datakit-dworld 1.0.0

Features

• Create a project (dataset) on data.world

• Push data files from a local directory to a data.world project

Setup instructions

Assuming you have DataKit installed, run the following to install the datakit-dworld plugin:

$ pip install -e 'git+https://github.com/associatedpress/datakit-dworld.git#egg=datakit-dworld'

Create a configuration file at ~/.datakit/plugins/datakit-dworld/config.json to tell the plugin about your data.world account. It should look like this:

{
  "username": "USERNAME",
  "api_token": "API_TOKEN"
}

Replace USERNAME with the username you will use on data.world to publish datasets.

Log in to data.world as that same user, and go to your Advanced settings page. You’ll see two long tokens under the “API Token” section. Copy the one labeled “Admin”, and paste that into your configuration file in place of API_TOKEN.

How to use

When in a datakit project, you’ll have two new commands:

• datakit dworld create, which creates a new project (dataset) on data.world. This requires a --slug option to help set the project’s URL. For example, if your username is example-human, then running datakit dworld create --slug example-dataset will create an empty dataset at https://data.world/example-human/example-dataset.

• datakit dworld push, which uploads data files (CSV, etc.) from your project to data.world. This defaults to uploading any files in your project’s data/public directory, but you can specify a different directory with --source-dir.

• datakit dworld summary, which updates the dataset’s overall Markdown summary on data.world. This assumes a Markdown file exists at publish/distro_summary.md; if that is not present, it creates one based on the dataset’s existing summary.

Credits

This plugin was created with Cookiecutter and the associatedpress/cookiecutter-datakit-plugin project template (a modified version of the most excellent audreyr/cookiecutter-pypackage).