A functional RAG app using D1 , Vectorize , Queues .

• Import pdf , webpage data using a url
• Import long texts
• Supports big imports using queues

1. Install packages

  npm install

2. Create d1, queue, vectorize

  npx wrangler d1 create <db_name>
  npx wrangler create queue <queue_name>
  npx wrangler vectorize create <index_name> --preset @cf/baai/bge-base-en-v1.5

( Update the values like id and name in wrangler.toml ) (Note : Don't change the bindings)

3. Create a table in the D1 database

npx wrangler d1 execute <db_name> --command "CREATE TABLE IF NOT EXISTS texts (id INTEGER PRIMARY KEY, text TEXT NOT NULL)"

4. Deploy (This won't work on local because queues and vectorize is not supported)

npx wrangler deploy --minify