Spark is a Python3 project pattern. It can be considered as a starting point for new Python3 applications using state-of-the-art technologies.

• Runnable as python virtual env and system wide
• Unit-Tests (using tox, unittest)
• Code coverage
• CI-Support (Travis CI)
• Debian packaging
• Documentation generation
• Git version awareness

The only requirement for building the application is

• python3.5
• curl

Everything else is being downloaded automatic to a Python virtual environment.

$ git clone https://github.com/masmu/spark.git
$ cd spark/

$ make

$ bin/spark

$ make test

The test suite includes unit tests, linting, code coverage and manifest checking. You can find a HTML coverage report in the htmlcov/ folder.

$ make docs

You can find the generated documentation in the docs/ folder of the source directory.

You need to setup the debian packaging environment before: On Ubuntu you can do this via:

$ sudo apt-get install pbuilder debootstrap devscripts lintian
$ sudo pbuilder create --debootstrapopts --variant=buildd

After that you should enable Universe support for pbuilder. For that add the following to your ~/.pbuilderrc

COMPONENTS="main restricted universe multiverse"

if [ "$DIST" == "squeeze" ]; then
    echo "Using a Debian pbuilder environment because DIST is $DIST"
    COMPONENTS="main contrib non-free"
fi

After adding those new sources run the following command to update pbuilder:

$ sudo pbuilder update --override-config

Now you should be able to create Debian packages via:

$ make deb