Delaunay points are a list of NaN

[Chad Hartman]

Greetings; I'm trying to learn about this library and am seeing strange behavior. Regardless of the ranges and number of values passed, the resulting Delaunay object's points are NaN. Steps to reproduce:

$ node -v
v14.4.0

$ npm -version
6.14.5

./package.json:

{
  "type": "module",
  "dependencies": {
    "d3-delaunay": "^5.3.0"
  }
}

./index.js:

import d3 from "./node_modules/d3-delaunay/dist/d3-delaunay.min.js";

let coords = Float64Array.of(69, 82, 203, 18, 320, 116, 359, 88, 302, 37, 16, 108, 255, 207, 199, 76, 340, 142, 311, 139, 381, 37, 340, 82);

console.log(d3.Delaunay.from(coords).points);

$ node index.js outputs:

Float64Array(48) [
  NaN, NaN, NaN, NaN, NaN, NaN, NaN, NaN,
  NaN, NaN, NaN, NaN, NaN, NaN, NaN, NaN,
  NaN, NaN, NaN, NaN, NaN, NaN, NaN, NaN,
  NaN, NaN, NaN, NaN, NaN, NaN, NaN, NaN,
  NaN, NaN, NaN, NaN, NaN, NaN, NaN, NaN,
  NaN, NaN, NaN, NaN, NaN, NaN, NaN, NaN
]

[Mike Bostock]

Delaunay.from takes an array [[x1, y1], [x2, y2], …].

Use new Delaunay if you want to pass in an array [x1, y1, x2, y2, …].