Skip to content

A Go queue management library designed for robust and scalable job processing, leveraging Asynq and Redis. It supports advanced features such as error handling, retries, priority queues, and rate limiting, ideal for tasks from simple to complex in distributed systems.

Notifications You must be signed in to change notification settings

kaptinlin/queue

Repository files navigation

Golang Queue Processing Library

This library offers a robust and flexible solution for managing and processing queued jobs in Go applications. Built on top of the Asynq task processing library, which uses Redis for storage, it provides advanced features like custom error handling, retries, priority queues, rate limiting, and job retention. Whether you're building a simple task runner or a complex distributed system, this library is designed to meet your needs with efficiency and ease. It also supports the setup of multiple workers across different machines, allowing for scalable and distributed job processing.

Getting Started

Installation

Ensure your Go environment is ready (requires Go version 1.21.4 or higher), then install the library:

go get -u github.com/kaptinlin/queue

Configuring Redis

Set up your Redis connection with minimal hassle:

import "github.com/kaptinlin/queue"

redisConfig := queue.NewRedisConfig(
    queue.WithRedisAddress("localhost:6379"),
    queue.WithRedisDB(0),
    queue.WithRedisPassword("your_password"),
)

Client Initialization

Create a client using the Redis configuration:

client, err := queue.NewClient(redisConfig)
if err != nil {
    log.Fatalf("Error initializing client: %v", err)
}

Job Enqueueing

Enqueue jobs by specifying their type and a structured payload for clear and concise data handling:

type EmailPayload struct {
    Email   string `json:"email"`
    Content string `json:"content"`
}

jobType := "email:send"
payload := EmailPayload{Email: "[email protected]", Content: "Welcome to our service!"}

_, err = client.Enqueue(jobType, payload, queue.WithDelay(5*time.Second))
if err != nil {
    log.Printf("Failed to enqueue job: %v", err)
}

Alternatively, for direct control over job configuration, use a Job instance:

job := queue.NewJob(jobType, payload, queue.WithDelay(5*time.Second))
if _, err := client.EnqueueJob(job); err != nil {
    log.Printf("Failed to enqueue job: %v", err)
}

This approach allows you to specify additional job options such as execution delay, directly within the Job object.

Handling Jobs

Define a function to process jobs of a specific type. Utilize the EmailPayload struct for type-safe payload handling:

func handleEmailSendJob(ctx context.Context, job *queue.Job) error {
    var payload EmailPayload
    if err := job.DecodePayload(&payload); err != nil {
        return fmt.Errorf("failed to decode payload: %w", err)
    }

    log.Printf("Sending email to: %s with content: %s", payload.Email, payload.Content)
    // Implement the email sending logic here.
    return nil
}

To achieve scalable and distributed job processing, you can register your function and start workers on different machines. Each worker independently processes jobs enqueued by the client:

worker, err := queue.NewWorker(redisConfig, queue.WithWorkerQueue("default", 1))
if err != nil {
    log.Fatalf("Error creating worker: %v", err)
}

err = worker.Register("email:send", handleEmailSendJob)
if err != nil {
    log.Fatalf("Failed to register handler: %v", err)
}

if err := worker.Start(); err != nil {
    log.Fatalf("Failed to start worker: %v", err)
}

Graceful Shutdown

Ensure a clean shutdown process:

func main() {
    // Initialization...

    c := make(chan os.Signal, 1)
    signal.Notify(c, os.Interrupt)
    <-c

    if err := client.Stop(); err != nil {
        log.Fatalf("Failed to stop client: %v", err)
    }
    worker.Stop()
}

Advanced Features

Learn more about the library's advanced features by exploring our documentation on:

Contributing

We welcome contributions! Please submit issues or pull requests on GitHub.

License

This library is licensed under the MIT License.

Credits

Special thanks to the creators of neoq and Asynq for inspiring this library.

About

A Go queue management library designed for robust and scalable job processing, leveraging Asynq and Redis. It supports advanced features such as error handling, retries, priority queues, and rate limiting, ideal for tasks from simple to complex in distributed systems.

Topics

Resources

Stars

Watchers

Forks

Packages

No packages published