Source code for our project «Object-Centric Comparative Process Mining» in the «Process Discovery Using Python» lab offered at RWTH Aachen University in WS21.
The code comes with a Dockerfile to build and run the application inside a docker container To build the container run
docker build --rm -t occpm:latest .
After the container is build the webapp can be run using
docker run -p 8000:8000 occpm
When using conda you can create an environment from the provided environment.yml by:
$ conda env create -f environment.yml
$ conda activate occpm
If any dependency is missing please add it and only it to environment.yml without specifying a version number and please do not use conda env export > environment.yml since one should avoid hard references to specific package versions unless explicitly necessary.
Aside from the python packages the application also requires graphviz to be installed on your system, on most Linux distros it is included by default, it is also included in the docker build.
If the environment created from environment.yml does not work you can try using environment_static.yml(which was created using conda env export > environment_static.yml) to obtain an environment with the exact versions used during development, however this is not guaranteed to work across OS's and if possible one should not use out-of-date versions of packages.
To run the application execute:
$ python manage.py runserver
To properly configure the databases behind django whenever the INSTALLED_APPS or models of apps are changed, the following commands should be executed
python manage.py makemigrations
python manage.py migrate
When one wants to update the environment, either because some packages were added, or just to get the newest versions of packages without recreating the environment execute:
conda env update --file environment.yml --prune
This project uses black's code styling. It is included as a dependency in the provided environment and can be run using:
black .
Alternatively if you want to avoid the hassle of remembering to run black all the time a pre-commit configuration is provided as well, which will run black before every commit. You can enable this configuration using:
pre-commit install
A test suit is included that can be run using
python manage.py test apps