Serverless, SQL based API for serving Unchained Index.

The project is structured as a collection of modules, which can share code but run independently. They also have different dependencies - this guarantees small lambda (serverless) executable size.

Structure

1. config handles configuration files and env variables
2. database everything database-related
3. extract take whole index and convert it to SQL. Swap tables (staging -> live)
4. query lambda (serverless function) and a cmd to find appearances
5. scanner (deprecated) old lambda to perform appearance lookup
6. queue insert to/read from the queue that feeds SQL database
7. quicknode QuickNode integration related logic: handling accounts, authorization, provision and healthcheck
8. test/integration inntegration tests (they run mocked environment in Docker containers)

How it works

The data can be initially ingested into the queue (and then SQL database) using extract tool. Later on, the scraper sends new appearances as notifications to queue/insert.

User can sign up using QuickNode Marketplace. This is handled by quicknode/provision code.

After signing up, the user can ask for appearances by sending JSON-RPC request to our API. The request is handled by query/lambda.

Plans and API keys

Plans and API keys for the users are defined in SAM template. They include rate limits.