A framework for creating parent-child relationships with D3.js.
View a working example here.
If you have used d3-relationshipgraph, feel free to edit this readme and put a link and image for your example.
You can install d3-relationshipgraph via bower
$ bower install d3-relationshipgraph
You can also use npm
$ npm install d3-relationshipgraph
Or by downloading the repository and running
$ npm install
in the directory.
Since d3.relationshipgraph extends the D3.js framework, it can be easily added to an existing project by adding the following
<script src="https://d3js.org/d3.v3.min.js"></script>
<script src="path/to/scripts/d3.relationshipgraph.min.js"></script>
<link type="text/css" rel="stylesheet" href="path/to/css/d3.relationshipgraph.min.css">Once the framework is added to the HTML file, graphs can be created using familiar D3 selections
var json = [
{
movietitle: 'Avatar',
parent: '20th Century Fox',
value: '$2,787,965,087',
year: '2009'
},
{
movietitle: 'Titanic',
parent: '20th Century Fox',
value: '$2,186,772,302',
year: '1997'
}
];
var graph = d3.select('#graph').relationshipGraph({
showTooltips: true,
maxChildCount: 10,
showKeys: false,
thresholds: [1000000000, 2000000000, 3000000000]
}).data(json);This simple code will produce the example at the beginning of the readme.
Thresholds can be strings or numbers. If you use a string, only values that match exactly will be in that threshold. If you use a number, a numeric value will be in the smallest threshold that is greater than the value.
If the values are strings (such as in the example above), the number is extracted from the string and used. This allows you to use string values such as:
var json = [
{parent: 'a', value: '$100'},
{parent: 'b', value: '$100.15'},
{parent: 'c', value: '100%'},
{parent: 'd', value: '100.15%'}
];and thresholds such as:
var thresholds = [25, 50, 75, 100];Private data can be added to the JSON data by using the _private_ key. This allows you to pass private data into the onClick function that isn't shown in the tooltip.
To use private data, structure your JSON data so that it looks similar to
var myData = {
parent: 'parentA',
name: 'child1',
_private_: {
private1: 'Hidden from the tooltip.',
private2: 'Also hidden from the tooltip.'
}
}d3.relationshipgraph is configured by passing in a JavaScript object into the constructor. The object can have the following properties
config = {
showTooltips: true, // Whether or not to show tooltips when the child block is moused over.
maxChildCount: 10, // The maximum amount of children to show per row before wrapping.
onClick: function(obj) {}, // The callback function to call when a child block is clicked on. This gets passed the JSON for the object.
showKeys: true, // Whether or not to show the JSON keys in the tooltip
thresholds: [100, 200, 300], // The thresholds for the color changes. If the values are strings, the colors are determined by the value of the child being equal to the threshold. If the thresholds are numbers, the color is determined by the value being less than the threshold.
colors: ['red', 'green', 'blue'], // The custom color set to use for the child blocks. These can be color names, HEX values, or RGBA values.
transitionTime: 1000, // The time in milliseconds for the transitions. Set to 0 to disable.
truncate: 25, // The maximum length for the parent labels before they get truncated. Set to 0 to disable.
sortFunction: sortJson, // A custom sort function. The parent value must be sorted first.
valueKeyName: 'Worldwide Gross' // Set a custom key value in the tooltip instead of showing 'value'.
}None of the configurations are required and they all have default values
config = {
showTooltips: true,
maxChildCount: 0, // When the value is 0, the max count is determined by the width of the parent element.
onClick: function () { }, // no-op
showKeys: true,
thresholds: [], // All chiild blocks will be the same color.
transitionTime: 1500,
truncate: 0,
sortFunction: sortJson,
valueKeyName: 'value'
}If a custom sorting is used, the parent value MUST be sorted first.
To update the relationship graph with new data, store the RelationshipGraph object and call the data function with the updated JSON
var json = [
{
movietitle: 'Avatar',
parent: '20th Century Fox',
value: '$2,787,965,087',
year: '2009'
},
{
movietitle: 'Titanic',
parent: '20th Century Fox',
value: '$2,186,772,302',
year: '1997'
}
];
var graph = d3.select('#graph').relationshipGraph({
showTooltips: true,
maxChildCount: 10,
showKeys: false,
thresholds: [1000000000, 2000000000, 3000000000]
});
graph.data(json); // Add the first set of data.
json = [
{
movietitle: 'Avatar',
parent: '20th Century Fox',
value: '$2,787,965,087',
year: '2009'
},
{
movietitle: 'Titanic',
parent: '20th Century Fox',
value: '$2,186,772,302',
year: '1997'
},
{
movietitle: 'Star Wars: The Force Awakens',
parent: 'Walt Disney Studios',
value: '$2,066,247,462',
year: '2015'
}
];
graph.data(json); // Update the graph with new data.To interact with the child nodes once they are in the graph, you can query for them based on subobjects. Thiswill return the objects that match the query. Once these objects have been returned,
you can change the color of the nodes by using the setNodeColor method on the object, or change the stroke color of the node by using the setNodeStrokeColor method.
An example of a query is if I was lookingfor all nodes in the following example that are from the year 2009
var json = [
{
movietitle: 'Avatar',
parent: '20th Century Fox',
value: '$2,787,965,087',
year: '2009'
},
{
movietitle: 'Titanic',
parent: '20th Century Fox',
value: '$2,186,772,302',
year: '1997'
}
];
var graph = d3.select('#graph').relationshipGraph();
graph.query({year: '2009'});That would return the Javascript object(s) that match the query.
This project is licensed under the MIT license -- see the LICENSE.md file for details.
If you would like to contribute please ensure that the following passes
$ grunt test -v
before putting up a pull request.