Skip to content


Subversion checkout URL

You can clone with
Download ZIP
Django Dash.
Python JavaScript
branch: master

Fetching latest commit…

Cannot retrieve the latest commit at this time

Failed to load latest commit information.


If the mountain will not come to Mohamet, Mohamet must go to the mountain.

Journeyman is an attempt to create a build server environment that is reasonably easy to use and set up. Focussing on the notion that different applications need different build steps, Journeyman employs a domain specific language called "journey" to describe a project's build process.

Journeyman currently focusses on use with Python projects, as it will do the following:

  1. connect to a worker node via ssh
  2. create a virtualenv repository
  3. check out a given repository (currently git only)
  4. run build steps as defined in journey file
  5. archive virtualenv
  6. determine build result from xunit-test conforming xml

Journey files consist of named rules and steps that are formulated in YAML.

Let's take a look into such a journey file:

- dependencies
- install
- test
- fetch_results

- spielwiese/requirements.txt

- cd spielwiese && python install

- cd spielwiese && nosetests --with-xunit

- spielwiese/nosetests.xml

The journey build runner uses this file to determine the steps in a build. Journeyman will always look for a "build"-rule, if it does not exist, try to execute a default build rule:

- dependencies
- install
- test
- testresults

To execute a rule, journeyman will follow the steps in the build rule and run the associate step. The first rule to run obviously is "dependencies".

- spielwiese/requirements.txt

Rules can optionally use square brackets to specify a plugin to run the steps defined in them. When no plugin was specified, it is assumed that the steps are shell commands and will by run by the plugin "run_commands".

Journeyman comes with the following plugins:

  • run_commands
    • Runs every single command step, if one of the steps fails i.e. return code != 0, the build fails and execution ends immediately
  • run_tests
    • Since most testing tools return non-0 results for failing tests and we might want to run multiple test suites, this plugin keeps on executing all steps even if they fail.
  • fetch_pip_dependencies
    • Given one path to a dependency file per step, this plugin will run install -r $file
  • fetch_xunit_results
    • Given one path to an xml unittest file per step, this plugin will download the file and thus allow to see the unittest results in the build results view.


  • Add build configuration to project repository and let journeyman take care of the rest
  • When you don't have access to the repository, the build server can save your journey script and use it when appropriate
  • Support for xUnit XML test results
  • Support for future enhancements: new build steps, new result processing steps
Something went wrong with that request. Please try again.