dsm-72/featherplot-svelte

FeatherPlot

This is a lightweight wrapper around deepscatter created by Benjamin Schmidt (@bmschmidt).

Of note is that there is a companion library in python featherplot. If you look on GitHub, there is an example notebook that demonstrates a pipeline for converting a Pandas DataFrame to the feather format using quadfeather as well as how to add sidecars (extra data).

As for this library, it has three main purposes:

1. assist with debugging the deepscatter library. (The example notebooks even found one with using csv data.).

2. assist with understanding the plot.plotAPI({...args}) which are inspired by vega-lite.

3. provide some utility functionality for interactivity with a deepscatter plot e.g. axes selectors, color selectors, number of point sliders, etc.

Building

To build your library:

npm run package

To create a production version of your showcase app:

npm run build

You can preview the production build with npm run preview.

Publishing

To publish your library to npm:

npm publish

Deploying to GitHub Pages

Add this to package.json:

"scripts": {
  "deploy": "NODE_ENV=production; npm run build && npx gh-pages -d build -t true"
}

then:

$ npm run deploy

then add this .nojekyll file to gh-pages:

{
  "scripts": {
    "dev": "vite dev",
    "build": "vite build",
    "deploy": "touch build/.nojekyll && gh-pages -d build -t true"
  }
}