The Vega editor is a web application for authoring and testing Vega and Vega-Lite visualizations. It includes a number of example specifications that showcase both the visual encodings and interaction techniques. It is deployed at https://vega.github.io/editor/.

A back-end service at https://vega.now.sh/ has been integrated which lets a user log in through GitHub so that user can access his/her personal gists. To contribute to editor-backend, one can contribute here: https://github.com/vega/editor-backend.

You can reset the Vega Editor by going to https://vega.github.io/editor/#/reset and clicking the reset button. This will reset the saved editor state.

To run the editor locally, you must first install the dependencies and then launch a local web server. We assume you have yarn installed.

1. Install the dependencies:

$ yarn

2. Start the server:

$ yarn start

3. The local web server will be accessible from http://localhost:8080.

If you'd like to use Docker, there's a Docker Compose setup that you can use:

1. Build the docker container:

$ docker-compose build

2. Run the Docker Compose service:

$ docker-compose up

1. The local web server will be accessible from http://localhost:8080. You can run yarn commands with docker-compose run editor CMD.

The editor is useful for testing if you are involved in Vega and Vega-Lite development. To use Vega, Vega-Lite, or Vega Datasets from another directory on your computer, you need to link it. For this, run yarn link in the directory of the library that you want to link. Then, in this directory run yarn link <name of library>, e.g. yarn link vega.

For example, to link Vega, run

cd VEGA_DIR
yarn link

cd VEGA_EDITOR_DIR
yarn link vega

The Vega editor supports React Developer Tools and Redux DevTools.

We welcome contributions and promptly review pull requests. For instructions about how to contribute, please follow the Vega-Lite contributing guidelines.

• Tag a new version with yarn version. Pre 1.x, update the minor version if there is a new feature.
• Push the tag. Travis will automatically deploy the editor.