Skip to content
Student Login

Tiny React & D3 flamegraph tutorial

Last updated: June 2018 👉 livestreamed every last Sunday of the month. Join live or subscribe by email 💌

This is a Livecoding Recap – an almost-weekly post about interesting things discovered while livecoding. Usually shorter than 500 words. Often with pictures. Livecoding happens almost every Sunday at 2pm PDT on multiple channels. You should subscribe to My Youtube channel(https://www.youtube.com/user/theswizec) to catch me live.

Flamegraphs are great. They show you when you've been bad and made the CPU cry.

Here is a flamegraph built with React & D3. It shows some stack data I stole from the internet:

You can try it out here, and see the code on GitHub. Consider it a work in progress. Releasing as open source once the animation works smoothly :)

Here's how it works 👇

You have two components:

  1. <Flamegraph>, which recursively renders the data tree
  2. <FlameRect>, which renders a particular rectangle and its label

Flamegraph

The <Flamegraph> component without animation takes your data, loops it through the array of entries at this level, and calls itself on data.children.

import { scaleLinear, schemeOrRd, color as d3color } from "d3";
class Flamegraph extends React.Component {
render() {
const { x = 0, y = 0, width, level = 0, data } = this.props;
const xScale = scaleLinear()
.domain([0, data.reduce((sum, d) => sum + d.value, 0)])
.range([0, width]);
return (
<g transform="{`translate(${x}," ${y})`}="">
{data.map((d, i) => {
const start = data
.slice(0, i)
.reduce((sum, d) => sum + d.value, 0);
return (
<react class="fragment" key={`${level}-${d.name}`}>
<flamerect x={xScale(start)} y={0} width={xScale(d.value)} height={RowHeight} name={d.name}>
{d.children && (
<flamegraph data={d.children} x={xScale(start)} y={RowHeight} width={xScale(d.value)} level="{level" +="" 1}="">
)}
</flamegraph></flamerect></react>
);
})}
</g>
);
}
}

Our render method takes a bunch of params out of props, creates a linear D3 scale to make calculations easier, then renders an SVG grouping element.

Inside that element, we loop through the data, and for each entry, we create a new <React.Fragment>. The fragment contains <FlameRect> which represents the current datapoint, and a <Flamegraph> which renders all the child nodes.

We decide each element's x position based on the sum of all node values up to the current one. And we make sure the child <Flamegraph> uses the same width as the current node. This creates the neat stacking effect.

The <FlameRect> component takes care of choosing a random color on initial render, highlighting when clicked, and displaying a label if there's enough space.

It looks like this:

import { scaleLinear, schemeOrRd, color as d3color } from "d3";
class FlameRect extends React.Component {
state = {
hideLabel: false,
selected: false,
color: schemeOrRd[9][Math.floor(Math.random() * 9)]
};
labelRefCallback = element => {
if (
element &&
element.getBoundingClientRect().width > this.props.width
) {
this.setState({
hideLabel: true
});
}
};
onClick = () => this.setState({ selected: !this.state.selected });
render() {
const { x, y, width, height, name } = this.props,
{ hideLabel, selected } = this.state;
let { color } = this.state;
if (selected) {
color = d3color(color).brighter();
}
return (
<g transform="{`translate(${x}," ${y})`}="" style={{ cursor: "pointer" }} onclick={this.onClick}>
<rect x={0} y={0} width={width} height={height} style={{ stroke: "white", fill: color }}>
{!hideLabel && (
<text x={5} y={13} style={{ fontsize: "12px" }} ref={this.labelRefCallback}>
{name}
</text>
)}
</rect></g>
);
}
}

We render a grouping element that contains a <rect> and a <text>. The rect gets sizing and color information, and the label gets a ref callback and some text.

We use the ref callback to dynamically detect the size of our label and hide it if necessary. That happens in labelRefCallback.

onClick, we flip the selected state.

And well, that's it. You can think of this component as a basic toggle component. Uses all the same logic, just renders as a colorful rectangle instead of a button.

Fin

That's the basic <Flamegraph> component. Uses recursion to render a tree data structure and lets you highlight individual elements.

Where it gets tricky is adding animation that lets users explore their data. Turns out deeply nested React elements are hard to animate smoothly.

The worst part is how long it takes before React even propagates prop updates through the tree before the animation even starts. Once the animation is running, it's smooth as silk.

But getting it started, ho boy.

This warrants further research. I'll be back :)

PS: version 0.1.0 is in fact on npm if you want to play.

About the Author

Hi, I’m Swizec Teller. I help coders become software engineers.

Story time 👇

React+D3 started as a bet in April 2015. A friend wanted to learn React and challenged me to publish a book. A month later React+D3 launched with 79 pages of hard earned knowledge.

In April 2016 it became React+D3 ES6. 117 pages and growing beyond a single big project it was a huge success. I kept going, started live streaming, and publishing videos on YouTube.

In 2017, after 10 months of work, React + D3v4 became the best book I'd ever written. At 249 pages, many examples, and code to play with it was designed like a step-by-step course. But I felt something was missing.

So in late 2018 I rebuilt the entire thing as React for Data Visualization — a proper video course. Designed for busy people with real lives like you. Over 8 hours of video material, split into chunks no longer than 5 minutes, a bunch of new chapters, and techniques I discovered along the way.

React for Data Visualization is the best way to learn how to build scalable dataviz components your whole team can understand.

Some of my work has been featured in 👇

Created bySwizecwith ❤️