mirror of
https://github.com/lancedb/lancedb.git
synced 2026-07-08 21:40:41 +00:00
0abf641733
### Description `db://`-style connections that use the lance-namespace path (`LanceNamespaceDatabase` → `NativeTable` + the lance-namespace REST client) never sent a read-freshness signal. Against a server configured to serve cached table metadata up to some staleness window, this allows stale-read-after-write across handles and processes. The remote table path already solved this (#3439). This brings the namespace path to parity. The namespace REST client doesn't let callers attach headers directly, but it forwards a `DynamicContextProvider`'s `headers.*` context entries as HTTP headers per request. So: - A shared per-table baseline map is created before the namespace client. I built and installed on the `ConnectBuilder` via a context provider. - On read operations the provider emits ·x-lancedb-min-timestamp = max(baseline, now − read_consistency_interval)` (RFC3339), keyed by the operation's `object_id`. - Each table handle bumps its baseline (monotonically) on `checkout_latest()`, `restore()`, and every data/schema write. `checkout_latest()` is the primary hook: consumers refresh a handle there after writing elsewhere, then poll. Read operations that carry the floor: `describe_table`, `list_table_versions`, `query_table`, `list_tables`. `list_table_versions` is what resolves "latest" for managed-versioning tables (`get_latest_version`), so it's the op that makes `checkout_latest()` actually observe a prior write. `describe_table_version` is excluded (pinned to an immutable version). This mirrors #3439 (timestamp baseline, `max(baseline, now − interval)`, monotonic); no `min_version` and no body channel, since the namespace path has no version-returning write responses. ### Testing - Unit tests for `compute_min_timestamp` / `next_freshness_baseline` and the provider (header at/after a bumped baseline; nothing for an empty baseline + no interval; interval floor applies; non-read ops emit nothing; `list_tables` uses only the interval floor). - Verified end-to-end against a local server that honors the header: reads carry `x-lancedb-min-timestamp`, writes don't, and read-your-write holds.
The Multimodal AI Lakehouse
How to Install ✦ Detailed Documentation ✦ Tutorials and Recipes ✦ Contributors
The ultimate multimodal data platform for AI/ML applications.
LanceDB is designed for fast, scalable, and production-ready vector search. It is built on top of the Lance columnar format. You can store, index, and search over petabytes of multimodal data and vectors with ease. LanceDB is a central location where developers can build, train and analyze their AI workloads.
Demo: Multimodal Search by Keyword, Vector or with SQL
Star LanceDB to get updates!
⭐ Click here ⭐ to see how fast we're growing!
Key Features:
- Fast Vector Search: Search billions of vectors in milliseconds with state-of-the-art indexing.
- Comprehensive Search: Support for vector similarity search, full-text search and SQL.
- Multimodal Support: Store, query and filter vectors, metadata and multimodal data (text, images, videos, point clouds, and more).
- Advanced Features: Zero-copy, automatic versioning, manage versions of your data without needing extra infrastructure. GPU support in building vector index.
Products:
- Open Source & Local: 100% open source, runs locally or in your cloud. No vendor lock-in.
- Cloud and Enterprise: Production-scale vector search with no servers to manage. Complete data sovereignty and security.
Ecosystem:
- Columnar Storage: Built on the Lance columnar format for efficient storage and analytics.
- Seamless Integration: Python, Node.js, Rust, and REST APIs for easy integration. Native Python and Javascript/Typescript support.
- Rich Ecosystem: Integrations with LangChain 🦜️🔗, LlamaIndex 🦙, Apache-Arrow, Pandas, Polars, DuckDB and more on the way.
How to Install:
Follow the Quickstart doc to set up LanceDB locally.
API & SDK: We also support Python, Typescript and Rust SDKs
| Interface | Documentation |
|---|---|
| Python SDK | https://lancedb.github.io/lancedb/python/python/ |
| Typescript SDK | https://lancedb.github.io/lancedb/js/globals/ |
| Rust SDK | https://docs.rs/lancedb/latest/lancedb/index.html |
| REST API | https://docs.lancedb.com/api-reference/rest |
Join Us and Contribute
We welcome contributions from everyone! Whether you're a developer, researcher, or just someone who wants to help out.
If you have any suggestions or feature requests, please feel free to open an issue on GitHub or discuss it on our Discord server.
Check out the GitHub Issues if you would like to work on the features that are planned for the future. If you have any suggestions or feature requests, please feel free to open an issue on GitHub.
Contributors
Stay in Touch With Us