beefyfinance / beefy-db-dev

beefy-db

TL;DR

http://localhost:4000/api/v2/ranges?oracle=BNB&vault=venus-bnb

http://localhost:4000/api/v2/prices?oracle=BNB&bucket=4h_3M

http://localhost:4000/api/v2/apys?vault=venus-bnb&bucket=4h_3M

http://localhost:4000/api/v2/tvls?vault=venus-bnb&bucket=4h_3M

Overview

This app is composed of 2 simple modules:

1. snapshot periodically fetches info from the api and stores it in a database
2. api basic fastify server that allows us to query the stored data

Env

.env is automatically loaded when running the app via yarn dev:* commands.

.env.example contains an example set of variables for development

Variables

DATABASE_URL - postgres connection string

DATABASE_SSL - false: no SSL, true: SSL, default: SSL allowing self-signed certs

PORT - port to run the api on, default: 4000

API_CORS_ORIGIN - RegExp for CORS origin, default: ^(https:\/\/app\.beefy\.(com|finance)|http:\/\/localhost(:[0-9]+)?)$

SNAPSHOT_INTERVAL - interval in seconds between snapshots, default: 900 (15 minutes)

SNAPSHOT_RETRY_DELAY - delay in seconds between snapshot retries, default: 60 (1 minute)

SNAPSHOT_RETRY_MAX - max number of retries, default: 5 (SNAPSHOT_RETRY_DELAY * SNAPSHOT_RETRY_MAX must be less than SNAPSHOT_INTERVAL)

NODE_ENV - development / production, default: development - CSP/CORS is disabled when not production

LOG_LEVEL - 'fatal', 'error', 'warn', 'info', 'debug', 'trace', 'silent', default: 'info' - you may want to set this to warn or above for production

Commands

Development

yarn dev:infra:start - start postgres via docker-compose

yarn dev:migrate - run database migrations via ts-node

yarn dev:snapshot - run the snapshot module via ts-node

yarn dev:api - run the api module via ts-node

Production

yarn build - compile typescript to javascript

yarn migrate - run database migrations

yarn snapshot - run the snapshot module

yarn api - run the api module

Heroku

Heroku will automatically run the build command (via package.json) on deploy.

Heroku will automatically run the migrate command (via Procfile) on release.

The snapshot module is run via the worker process (see Procfile).

The api module is run via the web process (see Procfile).

Logging

Logging is handled by pino

Pino outputs logs in JSON format to stdout. In development this is then piped to pino-pretty for pretty printing.

Endpoints

Ranges

Url

/api/v2/ranges?oracle={oracleId}&vault={vaultId}

Query string

oracle - a price oracle id, e.g. BNB

vault - a vault id, e.g. venus-bnb

Response

{
  "apys": {
    "min": "1627759800",
    "max": "1679023800"
  },
  "tvls": {
    "min": "1627759800",
    "max": "1679023800"
  },
  "prices": {
    "min": "1627759800",
    "max": "1679023800"
  }
}

Prices

Url

/api/v2/prices?oracle={oracleId}&bucket={bucket}

Query string

oracle - a price oracle id, e.g. BNB

bucket - a bucket, size_range e.g. 1h_1d | 1h_1w | 1d_1M | 1d_1Y

Response

[
  {
    "t": 1671244200,
    "v": 0.002845395315750137
  }
]

t - timestamp in seconds, start of bucket

v - high for bucket

APYs

Url

/api/v2/apys?vault={vaultId}&bucket={bucket}

Query string

vault - a vault id, e.g. venus-bnb

bucket - as above

Response

as above for prices

TVLs

Url

/api/v2/tvls?vault={vaultId}&bucket={bucket}

Query string

vault - a vault id, e.g. venus-bnb

bucket - as above

Response

as above for prices

Migrations

Database migrations are handled by postgrator and are available in /migrations/