Doodles for working with Companies House Beneficial Ownership Register Data.
This repo creates a set of linked Docker containers to provide a working environment for exploring company data.
The configuration includes:
- a Jupyter notebook server for writing reproducible analysis scripts
- a mongodb container (handy for UK Companies House beneficial ownership JSON data)
- a PostgreSQL container, eg for Companies House company info
I also tried to add in a third party neo4j container seeded with Panama Papers data, but this caused even more memory issues...:-( See the
I assume you have
I needed to up the spec of my docker-machine to run this:
docker-machine rm default
docker-machine -D create -d virtualbox --virtualbox-disk-size "35000" --virtualbox-memory 2048 default
You should now be able to just run:
docker-compose up -d
and then launch a Jupyter notebook in your browser.
The notebook is connected to a mongo database container. The
Companies House - Significant Control Snapshot - Loader.ipynb notebook shows how to grab a copy of the UK Companies House beneficial ownership register and load it into the mongo db from a notebook.
To shutdown the containers:
and to bring them back up: