scrobble_key sk ON sk.artist = s.artist AND sk.track = s.track
+     WHERE sk.track_id IS NOT NULL AND s.uts >= :year_ago
+     GROUP BY sk.track_id
+)
+"""
+SELECT_TRACK = """
+SELECT t.id AS id, a.name AS artist, t.title AS title,
+       t.duration AS duration, t.path AS path
+  FROM lidarr_track t
+  JOIN lidarr_artist a ON a.id = t.artist_id
+"""
+PLAYABLE = " t.has_file = 1 AND t.path IS NOT NULL AND t.path <> '' "
+SHUFFLE = f" ((t.id * {SHUFFLE_MULTIPLIER}) + :week) % {SHUFFLE_MODULUS} "
+
+PLAYLISTS = (
+    (
+        "heavy-rotation",
+        "most played over the last twelve months",
+        RECENT_CTE + SELECT_TRACK + f"""
+          JOIN recent r ON r.track_id = t.id
+         WHERE {PLAYABLE}
+         ORDER BY r.plays DESC, a.name, t.title
+         LIMIT :limit
+        """,
+    ),
+    (
+        "all-time",
+        "most played ever",
+        PLAYED_CTE + SELECT_TRACK + f"""
+          JOIN played p ON p.track_id = t.id
+         WHERE {PLAYABLE}
+         ORDER BY p.plays DESC, a.name, t.title
+         LIMIT :limit
+        """,
+    ),
+    (
+        "neglected",
+        "played heavily once, silent for twelve months",
+        PLAYED_CTE + SELECT_TRACK + f"""
+          JOIN played p ON p.track_id = t.id
+         WHERE {PLAYABLE} AND p.last_uts < :year_ago
+         ORDER BY p.plays DESC, a.name, t.title
+         LIMIT :limit
+        """,
+    ),
+    (
+        "deep-cuts",
+        "never played, from albums whose other tracks you play constantly",
+        PLAYED_CTE + """,
+        album_plays AS (
+            SELECT t.album_id AS album_id, SUM(p.plays) AS plays
+              FROM played p JOIN lidarr_track t ON t.id = p.track_id
+             GROUP BY t.album_id
+        )
+        """ + SELECT_TRACK + f"""
+          JOIN album_plays ap ON ap.album_id = t.album_id
+          LEFT JOIN played p ON p.track_id = t.id
+         WHERE {PLAYABLE} AND p.track_id IS NULL
+         ORDER BY ap.plays DESC, t.id
+         LIMIT :limit
+        """,
+    ),
+    (
+        "unheard-favourites",
+        "never played, by the artists you play most; rotates weekly",
+        PLAYED_CTE + """,
+        artist_plays AS (
+            SELECT t.artist_id AS artist_id, SUM(p.plays) AS plays
+              FROM played p JOIN lidarr_track t ON t.id = p.track_id
+             GROUP BY t.artist_id
+        )
+        """ + SELECT_TRACK + f"""
+          LEFT JOIN played p ON p.track_id = t.id
+         WHERE {PLAYABLE} AND p.track_id IS NULL
+           AND t.artist_id IN (SELECT artist_id FROM artist_plays
+                                ORDER BY plays DESC LIMIT 50)
+         ORDER BY {SHUFFLE}
+         LIMIT :limit
+        """,
+    ),
+    (
+        "unheard",
+        "never played, anywhere in the library; rotates weekly",
+        PLAYED_CTE + SELECT_TRACK + f"""
+          LEFT JOIN played p ON p.track_id = t.id
+         WHERE {PLAYABLE} AND p.track_id IS NULL
+         ORDER BY {SHUFFLE}
+         LIMIT :limit
+        """,
+    ),
+)
+
+
+def library_root_of(store):
+    """Return the directory Lidarr's artist folders sit under.
+
+    Derived rather than configured, because it has to agree with what Lidarr
+    reports and no one wants to keep a second copy of that in step by hand.
+    """
+    paths = [
+        row["path"]
+        for row in store.connection.execute(
+            "SELECT path FROM lidarr_artist WHERE path IS NOT NULL AND path <> ''"
+        )
+    ]
+    if not paths:
+        return None
+    if len(paths) == 1:
+        return str(Path(paths[0]).parent)
+    try:
+        return os.path.commonpath(paths)
+    except ValueError:
+        return None
+
+
+def mirror_path_for(source, library_root, mirror_root):
+    """Return where music-mirror would have put the MP3 for a source file."""
+    try:
+        relative = Path(source).relative_to(library_root)
+    except ValueError:
+        return None
+    return (Path(mirror_root) / relative).with_suffix(MIRROR_SUFFIX)
+
+
+def write_playlist(path, entries):
+    """Write one extended M3U, atomically.
+
+    Paths are relative to the playlist file, so the same playlist works from the
+    NAS, from a Mac over SMB and from Linux without rewriting.
+    """
+    lines = ["#EXTM3U"]
+    for entry in entries:
+        seconds = round((entry["duration"] or 0) / 1000)
+        lines.append(f"#EXTINF:{seconds},{entry['artist']} - {entry['title']}")
+        lines.append(os.path.relpath(entry["mirror"], path.parent))
+
+    path.parent.mkdir(parents=True, exist_ok=True)
+    handle, temporary = tempfile.mkstemp(dir=path.parent, suffix=".m3u.part")
+    os.close(handle)
+    temporary = Path(temporary)
+    try:
+        temporary.write_text("\n".join(lines) + "\n", encoding="utf-8")
+        # The mirror is read back by something else; see music-mirror, which had
+        # to learn this the hard way.
+        mode = temporary.stat().st_mode
+        if not mode & GROUP_READ:
+            temporary.chmod(mode | GROUP_READ)
+        os.replace(temporary, path)
+    finally:
+        temporary.unlink(missing_ok=True)
+
+
+def build_playlists(store, mirror_root, library_root, limit, now):
+    """Write every playlist into the mirror. Returns how many tracks were listed.
+
+    A track is only listed once its mirror file has been confirmed to exist. The
+    index knows what Lidarr holds, which is the lossless source; whether the MP3
+    beside it has been encoded yet is music-mirror's business and is checked
+    rather than assumed.
+    """
+    directory = Path(mirror_root) / PLAYLIST_DIRECTORY
+    parameters = {
+        "limit": limit,
+        "year_ago": now - 365 * 86400,
+        "week": now // ROTATION_PERIOD_SECONDS,
+    }
+
+    total = 0
+    missing = 0
+    for name, description, sql in PLAYLISTS:
+        entries = []
+        for row in store.connection.execute(sql, parameters):
+            mirror = mirror_path_for(row["path"], library_root, mirror_root)
+            if mirror is None or not mirror.is_file():
+                missing += 1
+                continue
+            entries.append({**dict(row), "mirror": mirror})
+        write_playlist(directory / f"{name}.m3u", entries)
+        total += len(entries)
+        logger.info("playlist %-20s %4d tracks -- %s", name, len(entries), description)
+
+    if missing:
+        logger.warning(
+            "%d selected tracks had no file in the mirror and were left out."
+            " A handful means music-mirror has not encoded them yet; a large"
+            " number means --library-root or --mirror is pointing at the wrong"
+            " place, since the paths are then being mapped to nothing.",
+            missing,
+        )
+    return total
+
+
 def coverage_report(store):
     """Log how much of the listening history could be tied to the library.
 
@@ -1262,7 +1492,10 @@ def report(store, now):
         logger.info("  top track 90d %2d: %-50s %d", position, label[:50], row["plays"])
 
 
-def run_once(client, store, user, now, backfill_limit, lidarr=None):
+def run_once(
+    client, store, user, now, backfill_limit, lidarr=None, mirror=None,
+    library_root=None, playlist_limit=100,
+):
     """Run a single pass. Returns the number of scrobbles added."""
     started = time.monotonic()
     added = catch_up(client, store, user, now)
@@ -1275,6 +1508,15 @@ def run_once(client, store, user, now, backfill_limit, lidarr=None):
     # arrived, and they need a verdict too.
     if store.scalar("SELECT COUNT(*) FROM lidarr_track"):
         match_library(store)
+        if mirror is not None:
+            root = library_root or library_root_of(store)
+            if root is None:
+                logger.warning(
+                    "cannot work out where Lidarr's music lives, so no playlists were"
+                    " written; set --library-root"
+                )
+            else:
+                build_playlists(store, mirror, root, playlist_limit, now)
 
     logger.info(
         "pass complete in %.1fs: %d scrobbles added, %d loved",
@@ -1347,6 +1589,25 @@ def build_parser():
         default=os.getenv("MUSIC_CURATOR_LIDARR_API_KEY"),
         help="Lidarr API key (env MUSIC_CURATOR_LIDARR_API_KEY)",
     )
+    parser.add_argument(
+        "--mirror",
+        default=os.getenv("MUSIC_CURATOR_MIRROR"),
+        help="root of the MP3 mirror; playlists are written into it"
+        " (env MUSIC_CURATOR_MIRROR)",
+    )
+    parser.add_argument(
+        "--library-root",
+        default=os.getenv("MUSIC_CURATOR_LIBRARY_ROOT"),
+        help="prefix to strip from Lidarr's track paths when mapping them into the"
+        " mirror; derived from the indexed artist folders when unset"
+        " (env MUSIC_CURATOR_LIBRARY_ROOT)",
+    )
+    parser.add_argument(
+        "--playlist-limit",
+        type=int,
+        default=int(os.getenv("MUSIC_CURATOR_PLAYLIST_LIMIT", "100")),
+        help="most tracks to put in any one playlist (env MUSIC_CURATOR_PLAYLIST_LIMIT)",
+    )
     parser.add_argument(
         "--skip-index",
         action="store_true",
@@ -1411,7 +1672,17 @@ def main(argv=None, clock=time.time):
         while True:
             now = int(clock())
             try:
-                run_once(client, store, args.user, now, args.backfill_limit, lidarr)
+                run_once(
+                    client,
+                    store,
+                    args.user,
+                    now,
+                    args.backfill_limit,
+                    lidarr,
+                    args.mirror,
+                    args.library_root,
+                    args.playlist_limit,
+                )
             except (LastfmError, LidarrError) as error:
                 logger.error("%s", error)
                 if interval is None:

tests/test_music_curator.py

@@ -1,5 +1,7 @@
 import json
+import stat
 import urllib.error
+from pathlib import Path
 
 import pytest
 from conftest import FakeLastfm, FakeLidarr, make_loved, make_tracks
@@ -825,3 +827,159 @@ def test_the_title_index_is_used_for_the_report_lookup(tmp_path):
         )
     )
     assert "lidarr_track_title" in plan, plan
+
+
+def playlist_library(tmp_path):
+    """A library on disk, with mirror MP3s beside the Lidarr source paths."""
+    source = tmp_path / "music"
+    mirror = tmp_path / "mirror"
+    library = [
+        {
+            "name": "Played Band",
+            "albums": [
+                {
+                    "title": "Known",
+                    "tracks": [{"title": "Hit"}, {"title": "Album Track"}],
+                }
+            ],
+        },
+        {
+            "name": "Silent Band",
+            "albums": [{"title": "Unknown", "tracks": [{"title": "Never Heard"}]}],
+        },
+    ]
+    api = FakeLidarr(library)
+    # FakeLidarr invents /music/<artist>/<album>/<title>.flac; put the mirror
+    # MP3s at the paths music-mirror would have produced from those.
+    for handle in api.files:
+        relative = Path(handle["path"]).relative_to("/music")
+        handle["path"] = str(source / relative)
+        mp3 = (mirror / relative).with_suffix(".mp3")
+        mp3.parent.mkdir(parents=True, exist_ok=True)
+        mp3.write_bytes(b"not really an mp3")
+    for artist in api.artists:
+        artist["path"] = str(source / Path(artist["path"]).name)
+    return api, source, mirror
+
+
+def test_playlists_are_written_into_the_mirror(tmp_path):
+    api, source, mirror = playlist_library(tmp_path)
+    store = store_at(tmp_path)
+    ingest(store, FakeLastfm([scrobble_of("Played Band", "Hit")] * 1))
+    music_curator.index_library(
+        music_curator.Lidarr("http://lidarr", "key", transport=api), store
+    )
+    music_curator.match_library(store)
+
+    music_curator.build_playlists(store, mirror, str(source), 100, NOW)
+
+    written = sorted(p.name for p in (mirror / "_playlists").glob("*.m3u"))
+    assert written == [
+        "all-time.m3u",
+        "deep-cuts.m3u",
+        "heavy-rotation.m3u",
+        "neglected.m3u",
+        "unheard-favourites.m3u",
+        "unheard.m3u",
+    ]
+
+    played = (mirror / "_playlists" / "all-time.m3u").read_text().splitlines()
+    assert played[0] == "#EXTM3U"
+    assert played[1].startswith("#EXTINF:")
+    assert "Played Band - Hit" in played[1]
+    # Relative to the playlist file, so the same file works from any mount.
+    assert played[2] == "../Played Band/Known/Hit.mp3"
+    assert (mirror / "_playlists" / "all-time.m3u").parent.joinpath(played[2]).resolve().is_file()
+
+
+def test_an_unplayed_track_lands_in_the_unheard_playlist(tmp_path):
+    api, source, mirror = playlist_library(tmp_path)
+    store = store_at(tmp_path)
+    ingest(store, FakeLastfm([scrobble_of("Played Band", "Hit")]))
+    music_curator.index_library(
+        music_curator.Lidarr("http://lidarr", "key", transport=api), store
+    )
+    music_curator.match_library(store)
+
+    music_curator.build_playlists(store, mirror, str(source), 100, NOW)
+
+    unheard = (mirror / "_playlists" / "unheard.m3u").read_text()
+    assert "Never Heard" in unheard
+    assert "Album Track" in unheard
+    # The one thing that was played must not be in it.
+    assert "Played Band - Hit\n" not in unheard
+
+
+def test_a_track_with_no_mirror_file_is_left_out(tmp_path):
+    """The index knows what Lidarr holds; whether music-mirror has encoded the
+    MP3 yet is a different question, and is checked rather than assumed."""
+    api, source, mirror = playlist_library(tmp_path)
+    for mp3 in mirror.rglob("*.mp3"):
+        mp3.unlink()
+    store = store_at(tmp_path)
+    ingest(store, FakeLastfm([scrobble_of("Played Band", "Hit")]))
+    music_curator.index_library(
+        music_curator.Lidarr("http://lidarr", "key", transport=api), store
+    )
+    music_curator.match_library(store)
+
+    total = music_curator.build_playlists(store, mirror, str(source), 100, NOW)
+
+    assert total == 0
+    assert (mirror / "_playlists" / "all-time.m3u").read_text() == "#EXTM3U\n"
+
+
+def test_the_library_root_is_derived_from_the_artist_folders(tmp_path):
+    api, source, mirror = playlist_library(tmp_path)
+    store = store_at(tmp_path)
+    music_curator.index_library(
+        music_curator.Lidarr("http://lidarr", "key", transport=api), store
+    )
+
+    assert music_curator.library_root_of(store) == str(source)
+
+
+def test_the_playlist_limit_is_honoured(tmp_path):
+    api, source, mirror = playlist_library(tmp_path)
+    store = store_at(tmp_path)
+    music_curator.index_library(
+        music_curator.Lidarr("http://lidarr", "key", transport=api), store
+    )
+    music_curator.match_library(store)
+
+    music_curator.build_playlists(store, mirror, str(source), 1, NOW)
+
+    unheard = (mirror / "_playlists" / "unheard.m3u").read_text().splitlines()
+    assert len([line for line in unheard if line.startswith("#EXTINF")]) == 1
+
+
+def test_the_rotation_moves_weekly_not_every_pass(tmp_path):
+    """A playlist that reorders on every pass is one that has to be re-imported
+    on every pass; the Music app imports a snapshot, it does not track a file."""
+    api, source, mirror = playlist_library(tmp_path)
+    store = store_at(tmp_path)
+    music_curator.index_library(
+        music_curator.Lidarr("http://lidarr", "key", transport=api), store
+    )
+    music_curator.match_library(store)
+
+    def unheard_at(when):
+        music_curator.build_playlists(store, mirror, str(source), 100, when)
+        return (mirror / "_playlists" / "unheard.m3u").read_text()
+
+    same_week = unheard_at(NOW), unheard_at(NOW + 3600)
+    assert same_week[0] == same_week[1]
+
+
+def test_playlists_are_group_readable(tmp_path):
+    api, source, mirror = playlist_library(tmp_path)
+    store = store_at(tmp_path)
+    music_curator.index_library(
+        music_curator.Lidarr("http://lidarr", "key", transport=api), store
+    )
+    music_curator.match_library(store)
+
+    music_curator.build_playlists(store, mirror, str(source), 100, NOW)
+
+    for playlist in (mirror / "_playlists").glob("*.m3u"):
+        assert playlist.stat().st_mode & stat.S_IRGRP, playlist