Supabase Integration Question

[jameswong3388]

Hi, I was looking at how to integrate supabase with langchain, then I found this.

https://js.langchain.com/docs/modules/indexes/vector_stores/integrations/supabase

-- Enable the pgvector extension to work with embedding vectors
create extension vector;

-- Create a table to store your documents
create table documents (
  id bigserial primary key,
  content text, -- corresponds to Document.pageContent
  metadata jsonb, -- corresponds to Document.metadata
  embedding vector(1536) -- 1536 works for OpenAI embeddings, change if needed
);

-- Create a function to search for documents
create function match_documents (
  query_embedding vector(1536),
  match_count int
) returns table (
  id bigint,
  content text,
  metadata jsonb,
  similarity float
)
language plpgsql
as $$
#variable_conflict use_column
begin
  return query
  select
    id,
    content,
    metadata,
    1 - (documents.embedding <=> query_embedding) as similarity
  from documents
  order by documents.embedding <=> query_embedding
  limit match_count;
end;
$$;

-- Create an index to be used by the search function
create index on documents
  using ivfflat (embedding vector_cosine_ops)
  with (lists = 100);

Do i must to follow this ? The following is my table to store the embeddings, and function to do the vector search.
All this SQL scripts can be found here

https://github.com/supabase-community/nextjs-openai-doc-search/blob/main/supabase/migrations/20230406025118_init.sql

create table "public"."page_section" (
  id bigserial primary key,
  page_id bigint not null references public.page on delete cascade,
  content text,
  token_count int,
  embedding vector(1536)
);

create or replace function match_page_sections(embedding vector(1536), match_threshold float, match_count int, min_content_length int)
returns table (id bigint, page_id bigint, slug text, heading text, content text, similarity float)
language plpgsql
as $$
#variable_conflict use_variable
begin
  return query
  select
    page_section.id,
    page_section.page_id,
    page_section.slug,
    page_section.heading,
    page_section.content,
    (page_section.embedding <#> embedding) * -1 as similarity
  from page_section

  -- We only care about sections that have a useful amount of content
  where length(page_section.content) >= min_content_length

  -- The dot product is negative because of a Postgres limitation, so we negate it
  and (page_section.embedding <#> embedding) * -1 > match_threshold

  -- OpenAI embeddings are normalized to length 1, so
  -- cosine similarity and dot product will produce the same results.
  -- Using dot product which can be computed slightly faster.
  --
  -- For the different syntaxes, see https://github.com/pgvector/pgvector
  order by page_section.embedding <#> embedding
  
  limit match_count;
end;
$$;

And this is my vector store, but it returned

[Error: Error searching for documents: [object Object]]

const vectorStore = await SupabaseVectorStore.fromExistingIndex(
      new OpenAIEmbeddings(),
      {
        client,
        tableName: 'page_section',
        queryName: 'match_page_sections',
      }
    )

[j1philli]

The rework at #968 of how errors are displayed was just released and gave me more info than just [object, Object] in your exact scenario.