Docker image with Jupyter sandbox based on Keboola Custom Python which is shared between transformations and sandboxes. Public image is available on Quay.
To run locally use the .env.template file to create an .env file. Create a python transformation configuration. Set the following variables:
KBC_CONFIG_ID- Transformation configuration (bucket) ID.KBC_ROW_ID- Transformation configuration row ID.KBC_CONFIG_VERSION- Version of the configuration to use.KBC_STORAGEAPI_URL- Connection URL.KBC_TOKEN- Connection token with access to configuration and storage tables referenced there.PACKAGES- optional JSON array with packages to preinstall, use[]to install no packages
Run docker-compose up, the notebook will be avaiable at http://localhost:8888. Password is test, can be changed in docker-compose.yml.