I am writing a chrome extension that records your actions like ( mouse click, keyboard keyup ). The idea of the extension is to help me and my colleagues to reduce the boring testing of our web based project. I made it to record the events and store it on the dev server as mysql so i can use or share to them. But the problem is replaying the saved actions.
So how if there is a way to force mouse move, mouse click events. Can it be done from flash,java or something like that.
PS. The project is Extjs but i want to make the extension useful for developer using other frameworks and publish it.
Consider using Selenium for this. It has support for many languages, and you can script your whole test with it. You can for example set it to click on an element, wait for something to happen or fill text boxes.
Imagine some random website controlling your mouse ... not cool, is it? (That's why you cant force mousemove via javascript)
However, you can trigger clicks on elements. To achieve that, you need to save the event(mouse-over|out/(dbl)click/whatever) and the according element (in the eventfunction: this). That should be sufficient to simulate theworkflow.
jQuery-Example:
$('#item').click();
$('#item').trigger('click');
vanilla javascript:
document.querySelector("#item").click();
Related
so I want to create cross-platform app that will be invisible overlay over whole screen. When user interacts with whatever is under, I want to catch the event (mouse location, mouse click etc) so I can have info of what user attempted and where, and then "pass" the event through. Also, I might wanna display something in overlay as result of the event.
Coming from .NET, I understand how I would hook into mouse events. I also understand that it is possible to simulate user clicks with Mouse class in .NET Core. However I was wondering if this is possible with electron, as I was checking it for cross-platform solution.
I checked some topics (like this click-through discussion) that mostly deal with passing events through overlay, but I am not sure those do this catching-then-passing. So is there something like that and is it actually possible with only electron/nodeJS? Or would I have to use additional libs, and would they be different for Win/Linux/Mac?
I checked at:
Overlay Electron and
electron-overlay-window
but both seem complicated with attaching to game and following the window respectively. I was hoping it is possible and simple somehow.
IMHO, it's is possible, but will require a ton of OS-specific code and possibly some native modules.
I suggest to start with https://www.electronjs.org/docs/api/browser-window#wingetnativewindowhandle and https://www.electronjs.org/docs/api/browser-window#winsetignoremouseeventsignore-options
Since I'm searching for an answer for a while now and I'm still without any clue, I'll just describe my actual problem:
I need to build up automated touch/mouse gestures (tap,drag,pinch...) which I can fire on a webpage in order to test touch frameworks and their performance. Thus I want to trigger "global" touch/mouse events on a webpage with JavaScript without dispatching them from a specific element.
Does anyone know how I could achieve this or how these events are delegated in general?
If you need this just for the sake of emulating mobile touch actions on your browser, Chrome already has a tool for that. Check out Mobile Emulation.
I'm currently working on an automation script (JavaScript) for an iOS app. I'm not a programmer, just a tester who wants to seek bugs.
Software concerned: Xcode, Instruments.
Here is the problem: there are buttons in my application that had no name, which is important for automated tests, so I gave them a Label in the Accessibility field of Xcode.
But Instruments still doesn't recognize the name of the buttons.
e.g.: What I want, but doesn't work:
target.frontMostApp().mainWindow().scrollViews()[2].buttons()["Settings"].tap();
What I don't want, but works:
target.frontMostApp().mainWindow().scrollViews()[2].buttons()[1].tap();
This is just one of many situations I've encountered. I chose a simple one and it could seem trivial, but some others are really problematic.
Any idea how to fix this?
Do you create the code by typing it, or do you capture it?
If you havn't captured it: you can do that by using that small record button in the JavaScript editor. After capturing, Instruments shows you the code, that would automate the actions you did "by hand". Then you can find out, if there is a different way to adress the button. This way you can find out how to trigger your button I havn't set up my GUI elements with the acessibility properties, because this way I can find out how to trigger the elements.
It sometimes takes a few tries to get things running.
Also, if an animation is applied when the screen is changing, a delay is needed before you can trigger the next GUI event:
target.delay(1);
This creates a delay of 1 second until the next event is fired.
Hope this helps
Using Chrome's developer tools I am trying to determine what jQuery function is hooking an input button on the page for debugging purposes. I usually just keep searching until I find it, but I figured I'd ask this time.
Is there a way to find a jQuery button hook for a specific button in Chrome? I've tried looking through the Event Listener Breakpoints, but can never seem to find the right thing to pause it.
Basically, I need to know what jQuery / Javascript is being executed after the button is clicked.
The hooks are implemented in the application like so:
$('.button_class').click(function (){
$('#button_id').click(function(){
etc...
try this :
$(yourbutton).data('events');
Depending on the number of events/timers on the page this doesn't always work. But you can try "pausing" before clicking the button you want to debug in the JavaScript debug window. That way the debugger will pause on the next line that executes. The thing that occasionally prevents you from using that is if there is a "hover" or mouse move/in/out event tied on an element you have to pass over to get to the button (including the button itself). In that case I just remove those events (if I can) until I get the one I want. The event listener breakpoints would be more ideal but they're sometimes difficult when using jQuery or another library, I've actually put in a feature request to the Chrome Dev Tools team to address this very issue. (allowing you to specify what files are "yours" and only "breaking" in those specific files)
good luck -ck
How I do monitor all the clicks and inputs on my page? I would ideally want to make a video of what the user does on my app; the way they uses my app.
How do I go about this?
Any links or demos would be really great.
Edit : What I have in mind is:
log all the events into a cookie , and send it to the server in frequent intervals.
then startup a Firefox session and trigger these events on the page . and capture it as a video.
Edit: Basically I have an app, and I want to see what the user does on the app, so I will monitor their clicks and inputs (that's pretty much what they do on the click), and then trigger those events back on my web app. So I will pretty be able to replicate their actions on my page. Now the question is, how do I capture all the necessary events, how do I serialize and trigger these events for later?
Not sure that you can make a 'video' unless you tracked every single movement of the mouse, which would extremely unfriendly, probably take a lot of processing time and raise ethical questions.
If you want to know how a user interacts with your site so that you can improve it (presumably) then you need to get into web analytics.
There are many options. My personal favourite is Google Analytics
With JavaScript you can create what's known as a heatmap of use clicks:
http://tympanus.net/codrops/2010/02/08/a-jquery-heat-map/
You can't capture a video of the users desktop session using JavaScript.
I remember seeing ClickHeat (or a similar variant) a long time ago. Had no experience with it, so I can't help with pros and cons, and it doesn't answer your question entirely (don't think it does keystrokes, although maybe I'm wrong, like I said I don't really know).
Might not be a total solution, but should be a good starting point.
In a similar vein it wouldn't be hard to capture mouse position every so often along with a timestamp and recreate a given user's actions. Generating a general/average movement is much more difficult with this type of data, but it's useful on the single-unit basis. So too with keystrokes, they're all just events.