This repo contains the backend part of the Growthcast extension. The current repo contains the following main components:
- API endpoints to interact with the extension (more details about the API endpoints)
- The sign-in page that can be used to login to a Farcaster account
Growthcast login page is available by the following path:
http://localhost:3000/signin
Requirements:
- Cloud or hardware server to:
- Running Farcaster node
- Running PostgreSQL database (14.12+)
- Neynar API key. It could be generated on https://neynar.com
- npm: 10.5.0+
- Next.js: 14.2.4+
To run the backend you need to have the running Farcaster node. To get more details on how to launch your own Farcaster node, you can refer to the following documentation
The basic server setup to run the Farcaster node:
- 16 GB of RAM
- 4 CPU cores or vCPUs
- 200 GB of free storage
- A public IP address with ports 2282 - 2285 Exposed
To run the backend you also need to have a running PostgreSQL database. It could be on the same server or on a separate one
The requirement version is 14.12+
The database schema is described in the following section:
For simplicity, some of the endpoints use NeynarSDK to get Farcaster data. NeynarSDK also provides the login. If you don't want to use NeynarSDK, you need to re-write necessary endpoints to interact with Farcaster Node as well as re-write the login process without using the SDK
To get the Neynar API key, go to https://neynar.com and buy the subscription. The key is available in the dev portal: https://dev.neynar.com
To run the backend properly, you need to create a file with environment variables. You can use 2 different files for the development and production:
- .env - for the production
- .env.development - for the development
The more details about variables will be described in the table below:
Environment Variable | Description | Example Value |
---|---|---|
NEXT_PUBLIC_NEYNAR_CLIENT_ID |
Client ID for Neynar | 07082225-8b12-458f-b0ba-d5bf6f26a7e6 |
NEXT_PUBLIC_NEYNAR_API_KEY |
API Key for Neynar | D2DD113E-6156-4837-B1F7-1AF7F410036E |
NEXT_PUBLIC_DOMAIN |
Domain for local development | http://localhost:3000 |
NEXT_PUBLIC_ENCRYPTION_KEY |
Key used for encryption | 4ed2309caafc493d29432f16626163d8ff683de2522960e3765c3459777fce14 |
NEXT_PUBLIC_REFRESH_ENCRYPTION_KEY |
Key used for refreshing encryption | 2qaBFBAJ2UDha0hntZG+5TN7SjKcsNLOGQUgDN+Lul0= |
NEXT_PUBLIC_IMGUR_CLIENT_ID |
Imgur Client ID | 129b687c7b3f67b |
NEXT_PUBLIC_IMGUR_CLIENT_SECRET |
Imgur Client Secret | 993b7160c7aad652783d1bf168c256603ac1f339 |
NEXT_PUBLIC_WARPCAST_HOST |
Host URL for Warpcast API | https://api.warpcast.com |
NEXT_PUBLIC_OPENRANK_HOST |
Host URL for OpenRank API | https://graph.cast.k3l.io |
NEXT_PUBLIC_NEYNAR_HOST |
Host URL for Neynar API | https://api.neynar.com |
NEXT_PUBLIC_DEV_MODE |
Enable/Disable development mode | true |
SENTRY_AUTH_TOKEN |
Auth token for Sentry | sntrys_eyJpOPQiOjE3MjAzNzQ4NTkuMTE3OTI1LCJ1cmwiOiJodHRwczovL3NlbnRyeS5pbyIsInJlZ2lvbl91cmwiOiJodHRwczovL2RlLnNlbnRyeS5pbyIsIm9yZyI6Imdyb3d0aGNhc3QifQ==_DHBTfO7HLAjPtK8E042ws39AJh8rJ4fD1lQ+7i2kcp8 |
NEXT_PUBLIC_NODE_ADDRESS |
Node address for Warp Drive | http://hub.warp-drive.xyz |
POSTGRES_V2_URL |
PostgreSQL connection URL | postgres://manager:[email protected]/growthcast_dev |
POSTGRES_V2_URL_NON_POOLING |
PostgreSQL connection URL (non-pooling) | postgres://manager:[email protected]/growthcast_dev?sslmode=require |
POSTGRES_V2_URL_NO_SSL |
PostgreSQL connection URL (no SSL) | postgres://manager:[email protected]/growthcast_dev |
POSTGRES_V2_PRISMA_URL |
PostgreSQL connection URL for Prisma | postgres://manager:[email protected]/growthcast_dev?pgbouncer=true&connect_timeout=15&sslmode=require |
POSTGRES_V2_USER |
PostgreSQL user | manager |
POSTGRES_V2_PASSWORD |
PostgreSQL password | 3bCaP3Axg1jnqQw6NERg |
POSTGRES_V2_HOST |
PostgreSQL host | 135.181.142.175 |
POSTGRES_V2_DATABASE |
PostgreSQL database name | growthcast_dev |
The following steps should be completed before starting the further installation:
- Launch the server
- Run a Farcaster node
- Run a PostgreSQL database
- Create a file with environmental variables
- Create and add the Neynar API key to the .env / .env.development
Right after all the steps above are completed, you can launch the backend!
-
Install all necessary dependencies:
npm install
-
Run the backend locally
npm run dev
-
The backend API is available through http://localhost:3000/api. The sign-in is available at http://localhost:3000/signin URL
You can deploy the backend in one click:
Make sure to add the environment variables for your Vercel project. Get more details about the envs on the following section:
You can also deploy the project to Vercel manually using the bash command:
vercel
Make sure that you've installed the Vercel with the ... version
Below there will describe some technical details about API, PostgreSQL, and Farcaster data processing
The API endpoints swagger is available by the following URL:
http://localhost:3000/api-doc
Several database tables are mainly used by the backend:
- users_casts_historical_data Stores pre-processed information about user's casts for the whole user's cast history
- users_info Stores the basic pre-processed data about a user and his activity on Farcaster
All tables will be automatically created during the first launch. The only thing that you need to do is to create the database that should have the "growthcast" name and "growthcast_dev" for the development database
users_all_historical_data table schema:
CREATE TABLE users_casts_historical_data (
fid int4 NULL,
cast_timestamp timestamp NULL,
cast_text text NULL,
cast_hash text NULL,
cast_likes int4 NULL,
cast_replies int4 NULL,
cast_recasts int4 NULL,
row_created_date timestamp NULL DEFAULT CURRENT_TIMESTAMP
)
users_info table schema:
CREATE TABLE IF NOT EXISTS users_info (
id SERIAL PRIMARY KEY,
fid INTEGER NOT NULL,
username VARCHAR(255) NOT NULL,
display_name VARCHAR(255) NOT NULL,
pfp_url TEXT,
followers INTEGER DEFAULT 0,
followings INTEGER DEFAULT 0,
verified_address JSONB,
is_data_fetched BOOLEAN DEFAULT FALSE,
row_created_date TIMESTAMP DEFAULT CURRENT_TIMESTAMP
)
The logic of the FarcasterDataProcessor and FarcasterReactionsDataProcessor is described on the schema below
TLDR:
- User logins to the Growthcast
- Starting to fetch his data from the current day to his first day until his first day by batches
- Adding batches of data to the database, so a service can get access to data without waiting for fetching of all data
This work is licensed under a Creative Commons Attribution 4.0 International License.