Brylix

A Rust framework for building GraphQL APIs on AWS Lambda with SeaORM and multi-tenant support.

Features

• GraphQL API - Built on async-graphql with playground support
• AWS Lambda - Optimized for serverless deployment with cargo-lambda
• SeaORM - Type-safe database operations with MySQL/PostgreSQL support
• Multi-tenant - Pool-per-droplet architecture for SaaS applications
• JWT Authentication - Secure token-based authentication with multi-role support
• Validation - Built-in input validation utilities
• Email Provider - SMTP email support with attachments
• S3 Provider - Presigned URL generation for file uploads/downloads
• Pagination - Generic pagination utilities for GraphQL connections
• Helpers - JSON parsing, timestamps, soft delete, and ID parsing utilities

Installation

[dependencies]
brylix = "0.2"

Quick Start

use brylix::prelude::*;
use async_graphql::{EmptySubscription, Schema};

#[tokio::main]
async fn main() -> Result<(), brylix::Error> {
    Brylix::builder()
        .config_from_env()
        .with_jwt_auth()
        .with_migrations::<migration::Migrator>()
        .build_schema(|| {
            Schema::build(
                graphql::Query,
                graphql::Mutation,
                EmptySubscription,
            )
            .finish()
        })
        .build()?
        .run()
        .await
}

Feature Flags

# PostgreSQL instead of MySQL
brylix = { version = "0.2", default-features = false, features = ["postgres", "playground"] }

# Multi-tenant support
brylix = { version = "0.2", features = ["multi-tenant"] }

# Email support
brylix = { version = "0.2", features = ["email"] }

# S3 presigned URLs
brylix = { version = "0.2", features = ["s3"] }

Utilities

Pagination

use brylix::prelude::*;

// In a resolver:
async fn list_users(ctx: &Context<'_>, page: u64, per_page: u64) -> Result<Connection<UserDto>> {
    let data = ctx.data_unchecked::<ContextData>();
    let (items, total) = UserService::paginated(&data.db, page, per_page).await?;
    Ok(Connection::new(items, total, page, per_page))
}

// Or use IntoConnection trait with (Vec<T>, u64) tuples:
let result: (Vec<UserDto>, u64) = (items, total);
let connection = result.into_connection(page, per_page);

GraphQL ID Parsing

use brylix::prelude::*;

// Parse string IDs to i64
let user_id = parse_gql_id("123")?;          // Ok(123)
let id = parse_gql_id_field("42", "user_id")?; // Custom error field name

// Or use the macro
let id = gql_id!("123");
let id = gql_id!("123", "user_id");

JSON Column Helpers

use brylix::prelude::*;
use serde_json::json;

// Parse JSON database columns into typed structs
let json_col: Option<serde_json::Value> = Some(json!({"key": "value"}));
let parsed: Option<MyStruct> = json_col.parse_as();
let with_default: Vec<String> = json_col.parse_or_default();

Timestamp Helpers

use brylix::prelude::*;

let now = utc_now(); // chrono::DateTime<Utc>

// Implement Timestamped for your ActiveModel:
impl Timestamped for users::ActiveModel {
    fn set_created_at(&mut self) { self.created_at = Set(utc_now()); }
    fn set_updated_at(&mut self) { self.updated_at = Set(utc_now()); }
}

let mut model = users::ActiveModel { .. };
model.set_timestamps(); // Sets both created_at and updated_at

Soft Delete

use brylix::prelude::*;

// Use status constants
let active = status::ACTIVE;   // "active"
let deleted = status::DELETED;  // "deleted"

// Implement SoftDeletable for your models
impl SoftDeletable for Post {
    fn mark_deleted(&mut self) { self.status = status::DELETED.to_string(); }
    fn is_deleted(&self) -> bool { self.status == status::DELETED }
}

Multi-Role Authentication

use brylix::prelude::*;

// Configure multiple JWT secrets
let jwt_config = MultiRoleJwtConfig::new()
    .add_role("user", std::env::var("JWT_SECRET").unwrap())
    .add_role("admin", std::env::var("ADMIN_JWT_SECRET").unwrap());

// Validate token to determine role
let role = jwt_config.validate(token); // Option<AuthRole>

// Set role in context
let ctx = ContextData::single_tenant(db, user_id, Some(AuthRole::Admin(1)));

// Use guards in resolvers
let admin_id = require_admin(ctx)?;       // Errors if not admin
let role = get_auth_role(ctx);            // Option<&AuthRole>

Environment Variables

# Required
DATABASE_URL=mysql://user:password@host/database
JWT_SECRET=your-secret-key
JWT_EXP_DAYS=7

# Email (optional, requires `email` feature)
SMTP_HOST=smtp.example.com
SMTP_PORT=465
SMTP_USER=your-email@example.com
SMTP_PASSWORD=your-password
SMTP_FROM_NAME=Your App Name
SMTP_FROM_EMAIL=noreply@example.com

# S3 (optional, requires `s3` feature)
S3_BUCKET=my-bucket-name
S3_REGION=us-east-1
S3_UPLOAD_EXPIRES_SECS=3600
S3_DOWNLOAD_EXPIRES_SECS=3600
# Custom credentials for local development (optional)
# If not set, falls back to default AWS credential chain (IAM role for Lambda)
S3_ACCESS_KEY_ID=your-access-key
S3_SECRET_ACCESS_KEY=your-secret-key

Documentation

Full documentation is available at docs.rs/brylix.

License

Licensed under either of Apache License 2.0 or MIT License at your option.