A Python script to generate offline documentation for vega-lite, an expressive language for building data interactive visualizations.
- In-page table of contents for all pages, including images and examples for each Vega-Lite spec property.
- Mobile friendly layout (sidebars have been removed)
- Quick search of all pages on the public docs page, including examples. (Note examples that load remote data require an internet connection to load normally.)
This step is only necessary if you intend to modify the published docset locally. Otherwise, it can be installed from the Dash app or the Dash contributions repository.
- Clone
vega-lite
as a sibling directory of this repository using changes from this PR. - In that directory, run
yarn install && yarn docset
to generate the site
- You may need to install
rbenv
,Jekyll
, and a recent version of Ruby to build the documentation site.
- In your python virtual environment, add python dependencies:
pip install -r requirements.txt
- Using Jupyter is optional, but recommended for ease of debugging notebook output.
- Run
jupyter notebook
, and double-clickgenerate-vega-lite-docset.ipynb
. Run all cells. - Your generated docset will be this directory
vega-lite.docset
. Use "File > Open Local Docset" to test the file before submitting it.
See notebook for detailed instructions. The DEBUG
flag can be toggled to adjust the amount of logging.
- See vega-docset-generator for documentation about Vega, the language that vega-lite compiles to.