chart = StackedBarChart(politifact, {
x: d => d.proportion,
y: d => d.speaker,
z: d => d.ruling,
xFormat: "+%",
xLabel: "← more lies · Truthiness · more truths →",
yDomain: d3.groupSort(politifact, D => d3.sum(D, d => -Math.min(0, d.proportion)), d => d.speaker),
zDomain: politifact.rulings,
colors: d3.schemeSpectral[politifact.rulings.length],
width,
marginLeft: 70
})
politifact = {
// A map to convert PoltiFact codes into readable names, and whether
// the codes represent lies (negative) or truths (positive).
const rulings = {
"pants-fire": {name: "Pants on fire!", sign: -1},
"false": {name: "False", sign: -1},
"mostly-false": {name: "Mostly false", sign: -1},
"barely-true": {name: "Mostly false", sign: -1}, // pessimistic
"half-true": {name: "Half true", sign: 1},
"mostly-true": {name: "Mostly true", sign: 1},
"true": {name: "True", sign: 1}
};
// The PoltiFact data includes categories we don’t want to consider (namely
// full-flop, which isn’t really true or false), so filter.
const politifact = (await FileAttachment("politifact.csv").csv())
.filter(d => d.ruling in rulings);
// Compute the total number of rulings for each speaker.
const total = d3.rollup(politifact, D => d3.sum(D, d => d.count), d => d.speaker);
// Lastly, convert the counts to signed counts (negative for lies, positive for
// truths), and compute the normalized counts (bias; -1 for all lies, +1 for all
// truths). The returned array has an extra “rulings” property which we use to
// see the z-domain of the chart for stable ordering and color.
return Object.assign(politifact.map(d => ({
speaker: d.speaker,
ruling: rulings[d.ruling].name,
count: d.count * rulings[d.ruling].sign,
proportion: d.count / total.get(d.speaker) * rulings[d.ruling].sign
})), {
rulings: [...d3.union(Object.values(rulings).map(d => d.name))]
});
}
// Copyright 2021 Observable, Inc.
// Released under the ISC license.
// https://observablehq.com/@d3/diverging-stacked-bar-chart
function StackedBarChart(data, {
x = d => d, // given d in data, returns the (quantitative) x-value
y = (d, i) => i, // given d in data, returns the (ordinal) y-value
z = () => 1, // given d in data, returns the (categorical) z-value
title, // given d in data, returns the title text
marginTop = 30, // top margin, in pixels
marginRight = 0, // right margin, in pixels
marginBottom = 0, // bottom margin, in pixels
marginLeft = 40, // left margin, in pixels
width = 640, // outer width, in pixels
height, // outer height, in pixels
xType = d3.scaleLinear, // type of x-scale
xDomain, // [xmin, xmax]
xRange = [marginLeft, width - marginRight], // [left, right]
yDomain, // array of y-values
yRange, // [bottom, top]
yPadding = 0.1, // amount of y-range to reserve to separate bars
zDomain, // array of z-values
offset = d3.stackOffsetDiverging, // stack offset method
order = (series) => { // stack order method; try also d3.stackOffsetNone
return [ // by default, stack negative series in reverse order
...series.map((S, i) => S.some(([, y]) => y < 0) ? i : null).reverse(),
...series.map((S, i) => S.some(([, y]) => y < 0) ? null : i)
].filter(i => i !== null);
},
xFormat, // a format specifier string for the x-axis
xLabel, // a label for the x-axis
colors = d3.schemeTableau10, // array of colors
} = {}) {
// Compute values.
const X = d3.map(data, x);
const Y = d3.map(data, y);
const Z = d3.map(data, z);
// Compute default y- and z-domains, and unique them.
if (yDomain === undefined) yDomain = Y;
if (zDomain === undefined) zDomain = Z;
yDomain = new d3.InternSet(yDomain);
zDomain = new d3.InternSet(zDomain);
// Omit any data not present in the y- and z-domains.
const I = d3.range(X.length).filter(i => yDomain.has(Y[i]) && zDomain.has(Z[i]));
// If the height is not specified, derive it from the y-domain.
if (height === undefined) height = yDomain.size * 25 + marginTop + marginBottom;
if (yRange === undefined) yRange = [height - marginBottom, marginTop];
// Compute a nested array of series where each series is [[x1, x2], [x1, x2],
// [x1, x2], …] representing the x-extent of each stacked rect. In addition,
// each tuple has an i (index) property so that we can refer back to the
// original data point (data[i]). This code assumes that there is only one
// data point for a given unique y- and z-value.
const series = d3.stack()
.keys(zDomain)
.value(([, I], z) => X[I.get(z)])
.order(order)
.offset(offset)
(d3.rollup(I, ([i]) => i, i => Y[i], i => Z[i]))
.map(s => s.map(d => Object.assign(d, {i: d.data[1].get(s.key)})));
// Compute the default y-domain. Note: diverging stacks can be negative.
if (xDomain === undefined) xDomain = d3.extent(series.flat(2));
// Construct scales, axes, and formats.
const xScale = xType(xDomain, xRange);
const yScale = d3.scaleBand(yDomain, yRange).paddingInner(yPadding);
const color = d3.scaleOrdinal(zDomain, colors);
const xAxis = d3.axisTop(xScale).ticks(width / 80, xFormat);
const yAxis = d3.axisLeft(yScale).tickSize(0);
// Compute titles.
if (title === undefined) {
const formatValue = xScale.tickFormat(100, xFormat);
title = i => `${Y[i]}\n${Z[i]}\n${formatValue(X[i])}`;
} else {
const O = d3.map(data, d => d);
const T = title;
title = i => T(O[i], i, data);
}
const svg = d3.create("svg")
.attr("width", width)
.attr("height", height)
.attr("viewBox", [0, 0, width, height])
.attr("style", "max-width: 100%; height: auto; height: intrinsic;");
svg.append("g")
.attr("transform", `translate(0,${marginTop})`)
.call(xAxis)
.call(g => g.select(".domain").remove())
.call(g => g.selectAll(".tick line").clone()
.attr("y2", height - marginTop - marginBottom)
.attr("stroke-opacity", 0.1))
.call(g => g.append("text")
.attr("x", xScale(0))
.attr("y", -22)
.attr("fill", "currentColor")
.attr("text-anchor", "middle")
.text(xLabel));
const bar = svg.append("g")
.selectAll("g")
.data(series)
.join("g")
.attr("fill", ([{i}]) => color(Z[i]))
.selectAll("rect")
.data(d => d)
.join("rect")
.attr("x", ([x1, x2]) => Math.min(xScale(x1), xScale(x2)))
.attr("y", ({i}) => yScale(Y[i]))
.attr("width", ([x1, x2]) => Math.abs(xScale(x1) - xScale(x2)))
.attr("height", yScale.bandwidth());
if (title) bar.append("title")
.text(({i}) => title(i));
svg.append("g")
.attr("transform", `translate(${xScale(0)},0)`)
.call(yAxis)
.call(g => g.selectAll(".tick text")
.attr("dx", -3)
.attr("x", y => { // Find the minimum x-value for the corresponding y-value.
const x = d3.min(series, S => S.find(d => Y[d.i] === y)?.[0]);
return xScale(x) - xScale(0);
}));
return Object.assign(svg.node(), {scales: {color}});
}
import {Swatches} from "@d3/color-legend"
import {howto, altplot} from "@d3/example-components"
One platform to build and deploy the best data apps
Experiment and prototype by building visualizations in live JavaScript notebooks. Collaborate with your team and decide which concepts to build out.
Use Observable Framework to build data apps locally. Use data loaders to build in any language or library, including Python, SQL, and R.
Seamlessly deploy to Observable. Test before you ship, use automatic deploy-on-commit, and ensure your projects are always up-to-date.