This library is intended to replace the old Bookworm-D3 library. It translates a variety of simple plot types from the old library into Vega-Lite specs, and then embeds them on a target div.

It's still in a very early stage of development, but is suitable for building a few visualizations. I also hope to make it embeddable in observable notebooks pretty soon.

It should really be an ES6 module, but for now it's just a single 'index.js' file thaht you can import.

API

Base keys

• plottype: ["linechart", "heatmap", "pointchart"]
• vega: Any other arguments can be passed straight through to the vega spec. This overwrites any default fields. E.g.:

  {"vega": {"title": "My plot title"}}

• aesthetic: ["x", "y", "color"], etc. See plots for supported elements.
  • This is an object map.
  • When a bookworm contains multiple search limits, the aesthetic Search (capitalized) can be used to refer to the different search terms.
  • E.g., "aesthetic": {"x": "date_year", "y": "classification", "color": "Search"}

Supported plots

Vega supports a powerful grammar of graphics, which can be accessed directly through the vega key. The 'plottype'

• heatmap Required aesthetics: 'x' and 'y' (ordinal, numeric, or date); 'fill' (ordinal)
• linechart. 'x', 'y', and 'color' supported.
• barchart. Required aesthetics 'y' and 'x'. I generally make the x axis numeric and the y axis categorical for text plotting.
• pointchart: 'x', 'y', and 'color'. This works better than a line chart with categorial variables. By default, the y axis is sorted by count.
• USchoropleth. A US state choropleth map.
  • Required aesthetics 'state', which gives a database field with the full name of the state. (It would be nice to include postal code, but that's for later).