#867: Tidal playlist discovery shows all tracks (was capped to ~21)

Two issues in the same path:
1. The shared discovery modal pre-renders one row per track from a
   separately-fetched frontend track list, then the poll dropped any backend
   result without a pre-rendered row (if (!row) return). When the frontend's
   track fetch came back rate-limited/partial (~21) while discovery's own fetch
   got all 59, the surplus results vanished. Now the modal CREATES a row for any
   result lacking one, so authoritative backend results drive the list (fixes
   all sources sharing the modal).
2. get_playlist hydrated a whole relationships page in one _get_tracks_batch
   call, but Tidal caps filter[id] at 20/request, silently truncating larger
   pages. Chunk to the cap like get_album_tracks already does.

Seam + regression tests (tests/test_tidal_playlist_batch_chunking.py).

core/tidal_client.py

@@ -1324,13 +1324,20 @@ class TidalClient:
                         track_ids.append(item.get("id"))
 
                 if track_ids:
-                    # Batch fetch full track details with artists and albums
-                    try:
-                        batch_tracks = self._get_tracks_batch(track_ids)
-                    except Exception as e:
-                        logger.error(f"Error fetching track details for page {page_num}: {e}")
-                        # Continue pagination — we lose this batch but can still get remaining
-                        batch_tracks = []
+                    # Batch fetch full track details with artists and albums.
+                    # Chunk to the filter[id] page cap — Tidal returns at most
+                    # _COLLECTION_BATCH_SIZE tracks per request, so a relationships
+                    # page larger than the cap would be silently truncated if sent
+                    # in one shot. Mirrors get_album_tracks. (#867)
+                    batch_tracks = []
+                    for j in range(0, len(track_ids), self._COLLECTION_BATCH_SIZE):
+                        chunk_ids = track_ids[j:j + self._COLLECTION_BATCH_SIZE]
+                        try:
+                            batch_tracks.extend(self._get_tracks_batch(chunk_ids))
+                        except Exception as e:
+                            logger.error(f"Error fetching track details for page {page_num}: {e}")
+                            # Lose this chunk but keep going — remaining chunks/pages can still load
+                            continue
 
                     if len(batch_tracks) < len(track_ids):
                         logger.warning(f"Page {page_num}: requested {len(track_ids)} tracks but only {len(batch_tracks)} returned (some may be unavailable in your region)")

tests/test_tidal_playlist_batch_chunking.py

@@ -0,0 +1,107 @@
+"""#867 regression: Tidal playlist track hydration must chunk to the API cap.
+
+Tidal's ``/tracks?filter[id]=...`` endpoint returns at most ``_COLLECTION_BATCH_SIZE``
+(20) tracks per request. ``get_playlist`` fetches a page of track-ID links and then
+hydrates them via ``_get_tracks_batch``. If it sends a relationships page with more
+than 20 IDs in a single hydration call, the surplus is silently dropped — a 59-track
+playlist would surface as ~20. ``get_album_tracks`` already chunks; this pins the same
+behavior for ``get_playlist`` so discovery sees every track.
+"""
+
+from __future__ import annotations
+
+from core.tidal_client import TidalClient, Track
+
+
+class _FakeResp:
+    status_code = 200
+
+    def raise_for_status(self):
+        return None
+
+    def json(self):
+        return {
+            "data": {
+                "id": "PL1",
+                "attributes": {"name": "My Playlist", "accessType": "PRIVATE"},
+                "relationships": {},
+            }
+        }
+
+
+class _FakeSession:
+    def get(self, *args, **kwargs):
+        return _FakeResp()
+
+
+def _make_client():
+    """Build a TidalClient without running its network/config __init__."""
+    client = object.__new__(TidalClient)
+    client.base_url = "https://api.tidal.test"
+    client.session = _FakeSession()
+    return client
+
+
+def test_get_playlist_chunks_oversized_relationships_page(monkeypatch):
+    client = _make_client()
+    cap = TidalClient._COLLECTION_BATCH_SIZE  # 20
+    n = 59
+    ids = [str(i) for i in range(n)]
+
+    monkeypatch.setattr(client, "_ensure_valid_token", lambda: True)
+    # One relationships page returns ALL 59 ID links at once, then no cursor.
+    monkeypatch.setattr(
+        client,
+        "_get_playlist_tracks_page",
+        lambda playlist_id, cursor=None: {
+            "data": [{"type": "tracks", "id": i} for i in ids],
+            "links": {"meta": {}},  # no nextCursor -> single page
+        },
+    )
+
+    seen_chunk_sizes = []
+
+    def fake_batch(chunk_ids):
+        # Simulate the real filter[id] cap: never return more than the cap, so a
+        # single oversized call would lose the surplus (the bug being guarded).
+        seen_chunk_sizes.append(len(chunk_ids))
+        capped = chunk_ids[:cap]
+        return [Track(id=i, name=f"t{i}", artists=["a"]) for i in capped]
+
+    monkeypatch.setattr(client, "_get_tracks_batch", fake_batch)
+
+    playlist = client.get_playlist("PL1")
+
+    assert playlist is not None
+    # The whole point: all 59 hydrate, not just the first 20.
+    assert len(playlist.tracks) == n
+    # Every hydration call stayed within the API cap (so none truncated).
+    assert seen_chunk_sizes and max(seen_chunk_sizes) <= cap
+
+
+def test_get_playlist_small_page_single_call(monkeypatch):
+    """A page at/under the cap still hydrates in one call (no behavior change)."""
+    client = _make_client()
+    ids = [str(i) for i in range(5)]
+
+    monkeypatch.setattr(client, "_ensure_valid_token", lambda: True)
+    monkeypatch.setattr(
+        client,
+        "_get_playlist_tracks_page",
+        lambda playlist_id, cursor=None: {
+            "data": [{"type": "tracks", "id": i} for i in ids],
+            "links": {"meta": {}},
+        },
+    )
+
+    calls = []
+
+    def fake_batch(chunk_ids):
+        calls.append(list(chunk_ids))
+        return [Track(id=i, name=f"t{i}", artists=["a"]) for i in chunk_ids]
+
+    monkeypatch.setattr(client, "_get_tracks_batch", fake_batch)
+
+    playlist = client.get_playlist("PL1")
+    assert len(playlist.tracks) == 5
+    assert len(calls) == 1  # one chunk, one hydration call

webui/static/sync-services.js

@@ -10186,8 +10186,30 @@ function updateYouTubeDiscoveryModal(urlHash, status) {
 
     // Update table rows
     status.results.forEach(result => {
-        const row = document.getElementById(`discovery-row-${urlHash}-${result.index}`);
-        if (!row) return;
+        let row = document.getElementById(`discovery-row-${urlHash}-${result.index}`);
+        if (!row) {
+            // #867: the initial rows are pre-rendered from a separately-fetched
+            // track list (state.playlist.tracks) that can be SHORTER than the
+            // backend's authoritative discovery results — e.g. a Tidal playlist
+            // whose discovery fetched 59 tracks while the modal's own track fetch
+            // returned a rate-limited partial (~21). The old `if (!row) return`
+            // silently dropped every result past the pre-rendered rows. Create
+            // the missing row instead so the authoritative results drive the
+            // list and no discovered track disappears. The existing cell-fill
+            // logic below then populates it like any other row.
+            const trackName = result.yt_track || result.lb_track || result.track_name || '-';
+            row = document.createElement('tr');
+            row.id = `discovery-row-${urlHash}-${result.index}`;
+            row.innerHTML =
+                `<td class="yt-track">${trackName}</td>` +
+                '<td class="yt-artist"></td>' +
+                '<td class="discovery-status"></td>' +
+                '<td class="spotify-track">-</td>' +
+                '<td class="spotify-artist">-</td>' +
+                '<td class="spotify-album">-</td>' +
+                '<td class="discovery-actions">-</td>';
+            tableBody.appendChild(row);
+        }
 
         const statusCell = row.querySelector('.discovery-status');
         const spotifyTrackCell = row.querySelector('.spotify-track');