Hijiki

Python Rabbit wrapper library to simplify to use Exchanges and Queues with decorators

Configurations

Hijiki uses environment variables to configure connection with BROKER.

COMMON

BROKER_PWD
BROKER_USERNAME

For single server

BROKER_PORT
BROKER_SERVER

For cluster server

BROKER_CLUSTER_SERVER

if BROKER_CLUSTER_SERVER is present the priority is generate URI using this list of servers, and not use the sever for single server. To user multiples server from cluster is necessary this environment variable has a list of server with port separates with comma.

Ex:

serverA:5672,serverB:5673

generating for uri connection string the value above

amqp://usr:password@server:5672;amqp://usr:password@serverB:5672;

The user and password is the same for all servers.

If server is not present, even BROKER_CLUSTER_SERVER, the connection url will be a default, and to others configs will be changed for "teste".

How to use

Publisher

The example demonstrate how to publish a simple message to topic "teste1_event" with a json message:

pub = Publisher("localhost", "rabbitmq", "rabbitmq", 5672, heartbeat=30)
pub.publish_message('teste1_event', '{"value": "Esta é a mensagem"}')

Consumer

Consumer uses a configuration to define QUEUES and Exchanges and the consumer is a decorator for the queue.

from hijiki.broker.hijiki_rabbit import HijikiQueueExchange, HijikiRabbit

qs = [HijikiQueueExchange('teste1', 'teste1_event'), HijikiQueueExchange('teste2', 'teste2_event')]
gr = HijikiRabbit().with_queues_exchange(qs) \
    .with_username("rabbitmq") \
    .with_password("rabbitmq") \
    .with_host("localhost") \
    .with_port(5672) \
    .with_heartbeat_interval(30)\
    .build()

class MyConsumer():
    @gr.task(queue_name='teste1')
    def my_consumer(data):
        print(f"consumer 1 executed with data : {data}")

    @gr.task(queue_name='teste2')
    def my_consumer2(data):
        print(f"consumer 2  executed with data : {data}")

if __name__ == '__main__':
    MyConsumer()
    gr.run()

asengardeon/hijiki