Creating a Dataset in HDX

In order to create a dataset on HDX programmatically, run the functions CreateDataset() and CreateResource() available in create/create.py script. That function takes two parameters as input:

• apikey: (String) Your HDX API key.
• path: (Dictionary) An object with all the fields necessary to create a dataset and add resources in HDX.

HDX has created a series of custom validators. Please refer to the file dataset.json for an example that includes all the required fields for a dataset and the resource.json file for the required fields for a resource.

You will notice that, in HDX, there are a number of fields that are required. For instance, you will need to add a group field which refers to a country in HDX by adding the country's ISO 3-letter-code. For more information about the HDX metadata, please refer to HDX's dataset metadata guide.

Usage

To experiment with the script, do the following:

• Add your HDX API Key to the create/create.py script.
• Edit the dataset.json file.

Setup

Create a virtual environment using virtualenv and install the dependencies from requirements.txt. The Makefile setup instruction contains those commands and can be run with the shortcut (run it inside the repository folder):

$ make setup

##TO-DO

• Add script to create GeoJSON resource for all projects on GeoTagX
• Refactor and add separate configuration file for params like author, email, etc