fix(updates): chunk bulk queries to avoid SQLite variable limit (#914)

_split _get_records_bulk into 500-id batches so the WHERE IN clause
never exceeds SQLite's 999-parameter ceiling.

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>

py/services/model_update_service.py

@@ -209,6 +209,8 @@ class ModelUpdateRecord:
 class ModelUpdateService:
     """Persist and query remote model version metadata."""
 
+    _SQLITE_MAX_VARIABLES = 500
+
     _SCHEMA = """
         PRAGMA foreign_keys = ON;
         CREATE TABLE IF NOT EXISTS model_update_status (
@@ -1439,33 +1441,41 @@ class ModelUpdateService:
         if not model_ids:
             return {}
 
-        params = tuple(model_ids)
-        placeholders = ",".join("?" for _ in params)
+        ids = list(model_ids)
+        status_rows: list = []
+        version_rows: list = []
 
         with self._connect() as conn:
-            status_rows = conn.execute(
-                f"""
-                SELECT model_id, model_type, last_checked_at, should_ignore_model
-                FROM model_update_status
-                WHERE model_id IN ({placeholders})
-                """,
-                params,
-            ).fetchall()
+            for start in range(0, len(ids), self._SQLITE_MAX_VARIABLES):
+                chunk = tuple(ids[start : start + self._SQLITE_MAX_VARIABLES])
+                placeholders = ",".join("?" for _ in chunk)
+
+                chunk_status = conn.execute(
+                    f"""
+                    SELECT model_id, model_type, last_checked_at, should_ignore_model
+                    FROM model_update_status
+                    WHERE model_id IN ({placeholders})
+                    """,
+                    chunk,
+                ).fetchall()
+                status_rows.extend(chunk_status)
+
+                chunk_versions = conn.execute(
+                    f"""
+                    SELECT model_id, version_id, sort_index, name, base_model, released_at,
+                           size_bytes, preview_url, is_in_library, should_ignore, early_access_ends_at,
+                           is_early_access
+                    FROM model_update_versions
+                    WHERE model_id IN ({placeholders})
+                    ORDER BY model_id ASC, sort_index ASC, version_id ASC
+                    """,
+                    chunk,
+                ).fetchall()
+                version_rows.extend(chunk_versions)
+
             if not status_rows:
                 return {}
 
-            version_rows = conn.execute(
-                f"""
-                SELECT model_id, version_id, sort_index, name, base_model, released_at,
-                       size_bytes, preview_url, is_in_library, should_ignore, early_access_ends_at,
-                       is_early_access
-                FROM model_update_versions
-                WHERE model_id IN ({placeholders})
-                ORDER BY model_id ASC, sort_index ASC, version_id ASC
-                """,
-                params,
-            ).fetchall()
-
         versions_by_model: Dict[int, List[ModelVersionRecord]] = {}
         for row in version_rows:
             model_id = int(row["model_id"])

tests/services/test_model_update_service.py

@@ -442,6 +442,42 @@ async def test_has_updates_bulk_returns_mapping(tmp_path):
     assert await service.has_update("lora", 9) is True
 
 
+@pytest.mark.asyncio
+async def test_has_updates_bulk_handles_more_than_sqlite_max_variables(tmp_path):
+    """Bulk query with >999 model IDs must not raise 'too many SQL variables'."""
+    db_path = tmp_path / "updates.sqlite"
+    service = ModelUpdateService(str(db_path), ttl_seconds=3600)
+
+    model_ids = list(range(1, 1201))
+    with sqlite3.connect(str(db_path)) as conn:
+        conn.execute("INSERT INTO model_update_status (model_id, model_type) VALUES (?, ?)", (1, "lora"))
+        conn.execute("INSERT INTO model_update_versions (model_id, version_id, sort_index, name) VALUES (?, ?, ?, ?)", (1, 10, 0, "v1"))
+
+    mapping = await service.has_updates_bulk("lora", model_ids)
+
+    assert mapping[1] is True
+    assert len(mapping) == len(model_ids)
+    assert all(v is False for k, v in mapping.items() if k != 1)
+
+
+@pytest.mark.asyncio
+async def test_get_records_bulk_handles_more_than_sqlite_max_variables(tmp_path):
+    """Bulk record fetch with >999 model IDs must not raise 'too many SQL variables'."""
+    db_path = tmp_path / "updates.sqlite"
+    service = ModelUpdateService(str(db_path), ttl_seconds=3600)
+
+    model_ids = list(range(1, 1201))
+    with sqlite3.connect(str(db_path)) as conn:
+        conn.execute("INSERT INTO model_update_status (model_id, model_type) VALUES (?, ?)", (1, "lora"))
+        conn.execute("INSERT INTO model_update_versions (model_id, version_id, sort_index, name) VALUES (?, ?, ?, ?)", (1, 10, 0, "v1"))
+
+    records = await service.get_records_bulk("lora", model_ids)
+
+    assert 1 in records
+    assert records[1].model_id == 1
+    assert len(records) == 1
+
+
 @pytest.mark.asyncio
 async def test_refresh_allows_duplicate_version_ids_across_models(tmp_path):
     db_path = tmp_path / "updates.sqlite"