jquery 1.8.0 $(window).height() and .width() is broken and does not work correctly (Firefox, IE8)

[Wikinaut]

as mentioned in  https://github.com/jquery/jquery/pull/764#issuecomment-8120831 the function of $(window) is broken compared to jquery 1.7.2

How to reproduce

================

See code in  http://dpaste.com/hold/793316/

When running 1.7.2 everything is perfect.

When running 1.8.0 IE8 returns $(window).height() and .width() == 0 (zero)

When running 1.8.0 Firefox returns $(window).height()= 8 or random values, and .width() okay (zero)

So please check your 1.8.0 code for $(window). The jquery 1.8.0 is broken !

See  http://stackoverflow.com/questions/1860524/javascript-how-to-find-out-width-and-height-of-viewpoint-in-browser-window for a *working* solution, tested with Safari, Firefox 14, Firefox 15, IE8, Firefox 10 ESR

[Wikinaut]

remark:

a working code which DOES NOT use jquery 1.8.0 for the viewport size detection is here  http://dpaste.com/hold/793319/

[timmywil]

Thanks for taking the time to contribute to the jQuery project! Please provide a complete reduced test case on jsFiddle to help us assess your ticket.

Additionally, be sure to test against the jQuery git("edge") version to ensure the issue still exists. To get you started, use this boilerplate:  http://jsfiddle.net/FrKyN/. Open the link and click to "Fork" (in the top menu) to get started.

[dmethvin]

That's cool, I double checked the code there.

It's in Quirks mode. The document type is invalid. Did you intend that?

[dmethvin]

Just to clarify, in that thread I asked.

@Wikinaut this isn't the place to report anything. Can you go to our bug tracker, http://bugs.jquery.com and provide a test case? Be sure you're not in Quirks mode, which jQuery has never supported.

Honestly. A couple of years ago I checked with John Resig to confirm that we never ever supported Quirks. Yes, sometimes things kinda sorta work. But we never supported Quirks mode, and have never run unit tests in them.

 http://docs.jquery.com/Won't_Fix

[dmethvin]

We need a test case that does not use Quirks that demonstrates this problem.

[farmdawgnation]

So, GitHub's recent notifications shuffle means I wasn't getting notifications from the Pull Request thread, so I'm just joining the conversation. Sorry, I'm late to the game. I'm the engineer who wrote the patch that allegedly broke IE8.

I wrote out this simple test example (fair warning, it uses an alert box :P):  http://jsfiddle.net/farmdawgnation/59XbC/

If what you're saying is true, then I should be getting different results in Firefox and IE8 when this code runs under jQuery 1.7.2 and jQuery 1.8.0 (by flipping the select drop down). It would seem that jQuery 1.7.2, jQuery 1.8.0, and jQuery Edge all yield the same results for $(window).height() under IE8 and Firefox when standards mode is in use.

I can only reproduce the behavior you're describing by running IE8 in quirks mode. I, too, suspect that is the cause of your troubles.

[mtkopone]

Google Chrome + jQuery 1.8.1, a window with a vertical scrollbar:

>> window.innerHeight --> 500 (the viewport height)

>> $(window).height --> 1000 (the document height)

Can't reproduce in jsfiddle, since I can't get it to show a scrollbar on the window.

[farmdawgnation]

Does my updated jsfiddle represent what you're testing?  http://jsfiddle.net/59XbC/1/

I manually set the height of the body such that the results pane scrolls, and I'm still seeing the same values for window.innerHeight and $(window).height().

[mikesherov]

This is definitely the Quirks mode issue. I see no reason to continue having this continue be pending or open. Can we please close?

[Wikinaut]

I am checking this currently. IE8 shows "console undefined" for the console.log() expression.

Should I replace this with an alert, or what is your recommendation for testing with IE8 ?

[farmdawgnation]

Ah, sorry, I updated that jsfiddle because multiple any alerts are annoying. But yes, for testing with IE8 change those console logs to alerts.

[dmethvin]

Bring up F12 Developer Tools in IE8, console.log is only available in IE8 while dev tools are open.

Still waiting on a test case that doesn't require Quirks to fail.

[Wikinaut]

Problem is solved now.

Report:

• I was until now fully unaware of Quirks mode, pls. apologize.
• I already knew from several jQuery reports, that jQuery never aimed to support Quirks mode.
• Until today I always used this document type declaration

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">

• I found that IE8 is bitchy and requires this form

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">

• My problem is reproducible in that way that a change between these two dtds makes the problem of $(window).height().width() not working in IE8 apparent or not:

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN"> - does not work

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd"> - works

Solution: Everyone with (such) jquery problems should be directly pointed to double-check whether the dtd is really

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">

You can close this bug now.

[anonymous]

The problem persist with firefox with all dtd.

[dmethvin]

#12942 is a duplicate of this ticket.

[dmethvin]

#12944 is a duplicate of this ticket.