Full deployment history for this project.
Merge pull request #21 from mbarbine/bolt/optimize-eventstore-index-16476925311189440537
⚡ Bolt: Index InMemoryEventStore events by aggregateId for O(1) lookups
Merge pull request #20 from mbarbine/bolt-optimize-interest-filtering-5027031729965625416
⚡ Bolt: Optimize tick loop interest filtering and entity iteration Avoid allocating intermediate `allEntities` arrays every tick and `relevantEntities` filter arrays/closures per connection by directly iterating `entities.values()` in single-pass replication loops.
Merge pull request #18 from mbarbine/perf/zero-alloc-binary-codec-6381673566413095821
⚡ Bolt: Fix binaryCodec count scoping and remove intermediate array allocations Fix missing count variable declarations in Snapshot and EntityDelta binary serialization loops. Remove intermediate Uint8Array array allocations during payload length sizing, leveraging fast getStringByteLength and TextEncoder.prototype.encodeInto() direct buffer write.
⚡ Bolt: Eliminate intermediate allocations during binary message encoding Fixed a reference bug where `count` was undefined and removed redundant `TextEncoder.encode()` calls and array allocations during message size calculations in `Snapshot` and `EntityDelta` encoders. Now uses `getStringByteLength()` and `encodeInto()` directly into output buffers without allocating intermediate Uint8Array instances.
⚡ Bolt: Optimize binaryCodec string length calculation & array allocations