296408 - CSS Class styles with a hover effect such as .test:hover {} fail on a div.

Status: RESOLVED INVALID

(Core :: CSS Parsing and Computation, defect)

Description

[lhunath]

User-Agent:       Opera/8.0 (X11; Linux i686; U; en)
Build Identifier: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.7.8) Gecko/20050530

.test:hover { background: red; }
<div class="test">Testing div</div>
Will not become red on Hover.
However, when we change .test:hover into div.test:hover; it does work!

Reproducible: Always

Steps to Reproduce:
1. Have a .classname:hover {} style
2. Hover the element in question
3. No hover effects are applied

Actual Results:  
No hover effects are applied

Expected Results:  
The specified CSS rules should be applied (in the example, the div should 
receive a red background).

Comment 1

[Steve England [:stevee]]

URL http://www.lyndir.com/bugs/two.html does not work for me.

Comment 2

[Martijn Wargers (dead)]

Sounds like bug 169078.
See also http://www.mozilla.org/docs/web-developer/quirks/quirklist.html

Comment 3

[lhunath]

The URL suffers resolution problems, it's basically exactly the same as the 
example I provided.

Comment 4

[Andrew Schultz]

comment 2 is correct.  the document at the URL has no doctype declaration. 
adding a proper doctype declaration to the page at the URL fixed it.

resolving INVALID -- the rendering is as intended.

Comment 5

[lhunath]

Hold on, you're saying you're not going to fix it; because it's operating in 
Quirks Mode? Or are you really saying in Quirks Mode, div.class and .class 
applied on a div, are two different things...?

Comment 6

[Hixie (not reading bugmail)]

It is intentional that :hover doesn't work on <div>s in quirks mode, for
compatibility with the many (many!) sites written for IE that depend on this.

Comment 7

[lhunath]

In theory, that evil behaviour is mostly of Mozilla's doing. Should a 
webdesigner bother to write a :hover on a DIV in the first place, it means he's 
got a good browser to test it with, or a good browser he's targetting. The only 
way we could assume the designer thinks is:
"IE doesn't support hover, so it's ok to make hovers on a DIV where to might 
screw up IE compatibility; since the Mozilla -being compatible with IE- ignores 
it too. So I don't need to make sure my things are Standardised, Mozilla'll make 
my site work anyway."

In a way it's good I suppose; for all the Mozilla users who can't stand seeing a 
site different than in IE.

Comment 8

[Hixie (not reading bugmail)]

Unfortunately no. People put :hover rules on all elements and expect them to
only apply to links, and when they see it applying to all elements (and
underlining massive blocks of text, e.g.) they blame us.

Comment 9

[lhunath]

You've a point there, seen that happen in Opera (which does allow for :hover in 
Quirks).
Sad how society is always ruled by the most stupid amoungst it =)