HMPlot

hmplot/compound

The built-in templates, and the experimental surface to write your own patterns.

import { Lollipop, composition } from "hmplot/compound";
const { ErrorBars, Points, ThemeDefaults, minimum, maximum } = composition;

Exports

Histogram, Violin, BoxPlot, Swarm, Bar, Lollipop, Pie, RuleX, RuleY, Crosshair, Callout, Smooth, Ecdf, DensityContour, DensityBands, ErrorBarscomponents
The built-in templates, the same components as hmplot/charts (each has its reference page), and their props types.
compositionnamespace (experimental)
What a pattern written in Svelte is built from: Transform, DataBoundary, ThemeDefaults, the options and data contexts, the data-bound primitive marks, the transforms the built-in templates wrap, framePointer and planeRule, expression builders and the input helpers. Names and behavior may change in any release.

Usage

The built-in templates, built from transforms and simpler marks, are exported from hmplot/charts (hm.charts in Python). hmplot/compound adds composition, everything a pattern written in Svelte may use: the Transform component that runs a named transform around its children, DataBoundary, ThemeDefaults, Undescribed (a helper part left out of the plot’s description), the options and data contexts, the data-bound primitive marks (Rect, Points, Line, ErrorBars, Polygon, …), the transforms the built-in templates wrap, planeRule and framePointer, and expression and input helpers.

composition is experimental: its names and behavior may change in any release, unlike the rest of the package. Its complete list, with what each part is for, a worked Lollipop and the conventions a pattern written in Svelte follows (forward the remaining props to the mark that carries color and interaction, re-export getLegendSource and getMark, give secondary parts legend={false}), is in docs/USAGE.md, “Writing a pattern”. In Python a template packages a pattern of existing components.

Related examples

Lollipop chart