langchain.js
    Preparing search index...

    Class for performing hybrid search operations on a Supabase database. It extends the BaseRetriever class and implements methods for similarity search, keyword search, and hybrid search.

    Hierarchy (View Summary)

    Index

    Constructors

    Properties

    client: SupabaseClient
    embeddings: EmbeddingsInterface
    keywordK: number
    keywordQueryName: string
    lc_namespace: string[] = ...
    query: string
    similarityK: number
    similarityQueryName: string
    tableName: string

    Methods

    • Parameters

      • query: string
      • OptionalrunManager: any

      Returns Promise<Document[]>

    • Combines the results of the similaritySearch and keywordSearch methods and returns the top 'k' documents based on a combination of similarity and keyword matching.

      Parameters

      • query: string

        The query to use for the hybrid search.

      • similarityK: number

        The number of top similar documents to return.

      • keywordK: number

        The number of top documents to return that match the keywords.

      • Optionalcallbacks: any

        Optional callbacks to pass to the similaritySearch method.

      Returns Promise<SearchResult[]>

      A promise that resolves to an array of search results. Each result is a tuple containing a Document, its combined score, and its ID.

    • Performs a keyword search on the Supabase database using the provided query and returns the top 'k' documents that match the keywords.

      Parameters

      • query: string

        The query to use for the keyword search.

      • k: number

        The number of top documents to return that match the keywords.

      Returns Promise<SearchResult[]>

      A promise that resolves to an array of search results. Each result is a tuple containing a Document, its similarity score multiplied by 10, and its ID.

    • Performs a similarity search on the Supabase database using the provided query and returns the top 'k' similar documents.

      Parameters

      • query: string

        The query to use for the similarity search.

      • k: number

        The number of top similar documents to return.

      • Optional_callbacks: any

        Optional callbacks to pass to the embedQuery method.

      Returns Promise<SearchResult[]>

      A promise that resolves to an array of search results. Each result is a tuple containing a Document, its similarity score, and its ID.

    • Returns string