Skip to main content
For the complete documentation index for agents and LLMs, see llms.txt.

SupabasePgvectorKeywordRetriever

Retrieve documents from a SupabasePgvectorDocumentStore using PostgreSQL full-text search. This component is a thin wrapper around PgvectorKeywordRetriever adapted for use with Supabase.

Key Features​

  • Performs keyword-based full-text search using PostgreSQL's ts_rank_cd function.
  • Ranks results based on term frequency, term proximity, and document section importance.
  • Reads the connection string from the SUPABASE_DB_URL environment variable by default.
  • Works without embeddings, making it suitable for pipelines that don't use dense retrieval.
  • Configurable filter policy to merge or replace filters at query time.

Configuration​

  1. Drag the SupabasePgvectorKeywordRetriever component onto the canvas from the Component Library.
  2. Click on the component to open the configuration panel.
  3. Configure the nested SupabasePgvectorDocumentStore:
    • Set connection_string as a secret called SUPABASE_DB_URL. The connection string format is postgresql://postgres.[project-ref]:[password]@aws-0-[region].pooler.supabase.com:5432/postgres. For instructions, see Add Secrets.
    • Use session mode (port 5432) or a direct connection for best compatibility.

Connections​

SupabasePgvectorKeywordRetriever receives a query string at runtime. It outputs a list of Document objects that you can connect to a PromptBuilder or other downstream components.

Source Code​

To check this component's source code, open keyword_retriever.py in the Haystack Core Integrations repository.

Usage Examples​

Basic Configuration​

  SupabasePgvectorKeywordRetriever:
    type: haystack_integrations.components.retrievers.supabase.keyword_retriever.SupabasePgvectorKeywordRetriever
    init_parameters:
      document_store:
        type: haystack_integrations.document_stores.supabase.document_store.SupabasePgvectorDocumentStore
        init_parameters:
          connection_string:
            type: env_var
            env_vars:
            - SUPABASE_DB_URL
            strict: false
          embedding_dimension: 768
      top_k: 10

Using the Component in a Pipeline​

# haystack-pipeline
components:
  document_store:
    type: haystack_integrations.document_stores.supabase.document_store.SupabasePgvectorDocumentStore
    init_parameters:
      connection_string:
        type: env_var
        env_vars:
        - SUPABASE_DB_URL
        strict: false
      embedding_dimension: 768

  retriever:
    type: haystack_integrations.components.retrievers.supabase.keyword_retriever.SupabasePgvectorKeywordRetriever
    init_parameters:
      document_store: document_store
      top_k: 10

connections: []

inputs:
  query:
  - retriever.query

outputs:
  documents: retriever.documents

Parameters​

Inputs​

ParameterTypeDescription
querystrThe keyword query string to search for.
filtersOptional[Dict[str, Any]]Filters to apply at query time.
top_kOptional[int]Maximum number of documents to retrieve. Overrides the init-time value.

Outputs​

ParameterTypeDescription
documentsList[Document]A list of matching documents ranked by relevance.

Init Parameters​

These are the parameters you can configure in Pipeline Builder:

ParameterTypeDefaultDescription
document_storeSupabasePgvectorDocumentStoreThe Supabase pgvector document store to retrieve documents from.
filtersOptional[Dict[str, Any]]NoneDefault filters applied to all searches.
top_kint10Maximum number of documents to return.
filter_policyFilterPolicyFilterPolicy.REPLACEHow to handle filters passed at query time. REPLACE replaces init-time filters; MERGE combines them.

Run Method Parameters​

These are the parameters you can configure for the component's run() method. This means you can pass these parameters at query time through the API, in Playground, or when running a job. For details, see Modify Pipeline Parameters at Query Time.

ParameterTypeDefaultDescription
querystrThe keyword query string.
filtersOptional[Dict[str, Any]]NoneRuntime filters to apply.
top_kOptional[int]NoneMaximum number of documents to retrieve. Overrides the init-time value.