User Guide

Anitya attempts to select reasonable default settings for projects and in many cases, these should work just fine. However, there are millions of software projects out there and some do not follow common release methods. In those cases, Anitya offers more flexible tools.

Creating a New Project

You can create new projects in the web interface if you have logged in. When you create a new project, you must select a backend to fetch the version with. If the project is released in several places, please use the backend for the language ecosystem. For example, if a project is hosted on GitHub and publishes releases to PyPI, use the PyPI backend.


Occasionally projects stop publishing releases to their ecosystem’s package index, but continue to tag releases. In those cases, it’s fine to use the backend for the source hosting service (GitHub, BitBucket, etc.) the project is using.


The backend of a project tells Anitya how to retrieve the versions of the project.

The backends available are:

  • for perl projects hosted on CPAN
  • for R projects hosted on CRAN
  • for crates hosted on
  • for projects hosted on the Debian ftp
  • for Drupal6 modules hosted on
  • for Drupal7 modules hosted on
  • for projects whose release archives are provided in a folder basic apache folder or modified one.
  • for projects hosted on /
  • for projects hosted on using Github v4 API <>
  • for projects hosted on GitLab server <>_. This backend is using GitLab API v4 <>_
  • for projects hosted on
  • for projects hosted on
  • for projects hosted on
  • for projects hosted on
  • for projects hosted on
  • for projects hosted on
  • for projects hosted on
  • for projects hosted on
  • for projects hosted on
  • for projects hosted on
  • for projects hosted on
  • for projects hosted on

If your project cannot be used with any of these backend you can always try the custom backend.

  • for projects who require a more flexible way of finding their version.

The custom backend requires two arguments:

  • version_url the url of the page where the versions information can be found, for example for banshee that would be their download page


It’s possible to provide a “glob” for projects that place their tarballs in multiple directories. For example, gcc uses*/ to find the tarballs in each version directory.

  • regex a regular expression using the Python re syntax to find the releases on the version_url page.


In most cases, you can set the regex to DEFAULT which will make anitya use its default regular expression:

<package name>(?:[-_]?(?:minsrc|src|source))?[-_]([^-/_\s]+?)(?i)(?:[-_](?:minsrc|src|source|asc|release))?\.(?:tar|t[bglx]z|tbz2|zip)

Project Name

The project name should match the upstream project name. Duplicate project names are allowed as long as the projects are not part of the same ecosystem. That is, you can have two projects called msgpack, but you cannot have two projects called msgpack that are both in the PyPI ecosystem.


When project is not part of any ecosystem, duplicate projects are detected based on the homepage of project.

Version Prefix

The version prefix can be used to retrieve the exact version number when the upstream maintainer prefixes its versions.

For example, if the project’s version are: foo-1.2, you can set the version prefix to foo- to tell Anitya how to get the version 1.2.


It’s common for projects to prefix their source control tags with a v when making a release. Anitya will automatically strip this from versions it finds.

More concrete examples:

  • junit tags are of the form: r<version>, to retrieve the version, one can set the version prefix to r.
  • fdupes tags are of the form fdupes-<version>, for this project, the version prefix can be set to fdupes-.

Regular Expressions

Sometimes you need to use a custom regular expression to find the version on a page. Anitya accepts user-defined regular expressions using the Python re syntax.

The simplest way to check your regular expression is to open a python shell and test it.

Below is an example on how it can be done:

>>> url = ''
>>> import requests
>>> import re
>>> text = requests.get(url).text
>>> re.findall(' ([\d]*)\.', text)
>>> re.findall(' ([\d\.-]*)\.', text)

If you prefer graphical representation you can use Debuggex.

The regular expression ([\d\.]*)\. can then be provided to anitya and used to find the new releases.


Only the captured groups are used as version, delimited by dot. For example: 1_2_3 could be captured by regular expression (\d)_(\d)_(\d). This will create version 1.2.3.

Integrating with Anitya


fedmsg is a message bus. In other words it is a system that allows for the sending and receiving of notifications between applications. For anitya, every action made on the application is announced/broadcasted on this bus, allowing anyone listening to it to act immediately instead of (for example) polling hourly all the data, looking for changes, and acting then. For the full list of messages Anitya sends, see the fedmsg topic documentation.

To start receiving fedmsg messages from anitya, it is as simple as:

  • install fedmsg the way you want:

On Fedora

dnf install fedmsg

On Debian

apt-get install fedmsg

Via pip

pip install fedmsg
  • in the configuration file: /etc/fedmsg.d/, make sure you activate the anitya endpoint

    "anitya-public-relay": [

From python

import fedmsg

# Yield messages as they're available from a generator
for name, endpoint, topic, msg in fedmsg.tail_messages():
    print topic, msg

From the shell

$ fedmsg-tail --really-pretty

Reporting Issues

You can report problems on our issue tracker. The source code is also available on GitHub. The development team hangs out in #fedora-apps on the freenode network. Please do stop by and say hello.