Le Post Infeeny

Les articles des consultants et experts Infeeny

WinJS Contrib Basics

Now that we set the stage for WinJS and WinJS Contrib, it’s time to dig into some code goodness.
In this episode, we will look at the fundamental of WinJS Contrib.

interactive elements

You have interactive elements in all apps. Weither it is with mouse or touch, it is good practice to provide visual feedback on interactions, otherwise the user feel trust in your application. WinJS Contrib has an abstraction over interactive element. Using it will provide an appropriate cursor with mouse, and a visual feedback with touch. It is as simple as :

WinJSContrib.UI.tap(domElement, callback, options);

the third argument is optional and allow you to specify such thing as « disableAnimation » to avoid the « pressed » effect with touch, or « lock » to add interactivity on elements in a scrollable are (like adding a button inside an item in a listview). The beauty is that « tap » is cross platform and will act as « fastclick » on Android and iOS.

When using it, a css class « tap » will be added on the element, and when element is pressed a « tapped » will be set as long as the element is pressed. You can use this to override the default animation.

In previous post, we explained that WinJS Contrib replaces the page implementation of WinJS. Our page control allow you to declare taps in your markup. You must simply add a « tap » attribute on an element, and as a value, you pass the name of the function on your page, like this :

<div class="button" tap="someFunctionOnMyPage"></div>

When called, your function will be bound to the page, it means that when the function is called, « this » refer to your page.

You could specify arguments to tap by adding a « tap-args » attribute. The value will be passed in the argument to the function on your page. The value could be a simple string, or use the resolution pipeline from WinJS Contrib. We will expand on this pipeline in another post, for now, just note that you can inject argument from a property or a function of your page.

<div class="button" tap="someFunctionOnMyPage" tap-args="ctrl:somePropertyOrFunctionOnMyPage"></div>

Another thing that is really common in all applications is to navigate after clicking a button. You can implement this as easily as tap with a « linkto » attribute. As a value, you will specify the url of the target page.

<div class="button" linkto="/pages/home/home.html"></div>

In case you want more, linkto can have arguments, just like tap. The arguments will be passed to the navigation

<div class="button" linkto="/pages/home/home.html" linktoargs="obj:{ index: 3}"></div>

interacting with DOM

When writing a WinJS page, you end up writing tons of DOM selector to get instances of the controls, or DOM element that you will interact with. Your page end up looking a bit ugly with things like this all around the place :

var listView = page.element.querySelector('#myListView').winControl;

If you are wandering why scoping selector to page, go have a look at some WinJS best practices.

WinJS Contrib page introduce a mecanism to map elements and controls as properties on your page by adding a « member » attribute on the node. You could leave it blank, and the name of the property will be the « id » on the node, or you could specify the name of the property.

<div class="button" member="btnClose"></div>

If the element has a control attached on it, the property will hold the control, otherwise it will be the DOM element :

<div member="listview" data-win-control="WinJS.UI.ListView"></div>

then you will use it magically in your page :

page.listview.itemDataSource = new WinJS.Binding.List().dataSource;

If you prefer having the declarations in your page, we still have some goodness for you. The page control has « q » and « qAll » functions for running querySelector and querySelectorAll from your page element, which is at least shorter than the full syntax. The « qAll » function is also kind enougth to map the result as an array, so you can use forEach on the result.

    //do something

preventing leaks

The first cause of memory leaks is unreleased event listeners having a closure on some global variable. To ensure releasing your listener without adding some (ugly) extra code to your page, WinJS Contrib provide an event tracker object. The page control holds an instance of it that you can use. You use this object to attach functions to event listeners, and when you dispose the tracker, it will dispose all attached events automatically. If you use the page instance, the page will take care of releasing the tracker.

instead of

ctrl.showBinded = ctrl.show.bind(ctrl);    		 
ctrl.searchInput.addEventListener('focus', ctrl.showBinded);
ctrl.searchInput.removeEventListener('focus', ctrl.showBinded);

you have :

ctrl.eventTracker.addEvent(ctrl.searchInput, 'focus', ctrl.show.bind(ctrl));    

does all this really help ?

Well, see for yourself by looking at the small sample application that we made. You have one version with WinJS only, and another with WinJS Contrib.

This application is for searching Flickr. If you look at the code for the search bar control and compare the pure WinJS and the WinJS Contrib version, you will see a real difference (it removed 45 lines of code on 154…)

Next time we will see other features enabled by the WinJS Contrib page

2 réponses à “WinJS Contrib Basics

  1. Pingback: WinJS and ES6 | Le Post de MCNEXT

  2. Pingback: Using WinJS controls as webcomponents | Le Post de MCNEXT

Laisser un commentaire

Entrez vos coordonnées ci-dessous ou cliquez sur une icône pour vous connecter:

Logo WordPress.com

Vous commentez à l'aide de votre compte WordPress.com. Déconnexion / Changer )

Image Twitter

Vous commentez à l'aide de votre compte Twitter. Déconnexion / Changer )

Photo Facebook

Vous commentez à l'aide de votre compte Facebook. Déconnexion / Changer )

Photo Google+

Vous commentez à l'aide de votre compte Google+. Déconnexion / Changer )

Connexion à %s

%d blogueurs aiment cette page :