mindtrace-jobs 0.9.4

Job management for Mindtrace

Mindtrace Jobs

A job queue system that works with different backends (local, Redis, RabbitMQ).

Backends

• Local
• Redis
• RabbitMQ

Setup Redis:

# Local installation
redis-server

# Using Docker
docker run -d --name redis -p 6379:6379 redis:latest

# Test connection
redis-cli ping 

RabbitMQ Backend

Setup RabbitMQ:

# Using Docker (recommended)
docker run -d --name rabbitmq -p 5672:5672 -p 15672:15672 -e RABBITMQ_DEFAULT_USER=user -e RABBITMQ_DEFAULT_PASS=password rabbitmq:3-management

Core Components:

• Consumer - Base class for processing jobs
• Orchestrator - Manages job queues and routing
• Job, JobSchema - Job data structures
• LocalClient, RedisClient, RabbitMQClient - Backend implementations

Architecture

graph TD
    Consumer["Your Consumer<br/>MathsConsumer('maths_operations')"] 
    Orchestrator["Orchestrator"]
    Schema["Job Schema<br/>(maths_operations)"]
    Job["Job Instance"]
    Queue["Queue<br/>(maths_operations)"]
    
    LocalClient["LocalClient"]
    RedisClient["RedisClient<br/>(requires Redis server)"] 
    RabbitMQClient["RabbitMQClient<br/>(requires RabbitMQ server)"]
    
    Consumer --> Orchestrator
    Schema --> Orchestrator
    Job --> Orchestrator
    Orchestrator --> Queue
    
    Orchestrator -.-> LocalClient
    Orchestrator -.-> RedisClient
    Orchestrator -.-> RabbitMQClient
    
    Queue --> Consumer

Basic Example

from mindtrace.jobs import Orchestrator, LocalClient, Consumer, JobSchema, job_from_schema
from pydantic import BaseModel

# Set up the orchestrator with local backend
orchestrator = Orchestrator(LocalClient())

# Define your job input/output models (inherit from BaseModel directly)
class MathsInput(BaseModel):
    operation: str = "add"
    a: float = 2.0
    b: float = 1.0

class MathsOutput(BaseModel):
    result: float = 0.0
    operation_performed: str = ""

schema = JobSchema(name="maths_operations", input_schema=MathsInput, output_schema=MathsOutput)
orchestrator.register(schema)

# Create a consumer
class MathsConsumer(Consumer):
    def run(self, job_dict: dict) -> dict:
        # Access input data from the dict
        input_data = job_dict.get('input_data', {})
        operation = input_data.get('operation', 'add')
        a = input_data.get('a')
        b = input_data.get('b')
        
        # Your processing logic here
        if operation == "add":
            result = a + b
        elif operation == "multiply":
            result = a * b
        elif operation == "power":
            result = a ** b
        else:
            raise ValueError(f"Unknown operation: {operation}")
        
        return {
            "result": result,
            "operation_performed": f"{operation}({a}, {b}) = {result}"
        }

# Connect and consume jobs
consumer = MathsConsumer("maths_operations")
consumer.connect(orchestrator)

# Add a job to the queue
job = job_from_schema(schema, MathsInput(
    operation="multiply",
    a=7.0,
    b=3.0
))
orchestrator.publish("maths_operations", job)

# Process jobs
consumer.consume(num_messages=1)

Using Different Backends

Redis Backend

from mindtrace.jobs import RedisClient

redis_backend = RedisClient(host="localhost", port=6379, db=0)
orchestrator = Orchestrator(redis_backend)

consumer = MathsConsumer("maths_operations")
consumer.connect(orchestrator)
consumer.consume()

RabbitMQ Backend

from mindtrace.jobs import RabbitMQClient

rabbitmq_backend = RabbitMQClient(
    host="localhost", 
    port=5672, 
    username="user", 
    password="password"
)
orchestrator = Orchestrator(rabbitmq_backend)

consumer = MathsConsumer("maths_operations")
consumer.connect(orchestrator)
consumer.consume()

Backend Switching

The job system supports seamless switching between backends without changing your consumer code:

# Development: Use local backend
backend = LocalClient()

# Testing: Switch to Redis
backend = RedisClient(host="localhost", port=6379, db=0)

# Production: Switch to RabbitMQ  
backend = RabbitMQClient(host="localhost", port=5672, username="user", password="password")

# Same orchestrator and consumer code works with any backend
orchestrator = Orchestrator(backend)
consumer = MathsConsumer("maths_operations")
consumer.connect(orchestrator)
consumer.consume()

Automatic Backend Detection

When you connect a consumer, the orchestrator automatically detects the backend type and creates the appropriate consumer backend:

consumer = MathsConsumer("maths_operations")
consumer.connect(orchestrator)  # Automatically detects backend type

Implementation:

1. The Orchestrator detects its backend type using type(self.backend).__name__
2. Creates the corresponding consumer backend:
   • LocalClient → LocalConsumerBackend
   • RedisClient → RedisConsumerBackend
   • RabbitMQClient → RabbitMQConsumerBackend

def create_consumer_backend_for_schema(self, schema: JobSchema) -> ConsumerBackendBase:
    backend_type = type(self.backend).__name__
    if backend_type == "LocalClient":
        return LocalConsumerBackend(queue_name, self)
    elif backend_type == "RedisClient":
        return RedisConsumerBackend(queue_name, self, poll_timeout=5)
    elif backend_type == "RabbitMQClient":
        return RabbitMQConsumerBackend(queue_name, self, prefetch_count=1)

Priority Queues

Local Priority Queue

# Declare priority queue
backend = LocalClient()
orchestrator = Orchestrator(backend)
backend.declare_queue("priority_tasks", queue_type="priority")

# Publish with different priorities (higher numbers = higher priority)
orchestrator.publish("priority_tasks", urgent_job, priority=10)
orchestrator.publish("priority_tasks", normal_job, priority=5)
orchestrator.publish("priority_tasks", background_job, priority=1)

Redis Priority Queue

# REQUIRES: Redis server running
redis_backend = RedisClient(host="localhost", port=6379, db=0)
orchestrator = Orchestrator(redis_backend)
redis_backend.declare_queue("redis_priority", queue_type="priority")

# Higher priority jobs processed first
orchestrator.publish("redis_priority", critical_job, priority=100)
orchestrator.publish("redis_priority", normal_job, priority=50)

RabbitMQ Priority Queue

# REQUIRES: RabbitMQ server running
rabbitmq_backend = RabbitMQClient(host="localhost", port=5672, username="user", password="password")
orchestrator = Orchestrator(rabbitmq_backend)
# RabbitMQ supports max priority 0-255
rabbitmq_backend.declare_queue("rabbitmq_priority", max_priority=255)

orchestrator.publish("rabbitmq_priority", critical_job, priority=255)
orchestrator.publish("rabbitmq_priority", normal_job, priority=128)

API Reference

Consumer

class Consumer:
    def __init__(self, job_type_name: str)
    def connect(self, orchestrator: Orchestrator)
    def consume(self, num_messages: Optional[int] = None)
    def run(self, job_dict: dict) -> dict  # Implement this method

Orchestrator

class Orchestrator:
    def __init__(self, backend)
    def register(self, schema: JobSchema) -> str
    def publish(self, queue_name: str, job: Job, **kwargs) -> str
    def receive_message(self, queue_name: str) -> Optional[dict]
    def count_queue_messages(self, queue_name: str) -> int