I have a very unusual behaviour when hovering my menu here:
http://corecruit.w6.ext.starberry.com/employers
It shudders and the links don't work.
You can see this by hovering over the 4 main links or any of the sub-nav links.
Any ideas?
I'm seeing an error in console: Uncaught SyntaxError: Unexpected token << jquery.main.js:1.
It looks like the hover state is being applied on mousemove when it should be applied on mouseover or mouseenter. Maybe it's an event bubbling issue that causes the mouseover event to fire constantly.
Upon refreshing the page the problem no longer appears. Did you fix it?
Related
Uncaught (in promise) Error: Found malformed component comment at Blazor:
{"sequence":0,"type":"server","prerenderId":"aa848eac6d3143c2b938c451316a207c"
Alright, so I keep running into this error. It's very inconsistent, and not always reproducible. I'm on .NET6, using Blazorise, and every once in a while on a page load, I get this error. The result is that the page will load, but without the sidebar, leaving me no ability to navigate. So far my notes of the error are as follows:
It happens in all browsers.
It happens more often when my screen size is smaller, or the window is not full screen.
It only seems to happen on screens that have an EditForm (Blazorise component).
It only seems to happen on screens that have multiple columns.
If I refresh the page between 2 and 370 times, eventually, the sidebar will load as usual.
Research into this issue has shown that sometimes it happens when there's a double body tag or something, but I have not seen anything of the sort on any of my pages thus far when examining the content in the browser. Does anyone have any tips or tricks on this?
Any input appreciated, thank you.
I am not sure whether its a usual behavior or not, but, on [WKWebView goBack] navigates to previous page from the Back-Forward list. but, in my case on very first back action, it does not calls didFinishNavigation delegate method.
So, if it's usual behavior then how to detects that page loading is completed after back the action.
Are you seeing this on a specific iOS version? Im seeing this too, only and error is being thrown and the page just stops loading, hence didFinish... not being called.
I've only started seeing this since 13.4 (or 13.4.1)
--- EDIT ---
I do have a workaround.
This appears to happen more often on slower networks and when the user interacts before a page is fully loaded. Normally if a user taps back before the page they visited finished loading then a "Cancelled" error is thrown.
What seems to be happening now is, if the page they are returning to never finished loading at the point they left, then the error is thrown again, but this time referencing the page they are returning to. and more importantly the web view is no longer loading at this point.
what we have done is monitor this and check to see if the page is not loading, and if it isn't, reload it. This seems to work and while not ideal, its buying us some time to investigate a proper fix.
In addition, I put a retry limit on this and we show an error with a reload button after the third attempt, though in practice we've never seen it go beyond the first reload attempt.
The fact that the error is thrown almost instantaneously means the reload adds no perceivable delay for the user.
I even see this on the simulator. Navigate to google.com, click privacy, then goBack and while the page loads normally, navigation reports a failure with error:
Error Domain=NSURLErrorDomain Code=-999 "(null)" UserInfo={NSErrorFailingURLStringKey=https://www.google.com/, NSErrorFailingURLKey=https://www.google.com/, _WKRecoveryAttempterErrorKey=<WKReloadFrameErrorRecoveryAttempter: 0x600003904e40>}
What you need to do is capture this with the didFailNavigation: delegate method and handle it gracefully (if you see a consistent error like the above you can even treat it exactly the same as success).
Is it possible to alter DOM in beforeunload handler? Chrome seems to apply DOM modifications after the user pressed "stay" in browser alert about leaving the page, and I want a part of the page to be hidden while this alert is visible.
See the demo:
<script>
window.onbeforeunload = function(e){ document.getElementById('a').style.display = "none"; e.returnValue = "adios";}
</script>
<div id=a >This text should disappear when the user tries to leave the page</div>
https://jsfiddle.net/qmatic/6sLp7rmq/
The only solution I have so far is to animate the div to opacity:0 and to have a code in setInterval that constantly resets this animation. While the code is running the div is always visible. Chrome stops all the code when it shows an alert, so the animation finally runs till the end and hides the div. But it's a terrible solution - I'm constantly updating the DOM to reset the animation. Does anyone have any better ideas?
The process being executed by Chrome appears to be:
1) call the handler, waiting synchronously for its return value
2) if the return (event.returnValue) is suitably undefined, continue the unload
3) otherwise present a modal popup of some kind, getting the user's permission to cancel the navigation.
4) if navigation is cancelled, stay on the page and apply a refresh cycle to the page view.
I have stepped through an onbeforeunload handler in the debugger, and when I do so, style changes WILL take effect before the modal dialog appears instead of after it closes. This seems to indicate that using the debugger introduces additional DOM refresh cycles that would otherwise not occur in normal running. I am trying to find a way to get one of these refresh cycles to occur programmatically from within the handler, but so far no luck.
In other browsers, there seems to be refresh cycles inserted between steps 1-4 above, giving the desired behavior.
The problem I have is pictured in this fiddle I created:
If you click on the Order coffee button, I've forced a Runtime Exception to be thrown within the doAfterCompose method of the modal window's controller. As you can see the modal window gets appended at the bottom of the page. This, aside from being ugly, allows the user to click again the Order Cofee button, which causes the famous "Not unique ID in space" error.
Is there any way to prevent the window from being created when an Exception is thrown?
You can call setPage(null) method on your component.
Actually, I want to store some data in background page and the popup page just show that part of data say Data as a div element created in background page document.createElement("div"). Here, the background page will register some listeners to the tab update and change the Data elements accordingly. What the popup will do is to get that Data and appendit use the document.appendChild(Data).
(The purpose I intend is this will cause the popup changes immediately while the tab updage is triggered.)
However, the elements are shown as usual, what I am facing very headache is I have registered the onclick for the div object in backgroundpage as onclick="chrome.extension.getBackgroundPage().somefunc()". However, the first time, all the click will triger the right behavior but after the popup loses foucs and get focus again, all the click won't work.
I try to put something like change the onclick="somefunc()" and leave the func within the script of popup page. And there I want to log whether it is called by console.log("clicked"). Here, something unbelievable happens, the function is succefully trigerred BUT the console is null here, I cannot even call chrome.extension.getBackgroundPage() as well.
Here are a list of questions, maybe very hard to express for me...
1. Whether I can reuse the DOM element from the background page to the popup page directly by appendChild(chrome.extension.getBackgroundPage().getElementById()?
2.Will the onclick event registered in the background page still work in the popup pages?
3. What's the problem with the problem I am encountering? I have tried many ways to find out the reason but all in vain at last...
Best Regards,
If you need any more information, please let me know.
(PS: I am wonderning if it is called something like the event propogation, however, I am not an expert in this two pages communicating...)