OGC Examples Jupyter Notebook

Available notebooks

This repository contains the following notebooks:

• Update pipeline example.ipynb: Example of a full data pipeline (from CSV to RDF) for the ISO19157-3 data quality measure properties.
• bblocks/: Notebooks related to the OGC Building Blocks
  • 01 Building Blocks basics.ipynb: Intro to using the Building Blocks.

Running the notebooks

Docker (recommended)

Since the notebook has external dependencies (mainly the ogc-na module), the easiest way to view the notebook is by using Docker.

After ensuring that you have a working Docker environment, run the following command:

docker run --rm --pull=always -p 8888:8888 -v "$(pwd)/work:/notebook/work" dockerogc/jupyter-notebooks

This will start the Notebook, making it available on port 8888. Check the console output for a link in the form http://127.0.0.1:8888/lab?token=abcdef123456; you will need to use it to access the Notebook. You can then open the Notebook file you want to use.

Notes:

• The --rm flag above will destroy the Notebook container once you are done with it.
• --pull=always guarantees that you are running the latest version of the Docker image.
• The -p 8888:8888 configuration is used to forward port 8888 on your machine to port 8888 in the container.
• -v "$(pwd)/work:/notebook/work" mounts a local (in the current directory) work subdirectory to the /notebook/work directory in the container, so that all files created when running the notebook will be available even after the Notebook is closed. The syntax maybe different for Windows environments.

Mounting other directories

Some notebooks may require additional directories to be mounted, such as a local copy of the OGC Building Blocks Register. You can add as many -v "/absolute/local/path:/notebook/notebook-path" parameters to the Docker command before dockerogc/jupyter-notebooks, replacing /absolute/local/path with an absolute path on your computer and notebook-path with the name that you want to give it inside the container, for example:

docker run --rm -p 8888:8888 -v "$(pwd)/work:/notebook/work" -v "/home/user/projects/bblocks:/notebook/bblocks" dockerogc/jupyter-notebooks

Locally

To run the Notebook locally, you first need to create a Python virtual environment and install the required packages:

1. Run python -m venv venv. This will create a venv subdirectory with a virtual environment.
2. Execute . venv/bin/activate to activate the virtual environment.
3. Install Jupyter Lab by running python -m pip install jupyterlab
4. Install the required dependencies with python -m pip install -r requirements.txt.
5. Start the Notebook: jupyter-lab

The Notebook will be available on port 8888 (the last of the commands above should automatically open a browser window). You can then open the Notebook file you want to use.