CSS 'cursor' property on root element should affect cursor for entire viewport

NEW
Unassigned

Status

()

defect
9 years ago
9 years ago

People

(Reporter: bugzilla33, Unassigned)

Tracking

Firefox Tracking Flags

(Not tracked)

Details

()

Attachments

(1 attachment, 1 obsolete attachment)

712 bytes, text/html
Details
User-Agent:       Mozilla/4.0 (compatible; MSIE 8.0; Windows NT 6.1; Trident/4.0; SLCC2; .NET CLR 2.0.50727; .NET CLR 3.5.30729; .NET CLR 3.0.30729; Media Center PC 6.0; InfoPath.2; .NET4.0C; .NET4.0E; FDM)
Build Identifier: Mozilla/5.0 (Windows; U; Windows NT 6.1; en-US; rv:1.9.2.4) Gecko/20100513 Firefox/3.6.4

only HTMLBodyElement has cursor:wait but HTMLHtmlElement should have

Reproducible: Always

Steps to Reproduce:

open attachment

Actual Results:  

only HTMLBodyElement has cursor:wait


Expected Results:  

HTMLHtmlElement has cursor:wait



open attachment
Posted file testcase (obsolete) —
<style type="text/css">
 html{cursor:wait}
</style>

---

only HTMLBodyElement has cursor:wait
Uh...  The cursor is wait in that testcase while you're hovering over the <html> element.  You can add a border on <html> to see where its boundaries are.

I have no idea why other browsers are doing something different here; what we're doing seems correct per spec.
But I have an idea. Other browsers distinguish html from the body element.
HTML fills the whole window vertically. Body no.

Try new testcase: http://pc44.one.pl/goorol/bugs/firefox/cursor.html

  <style type="text/css">
   html{background:#bbbbbb;cursor:wait}
   body{background:#ffff00}
  </style>
Comment on attachment 447728 [details]
testcase

see new testcase
Attachment #447728 - Attachment is obsolete: true
Posted file new testcase
Summary: cursor problem → cursor on HTML element problem
>> You can add a border on <html> to see where its boundaries
are.


A add grey background to see that HTML is whole window. Where is cursor:wait on grey area ??
> A add grey background to see that HTML is whole window.

Background on the root element is special-cased in CSS to apply to the viewport instead of the element itself.  This is why I told you to use border to see what the the actual size of the <html> is.
OK.

Imagine that cursor on root element SHOULD be also special-cased in CSS to apply to the viewport instead of the element itself.
It is normal behaviout - we want some cursor over background filled element.

WebKit, Trident, Presto and KHTML... As many as four independent engines can not go wrong at once. The Gecko takes bad foundation.
> Imagine that cursor on root element SHOULD be also special-cased in CSS

Then the CSS spec needs to say that.  Right now it doesn't.  I suggest mailing www-style@w3.org about that if you think the spec needs to change.

> As many as four independent engines can not go wrong at once

Sure they can.  It's happened before (with Gecko in that list of 4 at times).
Imagine a situation where a user makes an artistic side. He wants a beautiful decorative basic cursor. Unfortunately, not in firefox because the site is lower than the browser window.
I understand the use case.  I'm just saying that the spec doesn't allow that behavior at the moment.
(In reply to comment #12)
> Unfortunately, not in firefox because the site is lower than 

Workaround:     html { height:100% }
min-height, not height.
I think a special case for the root element would make sense.  Why don't you propose that on www-style?  (If you provide a testcase showing that the proposed change already matches the behavior of 4 browsers, it'll probably be received positively.)

However, you should probably also test if browsers special-case 'cursor' on body when it's not specified on the root.

http://lists.w3.org/Archives/Public/www-style/
Summary: cursor on HTML element problem → CSS 'cursor' property on root element should affect cursor for entire viewport
I'd like to include this detail in CSS3-UI as well, here:

http://dev.w3.org/csswg/css3-ui/#cursor
Status of this bug should change to NEW then?
OS: Windows 7 → All
Hardware: x86 → All
Status: UNCONFIRMED → NEW
Ever confirmed: true
You need to log in before you can comment on or make changes to this bug.