Bug Tracker

Opened 12 years ago

Closed 12 years ago

Last modified 10 years ago

#7486 closed bug (wontfix)


Reported by: anonymous Owned by: anonymous
Priority: low Milestone: 1.next
Component: css Version: 1.4.4
Keywords: regression Cc:
Blocked by: Blocking:


Until now "$('#x').css('height')" returned 'auto', if no css-height was defined. Now it returns always the height, e.g. '35 px';

Change History (19)

comment:1 Changed 12 years ago by addyosmani

Owner: set to anonymous
Status: newpending

Please supply a valid test case reproducing what you're trying to achieve, ideally on jsFiddle.net. This will allow us to investigate the problem described more thoroughly.

comment:2 Changed 12 years ago by anonymous

Status: pendingnew

<p id="test">blabla</p> <script> alert( $('#test').css('height') ); </script>

Until jQuery 1.4.3: "auto" jQuery 1.4.4: e.g. "22px"

comment:3 Changed 12 years ago by danheberden

Probably related to fixes for #7225 http://bugs.jquery.com/ticket/7225

comment:4 Changed 12 years ago by danheberden

Can you please provide a live demo via jsfiddle.net? That helps a ton in looking into the problem :)

comment:5 Changed 12 years ago by snover

Component: unfiledcss
Keywords: regression added
Status: newopen

This was noted on the commit notes for [53396b879bd29c090824da182e3cf69158829f82], though for whatever reason the ramifications of that change did not make it into the release notes. Now that there is an official report of this breaking someone’s stuff, I’m going to mark this and maybe we can revert this (though it is unlikely that the behaviour will change).

comment:6 Changed 12 years ago by anonymous


Yes, it broke my app, but it was easy to fix. Maybe you should update the release notes and the API-docs.

Thanks Marcus

comment:7 Changed 12 years ago by jitter

Keywords: needsdocs added

comment:8 Changed 12 years ago by Matt Palmerlee <[email protected]…>

I have this same problem, it seems that this problem started in jQuery 1.4.3, not 1.4.4

Please see my jsFiddle test case here: http://jsfiddle.net/mpalmerlee/DnRUG/

In 1.4.2 css("height") returns "auto" and in 1.4.3 and 1.4.4 it returns 0px in both IE and FF

comment:9 Changed 12 years ago by Matt Palmerlee <[email protected]…>

In addition to this not working with height:auto, My test case doesn't work with height:100% in the div either.

Here is a forked jsFiddle test case with height:100%


This also works as expected in jQuery 1.4.2

comment:10 Changed 12 years ago by mpalmerlee

Correction to my last comment, here is the jsFiddle for the height 100% version: http://jsfiddle.net/mpalmerlee/rrMTZ/1/

This breaks my app and is not as easy to fix for me as it was for anonymous, If I put in Height:16px for the div, my replaced input box looks ok in FF, but is too small in IE (IE wants input boxes ~ 20px high)

And if I specify Height:16px Firefox looks fine, but the input box is smaller in IE so this method doesn't work for me as well:


Last edited 12 years ago by mpalmerlee (previous) (diff)

comment:11 Changed 12 years ago by snover

mpalmerlee, investigate the box-sizing CSS property.

comment:12 Changed 12 years ago by mpalmerlee


Thanks for the pointer to the box-sizing CSS property, I was having trouble reproducing my application's behavior on jsFiddle in IE (the form elements looked fine when replaced when specifying an exact size like 16px). I finally realized that my application uses IE Quirks mode which I'm not able to simulate using jsFiddle, and it seems that the box-sizing properties don't work with IE quirks mode.

I've uploaded a test page to a server so that you can see the issue I have in IE: http://ea1.conversive.com/jsFiddleJQueryHeightBug.html

Since it is not feasible to change my application from quirks mode because it would mean a ton of layout refactoring, and since I can't find another work around, I won't be able to upgrade from jQuery 1.4.2 to 1.4.4 until css('height') works how it does in 1.4.2 (height:auto or height:100% elements return auto and 100% respectively).

Thanks, -Matt

comment:13 Changed 12 years ago by timmywil

#8741 is a duplicate of this ticket.

comment:14 Changed 12 years ago by timmywil

Priority: undecidedlow

jQuery.style gets uncomputed and looks like jQuery.css always gets computed. The question I think is, what should jQuery.fn.css return and is it specific to certain properties?

comment:15 Changed 12 years ago by mikesherov

Timmy, for what it's worth, isn't this exactly what cssHooks are for? Theoretically, a user could easily a cssHook to get the old behavior back:

Something like this... (untested)

jQuery.each(["height", "width"], function( i, name ) {
	jQuery.cssHooks[ name ] = {
		get: function( elem, computed, extra ) {
			return jQuery.style(elem, name);

		set: function( elem, value ) {
						return jQuery.style(elem, name, value);

comment:16 Changed 12 years ago by timmywil

Resolution: wontfix
Status: openclosed

I think we can safely say that this won't change, but a note on the docs would probably be good to indicate the change in behavior.

comment:17 Changed 12 years ago by anonymous

I need the ability to reset an element to 'height: auto' in some cases, but only if it was 'height: auto' or didn't have a height set to begin with.

jQuery.style still doesn't work for this. There does not appear to be a way to get information about styles that are not applied inline. In addition, I don't think that saying that this is "easily" achievable with an undocumented function is not really the best solution anyway...


comment:18 Changed 12 years ago by Rick Waldron

@anonymous - see above, this can be resolved with a custom cssHook

comment:19 Changed 10 years ago by mikesherov

Keywords: needsdocs removed
Note: See TracTickets for help on using tickets.