Repository navigation
Batch independent memory saves for faster imports - #4
Merged
Merged
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Legacy imports currently need a separate embedding operation and database transaction for every saved fact, making the first memory tool call slow. Add
rememberManyfor 1–20 independent saves: preserve each fact's source session and retry identity, batch embedding, and commit new records atomically with the existing revision guard.The SDK, HTTP client and Worker expose the same method. Built-in stores accept processed-ID batches across sessions; custom Store implementations must handle the documented list form before using it. Corrections retain the single-save API. Prepare the sole public package
@fluiddb/fluiddb@1.0.0-next.5.Validation:
bun run check,bun run test:runtime,bun run verify:packages,bun run release:check,./ref-check, all 65 real SQLite/Postgres/Firestore adapter checks, and Firestore REST in workerd passed. Regression coverage includes mixed retries, forgotten IDs, provider failure, cancellation/erasure, HTTP authorization and persistence across Worker restart. No paid model calls.The companion backend change reduced a synthetic first import page from 35.54 s to about 3 s at fixed artificial network delays; this is not a production measurement. The OpenAI adapter may split the batched invocation into concurrent HTTP requests. Independent correctness, security and API reviews completed; documented provider-call wording was corrected.