JavaScript Extensions with a Live Entity-based Chart in Oracle Policy Modelling

JavaScript Extensions with a Live Entity-based Chart in Oracle Policy Modelling

[NB : An updated post has been written in February 2019]

One of the most interesting questions I have been asked during my current assignment in beautiful Madrid has been the following :

“How can I make a Chart”

Well, that’s easy I said : pick your Charting platform (for example, you might choose D3 as the charting library as it is well-known and incredibly powerful. Or alternatively you could manipulate SVG arcs and lines yourself if you want to, to achieve the same thing.

“How can I make a Chart based on Oracle Policy Automation data”

Well, that’s easy I said : you even have an example in the Oracle Policy Automation Example Projects, called the Loan Advisor. You can see a screenshot from that project, right here:

JavaScript Extensions with a Live Entity-based Chart : Intro

Digging deeper into that Project, we discover that the attributes that are used to create the chart are global :

JavaScript Extensions with a Live Entity-based Chart : Intro 2

“How can I make a Chart that uses Entity data”

Well that’s easy I said, you could hook up the D3 library to the entity, and use a similar system to our custom Entity Container example to retrieve the data. After all, the Pie Chart in D3  accepts any array of data organized into labels and values. You may remember in that example, we sought the name of the entity using EntityId and then iterated through the instances in the Entity to display them.

But upon investigation this is not really satisfactory. Adding an instance to the entity container does not refresh the Chart. In fact, you have to navigate forward and then backwards to get the Chart to refresh. That’s because, of course, your new “instance” does not actually exist yet. It does not get added into the entity.instances[i] list until you leave the current screen.

“How Can I make JavaScript Extensions with a Live Entity-based Chart in Oracle Policy Modelling”

You will have guessed by now that my entourage here is pretty demanding! So here goes. Firstly, the scenario. I have created an Entity called the payment. It has two attributes, the identifier, a text attribute, and the payment amount, a currency or number as you wish.

The goal : on ONE Screen, enter data and view a live chart that updates live, without changing Screens, as data is added or removed.

To experiment further with this, I also decided to create another Entity called the shadow payment. This is a copy of the original, created by the simple rule in Word that follows:

JavaScript Extensions with a Live Entity-based Chart Intro 3

The idea was to test if it would work both on inferred and normal entity instances. It seemed to, so you could probably use the code with either. To enable my shadow entities, I set up the relationship for the above Word rules to function properly.

JavaScript Extensions with a Live Entity-based Chart in Oracle Policy Modelling Intro 5

The Code Concept

Now we come to the code itself. This is, as I always state clearly, just code that I have hacked together to see if a concept had the ability to be taken further. It is in no way production or even unit test ready. That’s your problem. But I hope you find it interesting and inspiring.

Another note, I have seen that this “code view” messes with some characters. The OPA Hub Shop has a PDF version which you can use to compare and correct anything that looks wrong, notably the > and < characters don’t show up correctly here.

The code uses D3 for the pie chart, and the explanation follows after the code. Pop D3 in a JS file in the resources folder, as well as the D3 library.

 * Richard Napier The OPA Hub Website April 2018
 * Educational Example of Custom D3 Chart Extension with live refresh on a screen
 * I will remember this is for demonstration and educational purposes only
	customContainer: function (control) {
		if (control.getProperty("name") === "xChart") {
			return {
				mount: function (el) {
				update: function (el) {
					if (document.readyState == "complete") {
						var entity = control._source.screen.serverState.shadowpayment;
						var myFlatList = [];
						var myObject;
						var width = 300, //width
						height = 300, //height
						radius = 150, //radius
						color = d3.scaleOrdinal(d3.schemeCategory10);
						var size = Object.keys(entity).length;
						for (i = 1; i < size+1; i++) {
							myObject = new Object();
							myObject.label =control._source.screen.serverState.shadowpayment["@_@opa-rel-target={payment[the payment" + i + "]}/global[global]"].shadow_payment;
							myObject.value = control._source.screen.serverState.shadowpayment["@_@opa-rel-target={payment[the payment" + i + "]}/global[global]"].shadow_amount;
							//console.log(" Flattened the list - item " + i);
						data = myFlatList;
						var vis =;"svg").remove();
						var vis =
							.attr("width", width)
							.attr("height", height)
							.attr('transform', 'translate(' + (width / 2) +
								',' + (height / 2) + ')'); 
						var arc = d3.arc().outerRadius(radius)
							.innerRadius(0); ;
						var pie = d3.pie()
							.value(function (d) {
								return d.value;
						var arcs = vis.selectAll("g.slice")
							.attr("class", "slice");
						.attr("fill", function (d, i) {
							return color(i);
						.attr("d", arc);
						.attr("transform", function (d) {
							d.innerRadius = 0;
							d.outerRadius = radius;
							return "translate(" + arc.centroid(d) + ")";
						.attr("text-anchor", "middle")
						.text(function (d, i) {
							return data[i].label;

Review of JavaScript Extensions with a Live Entity-based Chart

Line 8 : This is a Custom Container. So make sure that you drop a container into your Screen, and in the container make sure you add your Entity. In my example, I displayed both the original the payment entity, for the user to add or delete records, as well as the shadow payment which I leveraged to draw the pie chart.

Line 19 ; This example uses control._source.screen.serverState.shadowpayment. This JavaScript object contains the instances of your entity, and is updated as new instances are added, before you leave the page. Of course the name is the technical name  of your entity so don’t forget to add that to your Data tab.

Line 27 : Selecting a standard set of colors from the D3 color categories

Line 28: Finding how many objects are inside the control._source.screen.serverState.shadowpayment object. Each instance is a child object.

Lines 32 and 33 : Creating dynamically the identifiers of each row to select the label and the value using the names of the attributes in your entity, and copying them into a flat object with values and labels

Line 37 : Set the flat list as the data source for the Pie Chart

Line 40 : Add the Chart into the Container

Line 41 : Add the pie chart, setting the origin to the center of the Container, and hooking up to the data we created

Line 50 and beyond : Using D3, draw the arcs and fills that make the Pie Chart by going through the total set of data and dividing the pie into the right number of pieces.

This is the result, in full Hollywood glory, of JavaScript Extensions with a Live Entity-based Chart in Oracle Policy Modelling.

The PDF version is on the OPA Hub Shop, just search for the Pie Chart example.

We will of course be revisiting this to investigate making it more robust, but it’s a good starting point. Have fun and see you soon!

Richard Napier

Author: Richard Napier

After 8 years in case management and ERP software roles, Richard Napier joined Siebel Systems in 1999 and took up the role of managing the nascent Siebel University in Southern Europe. He subsequently was Director of Business Development and Education for InFact Group (now part of Business & Decisions) for 8 years. He now runs Intelligent Advisor IT Consulting OÜ. Owner of, he also is Co-Founder of the Siebel Hub.

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Serving Customers Worldwide
Hide picture