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Outdated suggestions cannot be applied.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.
DO NOT MERGE
This PR explores adding LanceDB as an embedded dependency. The goal is to be able to embed files and specific functions in a vector db which comes in handy for enforcing all sorts of rules, the most obvious of which is in detecting near duplicates for DRY-type rules.
We don't really need a vector database for this, but 1) it would be nice for the embeddings to be cached in-between runs and 2) we'd like to support larger codebases and more experimentation without worrying about the embeddings / vector search, so I'd prefer to use a production-grade solution.
Overall, I love that
lancedb
can be embedded locally or self-hosted or hosted in their cloud. BUT their JS/TS SDK is a complete mess, and I'm really not sure it's worth the trouble.I'm also considering Chroma and Qdrant, both of which are OSS and self-hostable. If we go with one of these non-embedded solutions, then I'd prefer to make it an optional dependency.
DO NOT MERGE