Interface MultiLevelCachingStoreOptions

L1 memory cache size in bytes (default: 100MB)

L2 OPFS cache size in bytes (default: 2GB)

Enable debug logging (default: false)

Disable both cache tiers (e.g. driven by ?noCache). Default false.

Skip the L2 OPFS tier entirely (e.g. driven by ?noOpfs); L1 stays on. The deterministic sibling of the OPFSStore circuit breaker for environments whose OPFS is known to stall. Default false.

Clear caches on init (e.g. driven by ?clearCache). Default false.

Background L2 write-queue concurrency cap. Defaults to config.cache.opfsWriteConcurrency. Exposed mainly so tests can inject a tiny cap.

Background L2 write-queue max pending depth. Defaults to config.cache.opfsWriteQueueMax.

Max bytes retained by pending L2 writes. Defaults to computeOpfsWriteQueueBudgetBytes — the non-cache heap remainder's share, no longer derived from the L1 budget because the pending buffer IS the L1 entry's buffer (see the enqueue site in getResult), so an L1-resident pending write costs a reference rather than a second copy. Pass a real budget when overriding: the queue clamps a non-positive or non-finite value to a 1-BYTE cap, i.e. every write drops — 0 does not mean "unbounded".