This tutorial requires you to have read Feeding JSON tree structures to the JIT and on controllers first.
Hi, this is going to be a quick tutorial on how to set the treemap up and running.
Note: You could change all calls to TM.Squarified to TM.SliceAndDice or TM.Strip and it should also work.
We are going to work with this tree JSON structure:
var json = {"id":"node02",
"name":"0.2",
"data":[
{"key":"key1",
"value":195},
{"key":"key2",
"value":5}],
"children":[
{"id":"node13",
"name":"1.3",
"data":[
{"key":"key1",
"value":23},
{"key":"key2",
"value":8}],
"children":[
{"id":"node24",
"name":"2.4",
"data":[
{"key":"key1",
"value":6},
{"key":"key2",
"value":-75}],
"children":[]},
{"id":"node25",
"name":"2.5",
"data":[
{"key":"key1",
"value":9},
{"key":"key2",
"value":-48}],
"children":[]},
{"id":"node26",
"name":"2.6",
"data":[
{"key":"key1",
"value":1},
{"key":"key2",
"value":-1}],
"children":[]},
{"id":"node27",
"name":"2.7",
"data":[
{"key":"key1",
"value":7},
{"key":"key2",
"value":25}],
"children":[]}]},
{"id":"node18",
"name":"1.8",
"data":[
{"key":"key1",
"value":17},
{"key":"key2",
"value":28}],
"children":[
{"id":"node29",
"name":"2.9",
"data":[
{"key":"key1",
"value":8},
{"key":"key2",
"value":-28}],
"children":[]},
{"id":"node210",
"name":"2.10",
"data":[
{"key":"key1",
"value":9},
{"key":"key2",
"value":-83}],
"children":[]}]},
{"id":"node111",
"name":"1.11",
"data":[
{"key":"key1",
"value":25},
{"key":"key2",
"value":-82}],
"children":[
{"id":"node212",
"name":"2.12",
"data":[
{"key":"key1",
"value":8},
{"key":"key2",
"value":-27}],
"children":[]},
{"id":"node213",
"name":"2.13",
"data":[
{"key":"key1",
"value":3},
{"key":"key2",
"value":-80}],
"children":[]},
{"id":"node214",
"name":"2.14",
"data":[
{"key":"key1",
"value":7},
{"key":"key2",
"value":-73}],
"children":[]},
{"id":"node215",
"name":"2.15",
"data":[
{"key":"key1",
"value":7},
{"key":"key2",
"value":26}],
"children":[]}]},
{"id":"node116",
"name":"1.16",
"data":[
{"key":"key1",
"value":17},
{"key":"key2",
"value":91}],
"children":[
{"id":"node217",
"name":"2.17",
"data":[
{"key":"key1",
"value":7},
{"key":"key2",
"value":48}],
"children":[]},
{"id":"node218",
"name":"2.18",
"data":[
{"key":"key1",
"value":10},
{"key":"key2",
"value":-86}],
"children":[]}]},
{"id":"node119",
"name":"1.19",
"data":[
{"key":"key1",
"value":52},
{"key":"key2",
"value":-77}],
"children":[
{"id":"node220",
"name":"2.20",
"data":[
{"key":"key1",
"value":8},
{"key":"key2",
"value":64}],
"children":[]},
{"id":"node221",
"name":"2.21",
"data":[
{"key":"key1",
"value":5},
{"key":"key2",
"value":84}],
"children":[]},
{"id":"node222",
"name":"2.22",
"data":[
{"key":"key1",
"value":6},
{"key":"key2",
"value":81}],
"children":[]},
{"id":"node223",
"name":"2.23",
"data":[
{"key":"key1",
"value":1},
{"key":"key2",
"value":25}],
"children":[]},
{"id":"node224",
"name":"2.24",
"data":[
{"key":"key1",
"value":4},
{"key":"key2",
"value":18}],
"children":[]},
{"id":"node225",
"name":"2.25",
"data":[
{"key":"key1",
"value":10},
{"key":"key2",
"value":37}],
"children":[]},
{"id":"node226",
"name":"2.26",
"data":[
{"key":"key1",
"value":8},
{"key":"key2",
"value":83}],
"children":[]},
{"id":"node227",
"name":"2.27",
"data":[
{"key":"key1",
"value":10},
{"key":"key2",
"value":-62}],
"children":[]}]},
{"id":"node128",
"name":"1.28",
"data":[
{"key":"key1",
"value":37},
{"key":"key2",
"value":-40}],
"children":[
{"id":"node229",
"name":"2.29",
"data":[
{"key":"key1",
"value":8},
{"key":"key2",
"value":-67}],
"children":[]},
{"id":"node230",
"name":"2.30",
"data":[
{"key":"key1",
"value":8},
{"key":"key2",
"value":46}],
"children":[]},
{"id":"node231",
"name":"2.31",
"data":[
{"key":"key1",
"value":4},
{"key":"key2",
"value":-99}],
"children":[]},
{"id":"node232",
"name":"2.32",
"data":[
{"key":"key1",
"value":8},
{"key":"key2",
"value":-38}],
"children":[]},
{"id":"node233",
"name":"2.33",
"data":[
{"key":"key1",
"value":1},
{"key":"key2",
"value":-3}],
"children":[]},
{"id":"node234",
"name":"2.34",
"data":[
{"key":"key1",
"value":8},
{"key":"key2",
"value":82}],
"children":[]}]},
{"id":"node135",
"name":"1.35",
"data":[
{"key":"key1",
"value":24},
{"key":"key2",
"value":63}],
"children":[
{"id":"node236",
"name":"2.36",
"data":[
{"key":"key1",
"value":10},
{"key":"key2",
"value":8}],
"children":[]},
{"id":"node237",
"name":"2.37",
"data":[
{"key":"key1",
"value":8},
{"key":"key2",
"value":63}],
"children":[]},
{"id":"node238",
"name":"2.38",
"data":[
{"key":"key1",
"value":6},
{"key":"key2",
"value":46}],
"children":[]}]}]};
Note: remember that the Treemap takes the first dataset object value to calculate the rectangles dimensions. That means that the values you setted on the first dataset object for all nodes must be coherent. Coherent means that, for example, if node A has nodes B and C as children, B having 3 as its first dataset object value and C having 2 as its first dataset value, then A must have 5 as its first dataset object value.
Put this HTML in your page:
<html> <head> <link type="text/css" rel="stylesheet" href="/static/css/example-treemap.css" /> <script type="text/javascript" src="/static/js/mootools-1.2.js"></script> <script type="text/javascript" src="/static/js/treemap/Treemap.js" ></script> <script type="text/javascript" src="/static/js/example/example-treemap.js" ></script> </head> <body onload="init();"> <div id="infovis" /> </body> </html>
Note: You’ll probably have to change the paths to the css and javascript files.
Note2: This is the only visualization that requires Mootools 1.2+ to run. I use version 1.2 which comes packaged with the library (in the extras folder). You can also download it from here.
Now, create a treemap-example.css and put this code in it:
html, body {
width:100%;
height:100%;
background-color:#222;
text-align:center;
overflow:hidden;
font-size:10px;
font-family:Verdana, Geneva, Arial, Helvetica, sans-serif;
}
#infovis {
position:relative;
width:900px;
height:500px;
margin:auto;
background-color:#1D1D20;
}
#infovis div {
position:absolute;
overflow:hidden;
}
#infovis .content {
background-color:#333;
border:0px solid #111;
}
#infovis .head {
height:12px;
color:white;
background-color:#444;
}
#infovis .head.in-path {
background-color:#655;
}
#infovis .body {
background-color:black;
}
#infovis .leaf {
color:white;
background-color:#111;
display:table-cell;
vertical-align:middle;
}
#infovis .over-leaf {
border:1px solid #9FD4FF;
}
#infovis .over-content {
background-color: #9FD4FF;
}
#infovis .over-head { /* ...boy i'm funny */
background-color:#A4D9FF;
color:black;
}
/*TOOLTIPS*/
.tool-tip {
color: #fff;
width: 139px;
z-index: 13000;
background-color: black;
}
.tip-title {
font-weight: bold;
font-size: 11px;
margin: 0;
color: #9FD4FF;
padding: 8px 8px 4px;
background-color: black;
}
.tip-text {
font-size: 11px;
padding: 4px 8px 8px;
background-color: black;
}
Note: I won’t explain all CSS classes, they are pretty much self explainable. However, you must put a position:relative; in the infovis container (the visualization container), just as specified on the CSS stylesheet.
Finally, create an example-treemap.js file and put this code in it:
function init() {
var json = //json data specified above...
var tm = new TM.Squarified();
tm.loadFromJSON(json);
}
You should see a treemap up and running.
Some notes:
- It’s mandatory to put the width and height style properties on the main div container, just as specified on the CSS stylesheet.
- You could change the main container id by setting the rootId property to whatever id you like in the controller object passed to the TM constructor.
Customizing the Treemap
Let’s add some colors!
In order to add colors you must know three things:
- Colors are calculated by taking the second dataset object value. You must also know the range of this value. In this example the second value ranges between [-100, 100]. You can change these values by setting the Color.minValue and Color.maxValue parameters for the treemap controller.
- You can also choose the color range for your visualization, this is done by setting the Color.minColorValue and Color.maxColorValue parameters. These properties take an array of 3 integers each describing an RGB value.
- Finally, to enable treemap coloring you must set the property Color.allow to true.
So the init javascript function would now look like:
function init() {
var json = //..json data
var tm = new TM.Squarified({
//main container id.
rootId: 'infovis',
//orientation
orientation: "v",
Color: {
//Allow coloring
allow: true,
//Set min value and max value for
//the second *dataset* object values.
//Default's to -100 and 100.
minValue: -100,
maxValue: 100,
//Set color range. Default's to reddish and
//greenish. It takes an array of three
//integers as R, G and B values.
maxColorValue: [0, 255, 50],
minColorValue: [255, 0, 50]
}
});
tm.loadFromJSON(json);
}
You should see a colored treemap now.
Finally, let’s put nice tooltips! We’ll do this by using the Tip Mootools class.
First, we’ll define an onAfterCompute controller method to query for all nodes that will be displaying information (like .head and .leaf nodes) and assign the information to be displayed. To know more about this please check the docs for the Tips Mootools plug in
I also define some extra methods that convert the dataset to HTML.
I also allow tooltips by setting tips to true.
So the JavaScript file should now look like this:
function init() {
var json = //...same json data as before...
var tm = new TM.Squarified({
//main container id.
rootId: 'infovis',
//orientation
orientation: "v",
//Allow tips for treemap
tips: true,
Color: {
//Allow coloring
allow: true,
//Set min value and max value for
//the second *dataset* object values.
//Default's to -100 and 100.
minValue: -100,
maxValue: 100,
//Set color range. Default's to reddish and
//greenish. It takes an array of three
//integers as R, G and B values.
maxColorValue: [0, 255, 50],
minColorValue: [255, 0, 50]
},
onAfterCompute: function() {
var that = this, parent;
$$('#infovis .leaf', '#infovis .head').each(function(elem, i) {
//get the JSON tree node element having the same id
//as the dom element queried and makeTip.
if(p = elem.getParent()) {
var sTree = TreeUtil.getSubtree(tm.tree, p.id);
if(sTree) that.makeTip(elem, sTree);
}
});
},
//Tooltip content is setted by setting the *title* of the element to be *tooltiped*.
//Read the mootools docs for further understanding.
makeTip: function(elem, json) {
var title = json.name;
var html = this.makeHTMLFromData(json.data);
elem.store('tip:title', title).store('tip:text', html);
},
//Take each dataset object key and value and make an HTML from it.
makeHTMLFromData: function(data) {
var html = '';
for(var i=0; i<data.length; i++) {
html += data[i].key + ': ' + data[i].value + '<br />';
}
return html;
}
});
tm.loadFromJSON(json);
}
You should see some tooltips now!
Remember that you have more controller methods to customize your visualization just as you want! Hope it was helpful.
Hello there, I'm Nicolas Garcia Belmonte, a Computer Science Engineer from the Buenos Aires Institute of Technology, in Argentina. I live in France now.
I was wondering if there is a practical limit to the number of elements in the treemap. I am looking at generating a Treemap with about 5000 nodes. About 70 are top level. The JSON string ends up being huge, about 800k, without any compression. Will the treemap be able to handle this size tree?
Thanks,
Ben
Hi Benjamin,
I think that the top feed for a treemap is 600 nodes (aprox.). However, you can set a controller to your treemap that allows it to bufferize nodes. That means that even when having a larger tree, you can force the TM to display only N levels below the actual root. As you traverse the tree, it will make requests for new nodes to be displayed while deleting unnecesary nodes and thus keeping a smaller number of nodes in memory. If you want to know more you can check the on controllers post and the example source for the Spacetree (packaged in the library). Hope this helps
Thanks for the advice, Nicolas. I’m still having some trouble figuring it out. It doesn’t help that I don’t have much JavaScript experience. I looked at the controller post. In order to ‘bufferize nodes’, should I implement the request function to only call onComplete() when the level is less than N? Should we continue this conversation on the Google Group? Thanks for your help.
Yes we should, I’ll post your question there: http://groups.google.com/group/javascript-information-visualization-toolkit