Changelog

All notable changes to this project are documented in this file.

The format is based on Keep a Changelog, and this project adheres to Semantic Versioning.

[Unreleased]

The paced consume engine (KeyValue.watchPaced / watchAllPaced, StreamContext.createOrderedPacedConsumer / getPacedConsumerContext) mistook the server's routine expiry of an idle pull for an early pull terminus: the local window deadline (armed at publish) always fired before the server's expiresIn clock (started at receipt), so the server's 408 landed at the head of the next window and tripped the 500 ms early-terminus guard — a non-draining sleep and a redundant second pull per idle window. Statuses are now matched to the pull they answer via jnats's per-pull reply subjects, and stale ones are skipped. No API or configuration change; the callback engine is unaffected.

Pure KV key validation on the KeyValue companion object: validateKey (write/read keys) and validateKeyWildcardAllowed (watch/keys filter patterns), returning Either[InvalidKeyError, Unit] with the offending key and the jnats reason. Both delegate to the jnats validators (KV key grammar plus strict subject grammar), so callers can validate keys before a write instead of catching and sniffing IllegalArgumentException from put.

KeyValue.keysDetailed(...) (three overloads mirroring keys) returning KeysResult(keys, warmup), which exposes the Warmup.Result of the underlying warmup. This lets callers detect when a key listing was cut short by the warmup timeout (Warmup.Result.Timeout) versus completed fully (Warmup.Result.Success).

KeyValue.keys(...) could silently return a partial key set when warmup hit its timeout before all pending messages were drained, with no signal to the caller — a silent-data-loss hazard for any list-style operation built on keys. Fixed additively (non-breaking): keys behaviour is unchanged and it now delegates to keysDetailed; callers that need completeness guarantees should switch to keysDetailed and inspect warmup (e.g. raise, retry with a longer timeout, or return partial-with-a-flag). No existing caller breaks. See Option A in TASK.md.