Yet another Python API for Jenkins.
Buy one now, you get a Docker image FOR FREE!
This library is built on top of python-jenkins, and his purpose is just that of offering a simple(r), more "pythonic" interface.
Build the image with (from the repo root directory):
docker build -f Dockerfile -t pumpkins .
Launch the container with:
docker run -h pumpkins --rm -it -p 8080:8080 -v $PWD:/staging pumpkins # $PWD:/staging or whatever
- Wait for Jenkins to startup and then access the instance at localhost:8080
- In the container run
python /staging/pumpkins.pyto verify that everything is working properly.
Jenkins will be configured with an 'admin:admin' user, the plugins to install can be specified in a 'plugins.txt' file that can be then mounted into the container aside to the pumpkins library itself.
The starting procedure will try to guess the position of the mounted endpoint inspecting the current
mounted volumes, and fallback to '/staging' eventually, see the start.sh script for more details.
If the environment variable '$INTERACTIVE' is set to any non empty value the Jenkins configuration can be executed manually, the terminal will show a password to Copy&Paste into the Web page and then it will be possible to configure Jenkins as needed:
docker run -h pumpkins -e INTERACTIVE=1 --rm -it -p 8080:8080 -v $PWD:/staging pumpkins
It is possible to pass any command to the container, if none is given a bash shell will be provided. On the bash shell you have ipython and python-jenkins ready to be used, (as well as VIM, you're welcome) so doing...
jenkins@pumpkins:~$ ipython3
In[1]: import pumpkins
In[2]: host = pumpkins.Host('http://localhost:8080', username='admin', password='admin')
In[3]: print(host.me)
...should do the work.
The library comes with a builtin simple unit test suite, it is possible to run it issuing a command like:
docker run --rm -it -v $PWD:/staging -w /staging pumpkins python3 -B -m pytest -p no:cacheprovider
(note that is not necessary to expose TCP/UDP ports for the tests to be executed).
The Docker image is used only for developing/purpose testing, as such it makes little sense to embed the Python library itself inside it, more useful is mounting the repository root folder into the container file system so to be able to edit the .py file on the fly.
Once logged into the container issuing export PYTHONPATH=$PYTHONPATH:/staging will save you from
having to type import sys; sys.append(...) every time you spawn a python/ipython shell.
The Docker image is built on top of the jenkins image (https://hub.docker.com/r/jenkins/jenkins) and as such it exposes the 50000 and 8080 TCP ports, only the 8080 is needed to access the web interface so I would not bother exposing the 50000 as well.
The Jenkins docker image
The Jenkins Python library
Thanks to Viktor Farcic for his great blog post on how to deploy Jenkins unsupervised.
