Treemaps are a splendid way to represent data that has a nested aspect to it. They allow for the easy display of complicated relative information, such as nested-part-to-whole relationships in a easy to grok fashion. Checkout the wikipedia page or Ben Shneiderman's excellent History of treemaps for more information.
The treemap
in react-vis builds a series of nested divs (allow for easy and highly restyleable trees). We offer ten different layout
strategies, enabling the construction of standard treemaps, circle packed treemaps, and partition trees (also called icicle diagrams).
Import the treemap
component:
import {Treemap} from 'react-vis';
Add the following code to your render function:
<Treemap
title={'My New Treemap'}
width={300}
height={300}
data={myData}
/>
Like other systems that make use of d3's hierarchy layout system we ask that our data be presented to us in a tree like structure. Here's a slice of the famous d3-flare dataset:
const myData = {
"title": "analytics",
"color": "#12939A",
"children": [
{
"title": "cluster",
"children": [
{"title": "AgglomerativeCluster", "color": "#12939A", "size": 3938},
{"title": "CommunityStructure", "color": "#12939A", "size": 3812},
{"title": "HierarchicalCluster", "color": "#12939A", "size": 6714},
{"title": "MergeEdge", "color": "#12939A", "size": 743}
]
},
{
"title": "graph",
"children": [
{"title": "BetweennessCentrality", "color": "#12939A", "size": 3534},
{"title": "LinkDistance", "color": "#12939A", "size": 5731},
{"title": "MaxFlowMinCut", "color": "#12939A", "size": 7840},
{"title": "ShortestPaths", "color": "#12939A", "size": 5914},
{"title": "SpanningTree", "color": "#12939A", "size": 3416}
]
},
{
"title": "optimization",
"children": [
{"title": "AspectRatioBanker", "color": "#12939A", "size": 7074}
]
}
]
}
First, note the recursive tree relationship: each node has a title, and an array of children. This pattern continues until we reach the leaves, where we declare the size of the leaves. This value is rolled up, so that the "cluster" node has 3938 + 3812 + 6714 + 743 = 15207 size units.
-
It is often quite effective to use the literal scale type for Treemap's color attribute, as this allows highly granular control over all of the nodes.
-
It can useful to encode opacity to indicate tree depth, however because each tree leaf is a nested div this gets a little tricky. One technique is to compute the effective RGBA to hex value, check out this link for more details.
-
If your not sure when to use a treemap, remember they provide an easy drop in relationship for pie charts.
Type: number
Width of the component.
Type: number
Height of the component.
Type: number
The padding between cells the cells of the heatmap in pixels.
Type: Object
The data for the component. The data
property is a tree-like structure.
Each point consists of following properties:
title
Type:string
The title to show inside the cell. Might be a string or a React component.size
Type:number
The relative size of the cell.opacity
(optional)
Type:number
The value to visualize the opacity with.color
(optional)
Type:number
orstring
The value to visualize the color with.style
(optional)
Type:object
style object to be added to the inline styles of the arraychildren
(optional)
Type:Array
The children for the leaf.
Type: boolean|Object
Please refer to Animation doc for more information.
Type: boolean
Simple boolean on whether or not to show the root node of the tree.
Type: function
- Should accept arguments (leafNode, domEvent)
Pass in a function that will be called on click on a given leaf.
Type: function
- Should accept arguments (leafNode, domEvent)
Pass in a function that will be called on mouseEnter on a given leaf.
Type: function
- Should accept arguments (leafNode, domEvent)
Pass in a function that will be called on mouseOut on a given leaf.
Type: string
- One of squarify, resquarify, slice, dice, slicedice, binary, circlePack, partition, partition-pivot
This modifies the tiling strategy for the treemap, for more information see the d3 hierarchy docs.
Type: function
- Should accept arguments (a, b)
Pass in a function that will be used to sort the nodes, for more information see the d3 hierarchy docs on sorting.
Scale properties for the color
scale. If color
property is not passed in the data object, each new section of the chart gets the next color (e. g. the 'category'
scale is applied).
Please refer to Scales and Data for more information about scales.