Skip to main content

ModelQueue is an Apache2 licensed task queue based on Django models.

For example, in appname/models.py:

import modelqueue
from django.db import models

class Task(models.Model):
    data = models.TextField()
    status = modelqueue.StatusField(
        # ^-- Just a models.BigIntegerField
        db_index=True,
        # ^-- Index for faster queries.
        default=modelqueue.Status.waiting,
        # ^-- Waiting state is ready to run.
    )

And in appname/management/commands/process_tasks.py:

import modelqueue, time
from django.core.management.base import BaseCommand
from .models import Task

class Command(BaseCommand):

    def handle(self, *args, **options):
        while True:
            task = modelqueue.run(
                Task.objects.all(),
                # ^-- Queryset of models to process.
                'status',
                # ^-- Field name for model queue.
                self.process,
                # ^-- Callable to process model.
            )
            if task is None:
                time.sleep(1)
                # ^-- Bring your own parallelism/concurrency.

    def process(self, task):
        pass  # Process task models.

And in appname/admin.py:

class TaskAdmin(admin.ModelAdmin):
    actions = [*modelqueue.admin_actions('status')]
    # ^-- Change task status in admin.
    list_filter = [
        modelqueue.admin_list_filter('status'),
        # ^-- Filter tasks in admin by queue state.
    ]

    def get_changeform_initial_data(self, request):
        # v-- Automatically fill in status field when adding a new task.
        return {'status': int(modelqueue.Status.waiting())}

ModelQueue is a hazardous project. It takes a bad idea and makes it easy and effective. You may come to regret using your database as a task queue but it won’t be today!

Testimonials

“I didn’t design relational database systems for this.” ~ Edgar Codd

“Well, at least you’re using transactions.” ~ Jim Gray

“You have successfully ignored most of what’s important in queueing theory.” ~ Agner Erlang

Does your company or website use ModelQueue? Send us a message and let us know.

Features

  • Pure-Python

  • Supports Django’s admin interface

  • Tasks can be retried, aborted, and canceled

  • Supports multiple attempts per task

  • Bring your own parallelism with threading, multiprocessing, or asyncio

  • Performance matters (add a single 64-bit field to models)

  • Fully documented

  • 100% test coverage

  • Years of stress testing in production

  • Developed on Python 3.10

  • Compatible with all Django versions

  • Tested on CPython 3.6, 3.7, 3.8, 3.9, 3.10

  • Tested on Linux, Mac OS X, and Windows

https://github.com/grantjenks/django-modelqueue/workflows/integration/badge.svg https://github.com/grantjenks/django-modelqueue/workflows/release/badge.svg

Quickstart

Installing ModelQueue is simple with pip:

$ python -m pip install modelqueue

You can access documentation in the interpreter with Python’s built-in help function:

>>> import modelqueue
>>> help(modelqueue)

User Guide

For those wanting more details, this part of the documentation describes introduction, benchmarks, development, and API.

Reference and Indices

ModelQueue License

Copyright 2022 Grant Jenks

Licensed under the Apache License, Version 2.0 (the “License”); you may not use this file except in compliance with the License. You may obtain a copy of the License at

http://www.apache.org/licenses/LICENSE-2.0

Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on an “AS IS” BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the specific language governing permissions and limitations under the License.

Metadata

Release files for modelqueue 2.2.1

For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.

Source distribution (sdist)

Source distribution for modelqueue 2.2.1
File Size Uploaded Provenance
modelqueue-2.2.1.tar.gz 8.5 kB Without provenance Details

Built distribution (wheel)

Table of built distributions (wheels) for modelqueue 2.2.1
File Interpreter ABI Platform
modelqueue-2.2.1-py3-none-any.whl Python 3 none any Details

Total release size: 17.3 kB

Release files / modelqueue-2.2.1.tar.gz

Download URL modelqueue-2.2.1.tar.gz
Size 8.5 kB
Tags Source
SHA-256 checksum
How to use checksums
11d325719933803dc0d79bce00cd4e2ab17224a0d6932a21bc46ae03fad275fd
BLAKE2b-256 checksum
How to use checksums
b56b1e6b48b717846e317685058144cd250adfda360ca61e93ac75d0549bb241
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
No
Uploaded via twine/4.0.0 CPython/3.10.2

Release files / modelqueue-2.2.1-py3-none-any.whl

Download URL modelqueue-2.2.1-py3-none-any.whl
Size 8.8 kB
Tags Python 3
SHA-256 checksum
How to use checksums
31a270939d997a1bf2587c4adfd8ad850438d5f0839b7c2d87f95a43d3d40ad3
BLAKE2b-256 checksum
How to use checksums
0c7c156b3c909c136cd05b6732949a87d1f3dc3e2de7f623aa814c4c2086939f
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
No
Uploaded via twine/4.0.0 CPython/3.10.2

Security for modelqueue 2.2.1

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

No provenance found

None of the files in this release were uploaded with provenance metadata.

A note on package security

PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.

To learn more about protecting yourself from malicious actors, visit our package security documentation.

Anthropic, PBC Visionary sponsor Bloomberg Visionary sponsor Hudson River Trading Visionary sponsor Meta Visionary sponsor NVIDIA Visionary sponsor Microsoft Sustainability sponsor Depot Continuous Integration AWS Cloud computing and Security Sponsor Datadog Monitoring Fastly CDN Google Download Analytics Sentry Error logging StatusPage Status page