September 3, 2015

Charting in Angular.js

It is a common scenario in our projects to display tons of data in a nice, elegant and informative way. The tool to do so exists since many decades: charts. In This blog post I give an overview of angular.js directives that can easily be used to create all sorts of charts.

There is a variety of JavaScript charting libraries out there and it feels like a new one pops up every month. At w11k we use some of these directives for over 2 years – of course also in production. I selected 4 directives that worked best for me in the past.

  • angularjs-nvd3-directives (NVD3)
  • angular-google-chart (Google Charts)
  • w11k-highcarts (Highcharts)
  • Angular-charts.js (Charts.js)

Overview

To give you a short overview of what the features (e.g. chart types) of the libraries are you can take a look at the following table.

Feature w11k-highchars angular-google-chart angular-nvd3-directives angular-charts.js
Free
Based on Highcharts Google Charts NVD3 Charts.js
Line Chart
Pie Chart
Area Chart
Bar Chart
Column Chart
Scatter Chart
Radar Chart
Combined Chart Types
Additional Chart Types Gauge, Doughnut Maps Bullet Doughnut

 

In-Depth View

w11k-highcharts

w11k-highcharts is a directive we wrote at w11k about two years ago. We recently release it as OpenSource on Github (https://github.com/w11k/w11k-highcharts).

The directive wraps Highcharts, a library which translates a JSON configuration object into a nice looking SVG. The top features besides the huge amount of chart types are that it supports animations, built-in drilldown, customizable all the way – from colors, to tooltips, labels, sizing and all the beauty you can imagine.

highcharts

With all the nice configuration options comes a obvious downer, you have to learn the Highcharts API, which is well documented, readable and filled with JSFiddle examples. The project is highly active and you can mostly find the solution to your problem on Stackoverflow, where some of the developers are active.

 

angular-google-chart

angular-google-chart wraps Google Chart and also accepts a configuration object, which gets rendered into a SVG. In my opinion angular-google-chart’s API isn’t as nice as for example the Highcharts API but can easily be used to display simple charts. Unlike the Google Charts JS Library the angular.js wrapper only supports 6 Chart Types including Area, Column, Pie, Line Charts. Nonetheless it is great if you want to dive right into creating simple Charts.

nvd3

 

angular-nvd3-directives

This module wraps around the most versatile visualization-library out there. Data-Driven-Documents (short: D3) is not a charting library but can pretty much display anything you want it to. With all that freedom comes a ton of complexity and a “not so easy to use” API.

Therefore some folks created NVD3 on top of D3 whose intention was to build re-usable chart components with all the power D3 has, but with far less complexity.

google-charts

The angular-nvd3-directives module wraps NVD3 and consists of multiple directives, one for each chart type. In difference to the previously described modules where you have one generic chart directive that is customizable by a configuration object, you use attributes to set some configuration options of the angular-nvd3-directives. The API Documentation talks a lot about the attributes that can be controlled and is structured the way I would want it to be.

 

angular-charts.js

angular-charts.js is pretty basic, looks awesome and is fairly easy to use. I personally like the look and feel of the charts. Like angular-nvd3-directives it requires you to use attribute rather than a huge configuration object in order to display a chart.

It is a wrapper around Charts.js that only supports 6 chart types and makes a good job in hiding complexity from the developer.

angular-charts

If you want to create a chart with basically no effort and no high requirements, like drilldown or instructiveness I recommend you use this library.

 

Summary and Recommendation

After we took a brief look at the different charting libraries it’s time to give you a little guideline on which library is best for you.

If you don’t want to spend any money on licenses and you have limited time – which seems to be the scenario in many cases – I would go with angular-charts.js. It has a small overhead of only a few Kilobytes and works pretty much out of the box and looks awesome in the default configuration. Another neat detail is, that it uses the HTML5 Canvas element if the Browser supports it – obviously fallbacks are provided for IE7/8.

But I found that in many other situations (mainly client projects) you need a more powerful charting library. After playing a bit with several D3-Wrappers we figured it wasn’t exactly what we imagined, cause we couldn’t create interactive charts without having to re-invent the wheel.

That was when we started to use Highcharts a bit more extensively. At this point I should tell you that Highcharts is free of charge if you use it in a non-commercial website, but if you use it for client projects you probably want to purchase a license. We wrote our own wrapper, because none existed at the time and the more we got used to Highcharts the more we loved it.  It allows you to modify pretty much everything, including

  • Adjusting colors to match the Corporate Identity of the client
  • Combining multiple chart types
  • Customizable tooltips
  • Event handlers
  • Drilldown/Rollup support
  • Support down to IE8

If you want to find out how easy it is to integrate w11k-highcharts into your project head over to https://github.com/w11k/w11k-highcharts and take a look at our directive. Furthermore I recommend you take a look at all the possibilities Highcharts offers at their demo page https://www.highcharts.com/demo

 

Further Reading

 

 

Related Posts

theCodeCampus Autor Kai Henzler

Kai Henzler
Developer at thecodecampus </>

Kai belongs to the generation, which grew up with the internet. His motto is „keep it simple“, which he applies both in web development with AngularJS and as a trainer.


Leave a Reply

Add code to your comment in Markdown syntax.
Like this:
`inline example`

```
code block
example
```

Your email address will not be published.