git-subtree-dir: foundations/net git-subtree-mainline:2b0d510202git-subtree-split:7722c95341
Huly Network Examples
This directory contains comprehensive examples demonstrating various aspects of the Huly Virtual Network.
🆕 Updated to Use Typed Proxies
All examples have been updated to demonstrate the new typed proxies using the cast<T>() method. This provides:
- ✅ Full type safety - TypeScript checks method signatures
- ✅ IDE autocomplete - Get suggestions for available methods
- ✅ Better documentation - Interfaces serve as contracts
- ✅ Refactoring safety - Compile-time errors catch issues
See PROXY_MIGRATION.md for detailed migration guide.
Running Examples
Example Usage
// Define your service interface
interface MyService {
getData: (id: string) => Promise<{ data: any }>
updateData: (id: string, data: any) => Promise<boolean>
}
// On client side - use cast() for typed proxy
const containerRef = await client.get('my-service', {})
const service = containerRef.cast<MyService>('MyService')
// Call methods with full type safety
const result = await service.getData('123')
await service.updateData('123', { updated: true })
Important Note: Auto-Disposal
All examples in this directory use the await using keyword for automatic client disposal. This ensures proper cleanup when the example completes or if an error occurs. For production services that need to run indefinitely, you should use manual lifecycle management instead. See docs/AUTO_DISPOSAL_GUIDE.md for detailed guidance.
Quick Start
Before running any example, make sure you have:
-
Built the project:
node common/scripts/install-run-rush.js build -
Either start a standalone network server:
cd pods/network-pod && rushx devOr let the example start its own network server (most examples do this).
-
Run an example:
cd examples rushx run:basic
Examples
01 - Basic Container with Request/Response
File: 01-basic-container-request-response.ts
Learn the fundamentals of creating containers with typed proxies. This example shows:
- Defining service interfaces for type safety
- Container implementation with
createProxyHandler - Using
cast<T>()method for typed proxies - Request/response pattern with full TypeScript support
- State management within containers
- Proper lifecycle management
Key concepts:
// Define interface
interface DataProcessorService {
store: (key: string, value: any) => Promise<{ success: boolean }>
retrieve: (key: string) => Promise<{ success: boolean; value: any }>
}
// Use typed proxy
const processor = containerRef.cast<DataProcessorService>('DataProcessorService')
await processor.store('key', value)
Use case: Simple data storage service, key-value stores, stateful services
02 - Event Broadcasting
File: 02-event-broadcasting.ts
Demonstrates real-time event broadcasting to multiple connected clients with typed proxies. Features:
- Defining chat service interface with typed methods
- Multiple clients connecting to the same container
- Broadcasting events to all connected clients
- Using typed proxies for chat operations
- Connection lifecycle management
Key concepts:
// Define interface
interface ChatRoomService {
sendMessage: (username: string, text: string) => Promise<{ success: boolean }>
getHistory: () => Promise<{ messages: ChatMessage[] }>
}
// Use typed proxy
const chat = connection.cast<ChatRoomService>('ChatRoomService')
await chat.sendMessage('Alice', 'Hello!')
const history = await chat.getHistory()
Use case: Chat systems, real-time notifications, collaborative editing, live dashboards
03 - Multi-Tenant Container Management
File: 03-multi-tenant.ts
Shows how to build multi-tenant applications with isolated workspaces. Includes:
- Per-tenant container isolation
- Label-based container selection
- Tenant-specific data management
- User and document management per tenant
Use case: SaaS applications, workspace management, multi-tenant platforms
04 - Complete Production Setup
File: 04-complete-production-setup.ts
A comprehensive production-ready example with:
- Multiple redundant agents for high availability
- Health monitoring and metrics collection
- Event monitoring for observability
- Graceful shutdown handling
- Signal handling (SIGTERM, SIGINT)
- Error handling and logging
Use case: Production deployments, enterprise systems, mission-critical services
05 - Error Handling and Retry Logic
File: 05-error-handling-retry.ts
Advanced error handling patterns including:
- Exponential backoff retry logic
- Timeout handling
- Graceful degradation with fallbacks
- Circuit breaker pattern
- Concurrent request handling with error isolation
Use case: Unreliable networks, external service integration, fault-tolerant systems
Custom Timeout Example
File: custom-timeout-example.ts
Shows how to configure different timeouts for various environments:
- Development vs production timeouts
- Connection timeout configuration
- Keep-alive timeout settings
Use case: Environment-specific configuration, debugging, performance tuning
High Availability (HA) Stateless Container Example
File: ha-stateless-container-example.ts
Demonstrates automatic failover with stateless containers:
- Multiple agents competing for the same container UUID
- Automatic leader election
- Failover when primary fails
- Standby agents taking over automatically
Use case: Leader election, active-passive HA, single-instance services with failover
Example Categories
For Beginners
Start with these examples to learn the basics:
01-basic-container-request-response.tscustom-timeout-example.ts
For Application Developers
Build real applications with these patterns:
02-event-broadcasting.ts- Real-time features03-multi-tenant.ts- SaaS applications
For DevOps/Production
Production-ready patterns:
04-complete-production-setup.ts- Full production setup05-error-handling-retry.ts- Robust error handlingha-stateless-container-example.ts- High availability
How to Run Examples
Prerequisites
Make sure you have installed dependencies:
rush install
Run a single example
cd examples
rushx run:basic # Run basic container request-response example
All available examples:
rushx run:basic- Basic container request-responserushx run:events- Event broadcastingrushx run:multi-tenant- Multi-tenant setuprushx run:production- Complete production setuprushx run:retry- Error handling and retryrushx run:timeout- Custom timeout configurationrushx run:ha- High-availability stateless containers
Run with custom network host
Most examples connect to localhost:3737 by default. To use a different host:
const client = createNetworkClient('your-host:3737')
Debug mode
Run with longer timeouts for debugging:
cd examples
NODE_ENV=development rushx run:basic
Common Patterns
Container Implementation Template
import type { Container, ContainerUuid, ClientUuid } from '@hcengineering/network-core'
class MyContainer implements Container {
constructor(readonly uuid: ContainerUuid) {}
async request(operation: string, data?: any, clientId?: ClientUuid): Promise<any> {
switch (operation) {
case 'myOperation':
return { success: true, result: 'data' }
default:
return { success: false, error: 'Unknown operation' }
}
}
async ping(): Promise<void> {}
async terminate(): Promise<void> {}
connect(clientId: ClientUuid, broadcast: (data: any) => Promise<void>): void {}
disconnect(clientId: ClientUuid): void {}
}
Client Usage Template
import { createNetworkClient } from '@hcengineering/network-client'
const client = createNetworkClient('localhost:3737')
await client.waitConnection(5000)
const containerRef = await client.get('my-service' as any, {})
const result = await containerRef.request('myOperation', { data: 'value' })
await containerRef.close()
await client.close()
Agent Setup Template
import { createNetworkClient } from '@hcengineering/network-client'
const client = createNetworkClient('localhost:3737')
await client.waitConnection(5000)
await client.serveAgent('localhost:3738', {
'my-service': async (options) => {
const uuid = options.uuid ?? generateUuid()
const container = new MyContainer(uuid)
return {
uuid,
container,
endpoint: `my-service://host/${uuid}` as any
}
}
})
// Agent server is already started and registered to network
Troubleshooting
Connection refused
- Ensure the network server is running on the specified port
- Check firewall settings
- Verify the correct host and port are being used
Container not found
- Ensure the agent is registered with the network
- Check that the container kind matches what the agent supports
- Verify the agent is still alive (check agent health)
Timeout errors
- Increase timeout values for debugging:
createNetworkClient('host:port', 3600) - Check network connectivity
- Verify services are responding to ping/health checks
Type errors in examples
The examples use dynamic container kinds (strings) which may show TypeScript errors. This is expected and doesn't affect functionality. In production, define proper types:
type MyContainerKind = 'my-service' as ContainerKind
Additional Resources
- Main README - Full project documentation
- HA Stateless Containers - Detailed HA documentation
- Quick Start HA - 5-minute HA guide
- API Reference - Complete API documentation
Contributing Examples
To contribute a new example:
- Create a new file:
XX-descriptive-name.ts - Include comprehensive comments and documentation
- Follow the existing example structure
- Add entry to this README
- Test thoroughly
- Submit a pull request
Example template structure:
- Header comment with description and usage
- Imports
- Container implementation (if applicable)
- Main function with clear sections
- Proper cleanup
- Export reusable components
Need help? Open an issue on GitHub or check the main documentation.