pgqueue 0.6

Light PgQ Framework - queuing system for PostgreSQL

This module provides a convenient Python API to integrate PostgreSQL PgQ features with any Python application.

Presentation of PgQ

(from SkyTools README)

PgQ is a queuing system written in PL/pgSQL, Python and C code. It is based on snapshot-based event handling ideas from Slony-I, and is written for general usage.

PgQ provides an efficient, transactional queueing system with multi-node support (including work sharing and splitting, failover and switchover, for queues and for consumers).

Rules:

• There can be several queues in a database.

• There can be several producers than can insert into any queue.

• There can be several consumers on one queue.

• There can be several subconsumers on a consumer.

PgQ is split into 3 layers: Producers, Ticker and Consumers.

Producers and Consumers respectively push and read events into a queue. Producers just need to call PostgreSQL stored procedures (like a trigger on a table or a PostgreSQL call from the application). Consumers are frequently written in Python, but any language able to run PostgreSQL stored procedures can be used.

Ticker is a daemon which splits the queues into batches of events and handle the maintenance of the system.

The PgQueue module

This module provides Python functions and classes to write Producers and Consumers. It contains also a Python implementation of the Ticker engine, which mimics the original C Ticker from SkyTools: it splits batches of events, and execute maintenance tasks.

Installation

Prerequisites:

• Python >= 2.6 or Python 3

• psycopg2 is automatically installed as a dependency

• (on the server) the PgQ extension version >= 3.1

On Debian / Ubuntu you will add the PostgreSQL APT repository, then install the package postgresql-x.x-pgq3 depending on the PostgreSQL version.

Finally create the extension in the database:

CREATE EXTENSION IF NOT EXISTS pgq;

You can install the pgqueue module into your environment.

pip install --update pgqueue

Example usage

You need to run the Ticker permanently. If the Ticker is off, the events will be stored into the queues, but no batch will be prepared for the consumers, and event tables will grow quickly.

For the Ticker, you have the choice between the optimized pgqd multi-database ticker written in C, and part of SkyTools, or use the simpler Python implementation provided with this module:

python -m pgqueue 'host=127.0.0.1 port=5432 user=jules password=xxxx dbname=test_db'

Let’s create a new queue, and register a consumer:

conn = psycopg2.connect("dbname=test user=postgres")
conn.autocommit = True
cursor = conn.cursor()

first_q = pgqueue.Queue('first_queue')
first_q.create(cursor, ticker_max_lag='4 seconds')

consum_q = pgqueue.Consumer('first_queue', 'consumer_one')
consum_q.register(cursor)

We’re ready to produce events into the queue, and consume events later in the application:

first_q.insert_event(cursor, 'announce', 'Hello ...')
first_q.insert_event(cursor, 'announce', 'Hello world!')

# ... wait a little bit

conn.autocommit = False
for event in consum_q.next_events(cursor, commit=True):
    print(event)

You can browse the source code for advanced usage, until we write more documentation (contributions are welcomed).

Also refer to the upstream documentation for more details.

Credits

PgQ is a PostgreSQL extension which is developed by Marko Kreen. It is part of SkyTools, a package of tools in use in Skype for replication and failover.

SkyTools embeds also a pgq Python framework which provides a slightly different API.

Links

• Fork me on GitHub