i have tried the following :
t.switchTo("frameID");
var ele = Selector("#ele");
await t.click(ele);
another attempt :
Selector(() =>{return document.getElementById("frameId").contentDocument.getElementById("#ele")});
This throws the following error : Function for selector should return DomElement, NodeList, HtmlCollection, NULL or undefined.
The HTML DOM is as follows :
<html>
<body>
<frameset>
<frame id='frame1'>
#document
<html>
<body>
<p id='ele1'>Hello World</p>
</body>
</html>
</frame>
</frameset>
</body>
</html>
The frame and frameset features are marked as obsolete in the W3C Recommendation. As TestCafe is based on HTML5 standards that do not support frameset, TestCafe does not support frameset either. So, we'd like to focus on up-to-date features, which will give the best experience in testing modern web applications.
As for the example you shared, you can use this test:
test('test', async t => {
await t.click('#ele1');
})
Related
When pages with Vue components are translated via chrome's translate option, the vue components stops re rendering and updating the view.
Ex: Translate https://v2.vuejs.org/v2/guide/#Handling-User-Input on chrome using the translate option from chromes's context menu into a different language, the reverse message demo stops working.
Since Google translate plugin updates DOM outside of Vue's control, this is sort of expected. Looking for any work arounds that let both co-exist. The sections can be marked with "notranslate" class but that would mean it is no longer translatable.
React inspite of being based on virtual DOM, works even with DOM being modified by translate plugin.
A possible workaround is to use the Vue special attributes key (as described here) and ref (as described here).
Here it is an example I did, starting from the link you provide above:
<!DOCTYPE html>
<html lang="en">
<head>
<meta charset="UTF-8">
<title>Vue.js Reverse Example</title>
<script src="https://unpkg.com/vue/dist/vue.js"></script>
</head>
<body>
<div id="app-5" class="demo">
<!-- Adding Vue attributes here -->
<p :key="message" ref="msg">{{ message }}</p>
<button v-on:click="reverseMessage">Reverse Message</button>
</div>
<script>
var vm = new Vue({
el: '#app-5',
data: {
message: 'Hello Vue.js!'
},
methods: {
reverseMessage: function () {
// vm.$refs.msg.innerText retrieves the translated content
this.message = vm.$refs.msg.innerText.split('').reverse().join('')
}
}
})
</script>
</body>
</html>
As you may notice, the DOM element where you'd like to maintain the Vue reactive behavior (i.e.: the reverse operation here), has been enriched with both a key attribute and a ref one.
The idea here is to use:
:key to force replacement of the element instead of reusing it;
ref to register a reference to the element: it is used in reverseMessage method in order to get the translated innerText content after the Google translation is performed.
For sure, this workaround would affect the performance, but at least it provides the expected behavior (i.e.: the reverse function properly working also after a page translation).
My component is loading fine but the styles are not loading, nor are the events firing. I am following the documentation and no errors are being thrown but it seems I might be missing something fundamental here?
View template rendered with res.marko:
import Explanation from "./components/explanation.marko";
<!DOCTYPE html>
<html lang="en">
<head>
...
</head>
<body>
...
<include(Explanation, input.explanation) />
...
</body>
</html>
explanation.marko file:
class {
onExplanationClick() {
console.log("Explanation clicked");
}
}
style {
.explanation-paragraph {
color: red;
}
}
<div id="explanation" on-click('onExplanationClick')>
<for (paragraph in input.content)>
<p class="explanation-paragraph">${paragraph}</p>
</for>
</div>
Server side:
app.get("/explanation/:id", async function(req, res) {
var explanation = await findExplanation(req.params.id);
var template = require("../../views/explanation/explanation.marko");
res.marko(template, { explanation, user: req.user });
});
Also using marko/node-require and marko/express.
You will need to integrate a module bundler/asset pipeline. In the sample marko-express app we are using Lasso (an asset pipeline + JavaScript module bundler).
There's also another sample app that integrates Webpack: https://github.com/marko-js-samples/marko-webpack
The Marko team supports both Lasso and Webpack, but we recommend Lasso because it is simpler and requires minimal configuration.
Please take a look at the marko-express app and feel free to ask questions in our Gitter chat room if you get stuck: https://gitter.im/marko-js/marko
I want to create a site with a 360 video thats streamed from my computers 360 cam (ricoh theta s) and uses the possibilites of a-frame. I did not have any luck with embedding youtube livestreaming within a-frame though. I would have a webserver and the stream would not be really public so maybe there is some solution that involves selfhosting? Does anybody have any experience with achieving sth like this? i cant find anything related so far and this plays a crucial role in my robotproject...
EDIT 1:
I might specify even more after some research:
it would be perfect if a-videosphere and a-video would support sth like hls or mpeg-dash - streams. since that would need some kind of player for chrome and android stuff i think the easiest route would be to support a motionjpg-support because they are very easy to create. flashstreams can be good too but i dont think there is a future for that.
is there anything like that in the planning by someone because i am quite sure that nothing like that exists yet... i took 2-3 days of researching and find nothing about that topic... just a getUserMedia for webcam example showed up but its not good for my purpose.
another approach i could live with would be a routine that just autoreloads pictures in a-sky (if possible without flickering). i tried to integrate javascripts that are supposed to do that within normal divs and so but nothing worked...
or did somewhere someone get a stream of any kind running in a-videosphere and if so how?
EDIT 2:
i got it working... somehow... not really but looks promising...
the stream is provided by "yawcam" which uploads a new picture every second to my ftp. if i now click on the red sphere the script starts via addEventListener 'click' and the content of a-sky gets updated... is there any way to make a loop out of that script so one doesnt have to click anymore and it just update itself every second?
<head>
<meta charset="utf-8">
<title>joeinterface</title>
<meta name="description" content="360 Video — A-Frame">
<script src="https://aframe.io/releases/0.3.2/aframe.min.js"></script>
</head>
<body>
<script>
AFRAME.registerComponent('set-sky', {
schema: {default:''},
init() {
const sky = document.querySelector('a-sky');
this.el.addEventListener('click', () => {
sky.setAttribute( 'src', this.data + "?" + Math.random());
});
}
});
</script>
<a-scene>
<a-camera position="0 0 0">
<a-cursor color="#4CC3D9" fuse="true" timeout="10"></a-cursor>
</a-camera>
<a-sphere color="#F44336" radius="8" position="-8 2 -8" set-sky="image1.jpg"></a-sphere>
<a-sky></a-sky>
</a-scene>
</body>
so here is the solution for the proposed "kind-of" solution...
the picture "out" is updated via "yawcam" on my server every second and updated in the a-sky tag - even without flickering... so no sound but at least kind of a live-video-feed in a-sky.
<head>
<meta charset="utf-8">
<title>joeinterface</title>
<meta name="description" content="360 Video � A-Frame">
<script src="https://aframe.io/releases/0.3.2/aframe.min.js"></script>
</head>
<body>
<script>
AFRAME.registerComponent('set-sky', {
schema: {default:''},
init: function() {
this.timeout = setInterval(this.updateSky.bind(this), 100);
this.sky = this.el;
},
remove: function() {
clearInterval(this.timeout);
this.el.removeObject3D(this.object3D);
},
updateSky: function() {
this.sky.setAttribute( 'src', this.data + "?" + Math.random());
}
});
</script>
<a-scene>
<a-camera position="0 0 0">
<a-cursor color="#4CC3D9 " fuse="true" timeout="10"></a-cursor>
</a-camera>
<a-sphere color="#F44336 " radius="2" position="-8 2 -8"></a-sphere>
<a-sky set-sky="out"></a-sky>
</a-scene>
</body>
I'm defining page with some vars and methods. Then I wanna use it in html markup (for example data-win-bind="textContent: myPage.variable). How can I access page variables in html markup?
In the JavaScript code behind your page, say default.js, you'd include your ViewModel for the data binding, something like:
(function(){
WinJS.Namespace.define("MyModel.myPage", {
variable : null
};
MyModel.myPage.variable = 'foo';
})();
Then in when the page is activated (in default.js), you'll need to initialize the bindings with a call like
WinJS.Binding.processAll(document.body, MyModel);
There's quite a bit more functionality available though, so this is just a simplistic one-way binding case that should get you started. For more info, check out the Quickstart: binding data and styles.
You have to use javascript here's the skeleton:
<html>
<head>
<script type="text/javascript">
//Code goes here
</script>
</head>
<body>
</body>
<html>
I have a test jsp with:
<head>
<script src="http://ajax.googleapis.com/ajax/libs/dojo/1.5/dojo/dojo.xd.js" type="text/javascript">
</script>
<script type="text/javascript">
dojo.require("dojo.widget.Tree");
dojo.require("dojo.widget.TreeSelector");
dojo.require("dojo.widget.TreeNode");
dojo.require("dojo.widget.TreeContextMenu");
</script>
</head>
<body>
<div dojoType="TreeSelector" widgetId="treeSelector"></div>
<div dojoType="Tree" widgetId="treeWidget" selector="treeSelector"toggler="wipe">
<div dojoType="TreeNode" widgetId="1" title="First node" isFolder="false"></div>
<div dojoType="TreeNode" widgetId="2" title="Second node">
<div dojoType="TreeNode" widgetId="2.1" title="Second node First Child"></div>
<div dojoType="TreeNode" widgetId="2.2" title="Second node Second Child"></div>
</div>
<div dojoType="TreeNode" widgetId="3" title="Third node" isFolder="false"></div>
</div>
This will not work in any browser.
I thought this would be easy, it seems the dojo library is not being downloaded/found?
Do I need to do anything else?
Also, my IDE, JDeveloper, reports that the attribute "dojoType" is not defined on element div.
I have to say, this example looks like it is taken from a very old version of dojo, but you're trying to run it against Dojo 1.5. That most likely won't work. dojo.widget hasn't existed since...0.4, 0.9 maybe.
You may be right in your comment to the previous answer in that no parseOnLoad: true was necessary in the original example, but I'd also assure you that that example was not running any version of Dojo anywhere near what you're running it with.
Based on what you're looking at there, you may want to start somewhere like here: http://www.dojotoolkit.org/reference-guide/dijit/Tree.html
I'm not sure what the default behavior is when it's not present, but you probably need to define a djConfig with parseOnLoad set to true (or call the parser directly). See the following links for more information:
http://docs.dojocampus.org/djConfig
http://dojocampus.org/content/2008/03/08/the-dojo-parser/
Follow the:
Google AJAX Libraries API Dev Guide,
and the Google API Loader's Guide.
You need to:
register for an API key (or use a direct link as you did),
if not using a direct link but google.load, you need to defer the execution of your code using an onload callback.
Personally, I would just do something like:
within the <head> section of my.html:
<script type="text/javascript" src="http://www.google.com/jsapi?key=MY_API_KEY_GOES_HERE"></script>
<script type="text/javascript" src="my.js"></script>
in my.js:
google.load("dojo", "1.5", {
uncompressed: true
});
function OnLoad() {
/* do stuff here */
}
google.setOnLoadCallback(OnLoad);