A flexible Python library for endpoint management and URI processing
Project description
UriPoint - Universal Network Endpoint Management
UriPoint is a comprehensive Python library designed for unified network endpoint management across multiple protocols. It simplifies the creation, management, and monitoring of network endpoints through a single, cohesive interface.
- Unified Management: Handle multiple protocols through one interface
- Simplified Configuration: Easy setup through CLI or Python API
- Protocol Agnostic: Support for web, streaming, IoT, and messaging protocols
- Production Ready: Built-in monitoring, testing, and persistence
Overview of UriPoint's Wide-Ranging Capabilities
┌─────────────────────────────┐
│ UriPoint Platform │
└─────────────────────────────┘
│
┌──────────────┬────────┴───────┬────────────────┐
▼ ▼ ▼ ▼
┌──────────────┐ ┌──────────────┐ ┌──────────────┐ ┌──────────────┐
│ Protocol │ │ System │ │ Management │ │ Integration │
│ Layer │ │ Architecture │ │ Tools │ │ Framework │
└──────────────┘ └──────────────┘ └──────────────┘ └──────────────┘
Comprehensive Protocol Support
Communication Protocols
- Web Protocols: HTTP/HTTPS, WebSocket, GraphQL
- Streaming: RTSP, HLS, DASH
- IoT: MQTT, CoAP
- Messaging: AMQP, ZeroMQ
- Storage: Redis, MongoDB
- Email: SMTP, IMAP
- File Transfer: FTP, SFTP
- Name Resolution: DNS
Protocol Features
- Protocol-specific optimizations
- Auto-negotiation
- Fallback mechanisms
- Protocol conversion
- Security layers
Technical Highlights
- Persistent Configuration: YAML-based storage (
~/.uripoint_config.yaml
) - Protocol Handlers: Dedicated handlers for each supported protocol
- Monitoring System: Real-time endpoint status tracking
- Testing Suite: Comprehensive testing framework with chaos engineering support
Target Audience
┌────────────────────────────────────────────────────────────────────┐
│ UriPoint Platform │
├────────────────────────────────────────────────────────────────────┤
│ │
│ ┌─────────────────┐ ┌─────────────────┐ │
│ │ Distributed │ ┌─────────────────┐ │ Streaming │ │
│ │ Systems │ │ DevOps │ │ Engineers │ │
│ │ Developers │ │ Engineers │ └────────┬────────┘ │
│ └────────┬────────┘ └────────┬────────┘ │ │
│ │ │ │ │
│ │ │ │ │
│ ▼ ▼ ▼ │
│ ┌─────────────────────────────────────────────────────────┐ │
│ │ │ │
│ │ ┌─────────────┐ ┌─────────────┐ ┌─────────────┐ │ │
│ │ │ Microservice│ │ Service │ │ Stream │ │ │
│ │ │ Architecture│ │ Deployment │ │ Processing │ │ │
│ │ └─────────────┘ └─────────────┘ └─────────────┘ │ │
│ │ │ │
│ │ ┌─────────────┐ ┌─────────────┐ ┌─────────────┐ │ │
│ │ │ Service │ │ Auto │ │ Live │ │ │
│ │ │ Discovery │ │ Scaling │ │ Streaming │ │ │
│ │ └─────────────┘ └─────────────┘ └─────────────┘ │ │
│ │ │ │
│ └─────────────────────────────────────────────────────────┘ │
│ ▲ ▲ │
│ │ │ │
│ ┌─────────────────┘ └───────┐ │
│ │ │ │
│ ┌───────┴────────┐ ┌───────┴──────┐ │
│ │ IoT │ │ Backend │ │
│ │ Specialists │ │ Developers │ │
│ └───────┬────────┘ └───────┬──────┘ │
│ │ │ │
│ │ ┌─────────────────────┐ │ │
│ └─────────►│ Core Features │◄─────────────┘ │
│ │ ┌─────────────┐ │ │
│ │ │ Device │ │ │
│ │ │ Management │ │ │
│ │ └─────────────┘ │ │
│ │ ┌─────────────┐ │ │
│ │ │ API │ │ │
│ │ │ Gateway │ │ │
│ │ └─────────────┘ │ │
│ │ ┌─────────────┐ │ │
│ │ │ Message │ │ │
│ │ │ Queue │ │ │
│ │ └─────────────┘ │ │
│ │ ┌─────────────┐ │ │
│ │ │ Cache │ │ │
│ │ │ Management │ │ │
│ │ └─────────────┘ │ │
│ └─────────────────────┘ │
│ │
└────────────────────────────────────────────────────────────────────┘
1. Distributed Systems Developers
- Building microservices architectures
- Managing service discovery
- Implementing API gateways
- Developing service meshes
2. DevOps Engineers
- Infrastructure automation
- Service deployment
- Monitoring setup
- Configuration management
3. IoT Specialists
- Device management
- Sensor networks
- Data collection systems
- Remote monitoring
4. Streaming Engineers
- Video streaming services
- Live broadcast systems
- Security camera networks
- Content delivery networks
5. Backend Developers
- API development
- Service integration
- Message queue implementation
- Cache management
Architecture
System Components
┌───────────────────────────────────────────────────────┐
│ UriPoint System │
├───────────────┬──────────────────┬────────────────────┤
│ Core Engine │ Protocol Layer │ Management Layer │
├───────────────┼──────────────────┼────────────────────┤
│ - Config │ - HTTP/HTTPS │ - CLI Interface │
│ - Persistence │ - MQTT │ - API Interface │
│ - Monitoring │ - RTSP │ - Web Dashboard │
│ - Testing │ - Redis │ - Monitoring UI │
└───────────────┴──────────────────┴────────────────────┘
Component Interaction
-
Core Engine
- Configuration management
- State persistence
- Event handling
- Resource management
-
Protocol Layer
- Protocol-specific handlers
- Data transformation
- Connection management
- Error handling
-
Management Layer
- User interfaces
- Monitoring tools
- Administration features
- Reporting systems
Protocol Support
Web & API Protocols
-
HTTP/HTTPS
- RESTful endpoints
- WebHooks
- Static content
- API proxying
-
WebSocket
- Real-time communication
- Bi-directional data flow
- Connection pooling
- Event streaming
Streaming Protocols
-
RTSP
- Camera streams
- Media servers
- Stream recording
- Transport selection
-
HLS/DASH
- Adaptive bitrate
- Multi-quality
- Live streaming
- VOD support
IoT & Messaging
-
MQTT
- Device management
- Sensor data
- Command control
- State synchronization
-
AMQP
- Message queuing
- Event routing
- Dead letter handling
- Queue management
Installation & Setup
Basic Installation
pip install uripoint
Configuration
# ~/.uripoint_config.yaml
endpoints:
- name: api-endpoint
uri: http://localhost:8080/api
protocol: http
methods: [GET, POST]
- name: mqtt-sensor
uri: mqtt://broker:1883/sensors
protocol: mqtt
qos: 1
Usage Guide
Service Lifecycle
┌──────────────┐ ┌──────────────┐ ┌──────────────┐
│ Create │ │ Configure │ │ Deploy │
│ Endpoint ├──►│ Protocol ├──►│ Service │
└──────────────┘ └──────────────┘ └──────────────┘
▲ │
│ ▼
┌──────────────┐ ┌──────────────┐
│ Update │◄───────────────────│ Monitor │
│ Configuration│ │ Status │
└──────────────┘ └──────────────┘
CLI Commands
uripoint [options] <command>
┌────────────────────────────────────────────────┐
│ --uri Define endpoint URI │
│ --method Specify HTTP methods │
│ --data Configure endpoint data │
│ --serve Start serving endpoints │
│ --list Show configured endpoints │
│ --detach Remove endpoints │
└────────────────────────────────────────────────┘
Endpoint Management
Options
# List all configured endpoints
uripoint --list
# Serve all endpoints as a live server
uripoint --serve
# Test endpoints
uripoint --test
# Detach specific endpoints
uripoint --detach "http://localhost:9000/api/hello" "http://localhost:9001/metrics"
# Detach all endpoints
uripoint --detach
Examples
uripoint create --uri http://localhost:8080/api --protocol http
uripoint list [--format json|yaml]
uripoint serve [--port 8000]
uripoint monitor [--interval 5]
uripoint remove [endpoint-name]
Protocol-Specific Commands
uripoint stream start --uri rtsp://camera:554/feed
uripoint mqtt publish --topic sensors/temp --message "23.5"
uripoint http test --endpoint api-endpoint
High Availability Setup
from uripoint import HighAvailability
ha = HighAvailability(
endpoints=['endpoint1', 'endpoint2'],
strategy='active-passive',
failover_timeout=30
)
ha.start()
Endpoint Creation Methods
- Full URI Approach with HTTP Methods
# Create an endpoint with full URI and specific HTTP methods
uripoint --uri http://localhost:8080/api/users --data '{"response": {"status": "OK"}}' --method GET POST PUT
- Component-Based Approach
# Create an endpoint using individual components
uripoint --hostname localhost --path /api/status --protocol http --port 8001 --data '{"status": "OK"}' --method GET
Protocol Examples
HTTP/REST API
# Python
from uripoint import UriPointCLI
cli = UriPointCLI()
cli.create_endpoint(
uri='http://localhost:8000/api/users',
data={
'response': {'users': []},
'methods': ['GET', 'POST', 'PUT', 'DELETE']
}
)
# CLI & curl
# Create endpoint
uripoint --uri http://localhost:8000/api/users --data '{"response": {"users": []}}' --method GET POST PUT DELETE
# Test endpoint
curl -X GET http://localhost:8000/api/users
curl -X POST http://localhost:8000/api/users -H "Content-Type: application/json" -d '{"name": "John"}'
curl -X PUT http://localhost:8000/api/users/1 -H "Content-Type: application/json" -d '{"name": "John Doe"}'
curl -X DELETE http://localhost:8000/api/users/1
MQTT IoT Device
# Python
cli.create_endpoint(
uri='mqtt://localhost:1883/sensors/temperature',
data={
'topic': 'sensors/temperature',
'qos': 1,
'device': {
'type': 'temperature',
'location': 'room1'
}
}
)
# CLI & curl/mosquitto
# Create endpoint
uripoint --uri mqtt://localhost:1883/sensors/temperature --data '{"topic": "sensors/temperature", "qos": 1}'
# Test endpoint
mosquitto_pub -h localhost -p 1883 -t sensors/temperature -m '{"value": 22.5}'
mosquitto_sub -h localhost -p 1883 -t sensors/temperature
Redis Cache
# Python
cli.create_endpoint(
uri='redis://localhost:6379/cache',
data={
'db': 0,
'decode_responses': True,
'max_connections': 10
}
)
# CLI & curl
# Create endpoint
uripoint --uri redis://localhost:6379/cache --data '{"db": 0, "decode_responses": true}'
# Test endpoint
curl -X GET http://localhost:6379/cache/user:123
curl -X PUT http://localhost:6379/cache/user:123 -d '{"name": "John", "age": 30}'
curl -X DELETE http://localhost:6379/cache/user:123
SMTP Email
# Python
cli.create_endpoint(
uri='smtp://smtp.gmail.com:587/mail',
data={
'use_tls': True,
'timeout': 30
}
)
# CLI & curl
# Create endpoint
uripoint --uri smtp://smtp.gmail.com:587/mail --data '{"use_tls": true, "timeout": 30}'
# Test endpoint
curl -X POST http://localhost:587/mail \
-H "Content-Type: application/json" \
-d '{
"to": "user@example.com",
"subject": "Test Email",
"body": "Hello from UriPoint"
}'
RTSP Stream
# Python
cli.create_endpoint(
uri='rtsp://localhost:8554/camera1',
data={
'stream_url': 'rtsp://camera.example.com/stream1',
'transport': 'tcp'
}
)
# CLI & ffmpeg/curl
# Create endpoint
uripoint --uri rtsp://localhost:8554/camera1 --data '{"stream_url": "rtsp://camera.example.com/stream1", "transport": "tcp"}'
# Test endpoint
ffplay rtsp://localhost:8554/camera1
curl http://localhost:8554/camera1/info # Get stream info
HLS Stream
# Python
cli.create_endpoint(
uri='http://localhost:8080/live/stream.m3u8',
data={
'manifest_url': '/live/stream.m3u8',
'segment_duration': 6,
'options': {
'bandwidth_variants': [
{'resolution': '1080p', 'bitrate': 5000000},
{'resolution': '720p', 'bitrate': 2500000}
]
}
}
)
# CLI & curl/ffmpeg
# Create endpoint
uripoint --uri http://localhost:8080/live/stream.m3u8 --data '{"manifest_url": "/live/stream.m3u8", "segment_duration": 6}'
# Test endpoint
curl http://localhost:8080/live/stream.m3u8 # Get manifest
ffplay http://localhost:8080/live/stream.m3u8 # Play stream
DASH Example
# Create DASH endpoint for video on demand
cli.create_endpoint(
uri='http://localhost:8080/vod/manifest.mpd',
data={
'mpd_url': '/vod/manifest.mpd',
'segment_duration': 4,
'options': {
'quality_levels': [
{'resolution': '2160p', 'bitrate': 15000000},
{'resolution': '1080p', 'bitrate': 4500000}
]
}
}
)
MQTT Example
# Create MQTT endpoint for temperature sensor
cli.create_endpoint(
uri='mqtt://localhost:1883/sensors/temperature',
data={
'topic': 'sensors/temperature',
'qos': 1,
'retain': True,
'device': {
'id': 'temp_sensor_01',
'type': 'temperature',
'location': 'living_room'
}
}
)
# Create MQTT endpoint for smart light control
cli.create_endpoint(
uri='mqtt://localhost:1883/devices/lights',
data={
'topic': 'devices/lights',
'qos': 1,
'retain': True,
'device': {
'id': 'smart_light_01',
'capabilities': ['dimming', 'color']
}
}
)
AMQP Message Queue
# Python
cli.create_endpoint(
uri='amqp://localhost:5672/orders',
data={
'exchange': 'orders',
'queue': 'new_orders',
'routing_key': 'order.new'
}
)
# CLI & curl
# Create endpoint
uripoint --uri amqp://localhost:5672/orders --data '{"exchange": "orders", "queue": "new_orders"}'
# Test endpoint
curl -X POST http://localhost:5672/orders \
-H "Content-Type: application/json" \
-d '{"order_id": "123", "items": ["item1", "item2"]}'
curl -X GET http://localhost:5672/orders/status
DNS Service
# Python
cli.create_endpoint(
uri='dns://localhost:53/lookup',
data={
'timeout': 5,
'cache_enabled': True
}
)
# CLI & curl/dig
# Create endpoint
uripoint --uri dns://localhost:53/lookup --data '{"timeout": 5, "cache_enabled": true}'
# Test endpoint
curl "http://localhost:53/lookup?domain=example.com"
dig @localhost example.com
WebSocket Chat
# Python
cli.create_endpoint(
uri='ws://localhost:8080/chat',
data={
'protocol': 'chat',
'max_connections': 100
}
)
# CLI & websocat
# Create endpoint
uripoint --uri ws://localhost:8080/chat --data '{"protocol": "chat", "max_connections": 100}'
# Test endpoint
websocat ws://localhost:8080/chat
wscat -c ws://localhost:8080/chat
Web API Examples
1. Basic REST API Endpoint
# CLI Command
uripoint --uri http://localhost:8080/api/users \
--method GET POST PUT DELETE \
--data '{
"response": {"users": []},
"cors": true,
"auth": {"type": "basic"}
}'
# Test Commands
curl -X GET http://localhost:8080/api/users
curl -X POST http://localhost:8080/api/users -d '{"name": "John"}'
curl -X PUT http://localhost:8080/api/users/1 -d '{"name": "John Doe"}'
curl -X DELETE http://localhost:8080/api/users/1
# Verification
uripoint --test http://localhost:8080/api/users --method GET
uripoint --verify http://localhost:8080/api/users
# Python Implementation
from uripoint import UriPointCLI, EndpointTest
# Create endpoint
cli = UriPointCLI()
endpoint = cli.create_endpoint(
uri='http://localhost:8080/api/users',
data={
'response': {'users': []},
'methods': ['GET', 'POST', 'PUT', 'DELETE'],
'cors': True,
'auth': {'type': 'basic'}
}
)
# Test endpoint
test = EndpointTest(endpoint)
test.run_http_tests()
test.verify_cors()
test.check_methods()
Streaming Examples
2. RTSP Camera Stream
# CLI Command
uripoint --uri rtsp://localhost:8554/camera1 \
--data '{
"stream_url": "rtsp://camera.example.com/stream1",
"transport": "tcp",
"authentication": {
"username": "${CAMERA_USER}",
"password": "${CAMERA_PASS}"
},
"options": {
"framerate": 30,
"resolution": "1920x1080"
}
}'
# Test Commands
ffplay rtsp://localhost:8554/camera1
ffprobe -v error -show_entries stream=codec_type rtsp://localhost:8554/camera1
# Verification
uripoint --test rtsp://localhost:8554/camera1 --check-stream
uripoint --verify rtsp://localhost:8554/camera1 --params framerate,resolution
# Python Implementation
from uripoint import UriPointCLI, StreamTest
# Create streaming endpoint
cli = UriPointCLI()
stream = cli.create_stream_endpoint(
uri='rtsp://localhost:8554/camera1',
data={
'stream_url': 'rtsp://camera.example.com/stream1',
'transport': 'tcp',
'authentication': {
'username': os.getenv('CAMERA_USER'),
'password': os.getenv('CAMERA_PASS')
},
'options': {
'framerate': 30,
'resolution': '1920x1080'
}
}
)
# Test stream
test = StreamTest(stream)
test.verify_stream_health()
test.check_parameters()
test.monitor_bandwidth()
IoT Examples
3. MQTT Temperature Sensor Network
# CLI Command
uripoint --uri mqtt://localhost:1883/sensors/temperature \
--data '{
"topics": ["sensors/+/temperature"],
"qos": 1,
"retain": true,
"device_schema": {
"type": "temperature",
"unit": "celsius",
"interval": 60
}
}'
# Test Commands
mosquitto_pub -h localhost -t sensors/room1/temperature -m '{"value": 22.5}'
mosquitto_sub -h localhost -t 'sensors/+/temperature'
# Verification
uripoint --test mqtt://localhost:1883/sensors/temperature
uripoint --verify mqtt://localhost:1883/sensors/temperature --check-schema
# Python Implementation
from uripoint import UriPointCLI, MQTTTest
# Create MQTT endpoint
cli = UriPointCLI()
mqtt = cli.create_mqtt_endpoint(
uri='mqtt://localhost:1883/sensors/temperature',
data={
'topics': ['sensors/+/temperature'],
'qos': 1,
'retain': True,
'device_schema': {
'type': 'temperature',
'unit': 'celsius',
'interval': 60
}
}
)
# Test MQTT endpoint
test = MQTTTest(mqtt)
test.verify_connectivity()
test.test_pub_sub()
test.validate_messages()
Message Queue Examples
4. AMQP Order Processing Queue
# CLI Command
uripoint --uri amqp://localhost:5672/orders \
--data '{
"exchange": "order_exchange",
"queues": {
"new_orders": {"durable": true},
"processed_orders": {"durable": true}
},
"routing_keys": ["order.new", "order.processed"],
"dlx": "order_dlx"
}'
# Test Commands
curl -X POST http://localhost:5672/orders/send \
-d '{"order_id": "123", "status": "new"}'
curl -X GET http://localhost:5672/orders/status
# Verification
uripoint --test amqp://localhost:5672/orders --check-queues
uripoint --verify amqp://localhost:5672/orders --check-bindings
# Python Implementation
from uripoint import UriPointCLI, AMQPTest
# Create AMQP endpoint
cli = UriPointCLI()
amqp = cli.create_amqp_endpoint(
uri='amqp://localhost:5672/orders',
data={
'exchange': 'order_exchange',
'queues': {
'new_orders': {'durable': True},
'processed_orders': {'durable': True}
},
'routing_keys': ['order.new', 'order.processed'],
'dlx': 'order_dlx'
}
)
# Test AMQP endpoint
test = AMQPTest(amqp)
test.verify_exchange()
test.check_queues()
test.test_message_flow()
Storage Examples
5. Redis Cache System
# CLI Command
uripoint --uri redis://localhost:6379/cache \
--data '{
"database": 0,
"key_prefix": "app:",
"max_connections": 10,
"timeout": 5,
"encoding": "utf-8"
}'
# Test Commands
curl -X SET http://localhost:6379/cache/user:123 -d '{"name": "John"}'
curl -X GET http://localhost:6379/cache/user:123
redis-cli -h localhost PING
# Verification
uripoint --test redis://localhost:6379/cache --check-connection
uripoint --verify redis://localhost:6379/cache --check-encoding
# Python Implementation
from uripoint import UriPointCLI, RedisTest
# Create Redis endpoint
cli = UriPointCLI()
redis = cli.create_redis_endpoint(
uri='redis://localhost:6379/cache',
data={
'database': 0,
'key_prefix': 'app:',
'max_connections': 10,
'timeout': 5,
'encoding': 'utf-8'
}
)
# Test Redis endpoint
test = RedisTest(redis)
test.verify_connectivity()
test.test_operations()
test.check_performance()
Email Examples
6. SMTP Email Service
# CLI Command
uripoint --uri smtp://smtp.gmail.com:587/mail \
--data '{
"auth": {
"username": "${SMTP_USER}",
"password": "${SMTP_PASS}"
},
"tls": true,
"templates": {
"welcome": {"path": "templates/welcome.html"},
"reset": {"path": "templates/reset.html"}
}
}'
# Test Commands
curl -X POST http://localhost:587/mail/send \
-d '{
"to": "user@example.com",
"template": "welcome",
"data": {"name": "John"}
}'
# Verification
uripoint --test smtp://smtp.gmail.com:587/mail --check-auth
uripoint --verify smtp://smtp.gmail.com:587/mail --check-templates
# Python Implementation
from uripoint import UriPointCLI, SMTPTest
# Create SMTP endpoint
cli = UriPointCLI()
smtp = cli.create_smtp_endpoint(
uri='smtp://smtp.gmail.com:587/mail',
data={
'auth': {
'username': os.getenv('SMTP_USER'),
'password': os.getenv('SMTP_PASS')
},
'tls': True,
'templates': {
'welcome': {'path': 'templates/welcome.html'},
'reset': {'path': 'templates/reset.html'}
}
}
)
# Test SMTP endpoint
test = SMTPTest(smtp)
test.verify_connection()
test.test_templates()
test.check_delivery()
Multi-Protocol Examples
7. Full IoT Monitoring System
# CLI Commands
# Create MQTT endpoint for sensor data
uripoint --uri mqtt://localhost:1883/sensors \
--data '{
"topics": ["sensors/#"],
"qos": 1
}'
# Create Redis endpoint for data storage
uripoint --uri redis://localhost:6379/sensor-data \
--data '{
"database": 1,
"expire": 3600
}'
# Create HTTP API endpoint
uripoint --uri http://localhost:8080/api/sensors \
--method GET POST \
--data '{
"cache": "redis://localhost:6379/sensor-data",
"mqtt": "mqtt://localhost:1883/sensors"
}'
# Test Commands
# Test MQTT data flow
mosquitto_pub -t sensors/temp1 -m '{"value": 24.5}'
# Check Redis storage
redis-cli -h localhost -n 1 GET sensor:temp1
# Verify HTTP API
curl http://localhost:8080/api/sensors/temp1
# Verification
uripoint --test-suite iot-monitoring --verify-all
# Python Implementation
from uripoint import UriPointCLI, SystemTest
# Create system endpoints
cli = UriPointCLI()
# MQTT endpoint
mqtt = cli.create_mqtt_endpoint(
uri='mqtt://localhost:1883/sensors',
data={
'topics': ['sensors/#'],
'qos': 1
}
)
# Redis endpoint
redis = cli.create_redis_endpoint(
uri='redis://localhost:6379/sensor-data',
data={
'database': 1,
'expire': 3600
}
)
# HTTP API endpoint
api = cli.create_endpoint(
uri='http://localhost:8080/api/sensors',
data={
'methods': ['GET', 'POST'],
'cache': 'redis://localhost:6379/sensor-data',
'mqtt': 'mqtt://localhost:1883/sensors'
}
)
# Create system test
test = SystemTest([mqtt, redis, api])
test.verify_integration()
test.test_data_flow()
test.check_performance()
Testing Framework Examples
Common Test Cases
# Basic endpoint test
test = EndpointTest(endpoint)
test.run()
# Protocol-specific tests
test.verify_protocol()
test.check_security()
test.validate_performance()
# Integration tests
test.verify_dependencies()
test.check_data_flow()
test.test_failover()
# Load tests
test.concurrent_access(num_clients=100)
test.sustained_load(duration=300)
test.peak_performance()
# Chaos tests
test.network_partition()
test.process_failure()
test.data_corruption()
Test Configuration
# test_config.yaml
tests:
performance:
concurrent_users: 100
duration: 300
thresholds:
latency_p95: 200
error_rate: 0.01
integration:
dependencies:
- redis
- mqtt
- http
timeout: 30
chaos:
scenarios:
- network_partition
- process_crash
- disk_full
duration: 600
See examples/protocol_examples/ for more comprehensive examples:
- streaming_example.py: RTSP, HLS, and DASH streaming
- mqtt_iot_example.py: IoT device communication
- redis_example.py: Data store access
- smtp_example.py: Email handling
- amqp_example.py: Message queuing
- dns_example.py: Domain resolution
Configuration File Location
- Path:
~/.uripoint_config.yaml
- Format: YAML
- Contents: List of endpoint configurations
Testing Framework
UriPoint includes a comprehensive testing framework that ensures reliability and performance across all components:
┌────────────────────────┐ ┌─────────────────────────┐
│ Performance Tests │ │ Integration Tests │
├────────────────────────┤ ├─────────────────────────┤
│ - Endpoint Creation │ │ - Component Interaction │
│ - Concurrent Access │ │ - Multi-Protocol │
│ - Memory Usage │ │ - Process Management │
│ - Protocol Handlers │ │ - Error Propagation │
└────────────────────────┘ └─────────────────────────┘
┌────────────────────────┐ ┌─────────────────────────┐
│ Chaos Tests │ │ Protocol Tests │
├────────────────────────┤ ├─────────────────────────┤
│ - Random Operations │ │ - Protocol Validation │
│ - Process Chaos │ │ - Handler Behavior │
│ - Network Simulation │ │ - Configuration │
│ - Data Input Chaos │ │ - Error Handling │
└────────────────────────┘ └─────────────────────────┘
Running Tests
# Run all tests
pytest tests/
# Run specific test categories
pytest tests/test_protocols.py
pytest tests/test_performance.py
pytest tests/test_integration.py
pytest tests/test_chaos.py
# Run with coverage report
pytest --cov=uripoint tests/
Test Categories
-
Performance Tests
- Endpoint creation benchmarks
- Concurrent access testing
- Memory usage monitoring
- Protocol handler performance
-
Integration Tests
- Component interaction verification
- Multi-protocol integration
- Process-endpoint integration
- Error propagation testing
-
Chaos Tests
- Random endpoint operations
- Process management chaos
- Network chaos simulation
- Data input chaos testing
-
Protocol-Specific Tests
- Protocol handler validation
- Configuration verification
- Error handling
- Protocol behavior testing
UriPoint: A Comprehensive Network Communication Solution
UriPoint provides a comprehensive solution that goes beyond simple protocol handling, offering a complete platform for building, managing, and operating network communications. Its wide-ranging capabilities make it suitable for enterprise applications while maintaining the simplicity needed for smaller projects.
Unique features:
- Unified interface for multiple protocols
- Built-in endpoint management
- Configuration persistence system
- Integrated testing framework
- CLI for easy management
- Streaming protocol support
- Comprehensive monitoring
Closest to UriPoint functionality are combinations of several libraries, e.g.:
- FastAPI + Mosquitto + aiohttp
- Flask + pika + SocketIO
- Django Channels + RabbitMQ
However, it would still require significant integration work to achieve the level of unification offered by UriPoint.
Comparing Integration Effort:
┌─────────────────────┐ ┌─────────────────────────┐
│ UriPoint │ │ Multiple Libraries │
├─────────────────────┤ ├─────────────────────────┤
│ - Single interface │ │ - Multiple interfaces │
│ - Unified config │ │ - Separate configs │
│ - Built-in testing │ │ - Custom testing needed │
│ - One installation │ │ - Multiple installs │
└─────────────────────┘ └─────────────────────────┘
Key Advantages of UriPoint:
-
Development Speed
- Single API to learn
- Consistent patterns
- Integrated tools
-
Maintenance
- Centralized updates
- Unified logging
- Single point of configuration
-
Integration
- Built-in protocol support
- Standardized error handling
- Common monitoring interface
-
Testing
- Comprehensive test suite
- Cross-protocol testing
- Integrated monitoring
While other libraries provide excellent functionality in their specific domains, UriPoint's value proposition lies in its unified approach to handling multiple protocols and endpoint management through a single, consistent interface.
This makes it particularly valuable for:
- Complex distributed systems
- Multi-protocol applications
- IoT ecosystems
- Streaming services
- Enterprise integrations
Others
There are several similar solutions, although none offers exactly the same range of functionality as UriPoint. Here are the most similar libraries:
- FastAPI (Python)
- Main features:
- REST API creation
- WebSocket support
- Automatic documentation
- Data validation
- Differences from UriPoint:
- Mainly focuses on HTTP/REST
- No support for MQTT, RTSP, etc.
- aiohttp (Python)
- Main features:
- Asynchronous HTTP connections
- WebSocket support
- HTTP client and server
- Differences:
- HTTP/WS protocols only
- No integrated endpoint management
- Mosquitto (with Python binding)
- Main features:
- MQTT broker
- IoT support
- Light and efficient
- Differences:
- MQTT protocol only
- No universal interface
- RabbitMQ (with pika for Python)
- Main features:
- Queuing system
- Various communication patterns
- AMQP support
- Differences:
- Focuses on queuing
- No support for other protocols
- ZeroMQ (pyzmq)
- Main features:
- Flexible communication
- Various communication patterns
- High performance
- Differences:
- Custom communication protocol
- Requires more configuration
- Flask-SocketIO
- Main features:
- WebSocket integration with Flask
- Event support
- Real-time communication
- Differences:
- Limited to HTTP and WebSocket
- No IoT support
- Django Channels
- Main features:
- WebSocket for Django
- Asynchronous communication
- ORM integration
- Differences:
- Limited to Django ecosystem
- No support for other protocols
2. Enterprise-Grade Architecture
System Components
┌────────────────────┐ ┌────────────────────┐
│ Load Balancer │ │ Service Mesh │
├────────────────────┤ ├────────────────────┤
│ - Round Robin │ │ - Service Discovery│
│ - Least Connection │ │ - Load Balancing │
│ - IP Hash │ │ - Circuit Breaking │
└────────────────────┘ └────────────────────┘
┌────────────────────┐ ┌────────────────────┐
│ Security Layer │ │ Monitoring │
├────────────────────┤ ├────────────────────┤
│ - Authentication │ │ - Health Checks │
│ - Authorization │ │ - Metrics │
│ - Encryption │ │ - Alerts │
└────────────────────┘ └────────────────────┘
Features
- High Availability
- Fault Tolerance
- Scalability
- Service Discovery
- Load Balancing
- Circuit Breaking
3. Advanced Management Capabilities
Administrative Tools
- CLI Interface
- Web Dashboard
- API Access
- Configuration Management
- Monitoring Tools
- Logging System
Management Features
┌──────────────────────────────────────────┐
│ Management Dashboard │
├───────────────┬───────────────┬──────────┤
│ Endpoints │ Protocols │ Metrics │
├───────────────┼───────────────┼──────────┤
│ Configuration │ Security │ Logging │
├───────────────┼───────────────┼──────────┤
│ Monitoring │ Analytics │ Alerts │
└───────────────┴───────────────┴──────────┘
4. Integration Framework
Integration Capabilities
- Service Integration
- API Gateway
- Protocol Bridge
- Data Transformation
- Message Routing
- Event Processing
Integration Patterns
┌───────────────────┐ ┌────────────────────┐
│ API Gateway │ │ Event Processing │
├───────────────────┤ ├────────────────────┤
│ - Routing │ │ - Event Sourcing │
│ - Transformation │ │ - Stream Processing│
│ - Authentication │ │ - Message Queuing │
└───────────────────┘ └────────────────────┘
5. Development Environment
Development Tools
- SDK Support
- API Documentation
- Testing Framework
- Debug Tools
- Performance Profiling
- Development Console
Testing Capabilities
┌────────────────────┐ ┌────────────────────┐
│ Unit Testing │ │ Integration Tests │
├────────────────────┤ ├────────────────────┤
│ - Protocol Tests │ │ - End-to-End Tests │
│ - Component Tests │ │ - Load Tests │
│ - Mock Services │ │ - Stress Tests │
└────────────────────┘ └────────────────────┘
6. Use Cases
Enterprise Applications
- Microservices Architecture
- Distributed Systems
- Cloud Applications
- IoT Platforms
- Streaming Services
Industry Solutions
┌────────────────────┐ ┌────────────────────┐
│ IoT Platform │ │ Media Streaming │
├────────────────────┤ ├────────────────────┤
│ - Device Mgmt │ │ - Live Streaming │
│ - Data Collection │ │ - VOD Services │
│ - Monitoring │ │ - CDN Integration │
└────────────────────┘ └────────────────────┘
┌────────────────────┐ ┌────────────────────┐
│ E-Commerce │ │ Finance Apps │
├────────────────────┤ ├────────────────────┤
│ - API Gateway │ │ - Real-time Data │
│ - Order Processing │ │ - Transactions │
│ - Inventory Mgmt │ │ - Security │
└────────────────────┘ └────────────────────┘
7. Operational Features
Operations Management
- Deployment Automation
- Configuration Management
- Monitoring and Alerting
- Log Management
- Backup and Recovery
- Security Management
Operational Tools
┌────────────────────┐ ┌────────────────────┐
│ Deployment │ │ Monitoring │
├────────────────────┤ ├────────────────────┤
│ - CI/CD Pipeline │ │ - Health Checks │
│ - Auto Scaling │ │ - Performance │
│ - Load Balancing │ │ - Resource Usage │
└────────────────────┘ └────────────────────┘
Advantages Over Point Solutions
Unified Platform Benefits
-
Simplified Management
- Single interface for all protocols
- Centralized configuration
- Unified monitoring
-
Cost Efficiency
- Reduced training needs
- Lower maintenance overhead
- Simplified licensing
-
Enhanced Security
- Consistent security policies
- Unified authentication
- Centralized monitoring
-
Better Performance
- Optimized communication
- Reduced overhead
- Integrated caching
Integration Advantages
- Seamless Protocol Integration
- Unified Configuration Management
- Consistent Security Framework
- Centralized Monitoring
- Standardized Development Patterns
License
This project is licensed under the terms of the LICENSE file in the project root.
Changelog
See CHANGELOG.md for version history and updates.
Contributing
We welcome contributions! See our Contributing Guide for details on:
- Code style
- Pull request process
- Development setup
- Testing requirements
Support
- Support: Community support via GitHub Issues
- Documentation: Full docs at docs.uripoint.com
- Updates: See CHANGELOG.md
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
File details
Details for the file uripoint-1.9.6.tar.gz
.
File metadata
- Download URL: uripoint-1.9.6.tar.gz
- Upload date:
- Size: 123.9 kB
- Tags: Source
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/5.1.1 CPython/3.11.3
File hashes
Algorithm | Hash digest | |
---|---|---|
SHA256 | 50d3ffe5bbfacdd9e4fc2743eb1b7946a769dfc09d40bbf5c502e2eac88599fb |
|
MD5 | 2e7a0033ef930cf8354dbffeb2d9c84d |
|
BLAKE2b-256 | d72e7589bb5a7d321ec0d427a31889585f1b174712314c292e851d7aa815fcab |
File details
Details for the file uripoint-1.9.6-py3-none-any.whl
.
File metadata
- Download URL: uripoint-1.9.6-py3-none-any.whl
- Upload date:
- Size: 25.3 kB
- Tags: Python 3
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/5.1.1 CPython/3.11.3
File hashes
Algorithm | Hash digest | |
---|---|---|
SHA256 | 87bcf6e21588eea67c5d3630e1449d6d8a74514f3de4d378b184d65e6eaa4fab |
|
MD5 | 420b4c3e059e2e75d21a0d3f8a671a60 |
|
BLAKE2b-256 | f4768f6b42a8fa9a2723df88e678555f951138d9f579e0d4d505702ba828f854 |