120834 - Implement :checked selector

Status: VERIFIED FIXED

(Core :: CSS Parsing and Computation, defect, P3)

Description

[John Keiser (jkeiser)]

The :checked selector is a CSS3 selector that is on when a checkbox is checked.

Comment 1

[Hixie (not reading bugmail)]

This is part of the fix for bug 115462.

Comment 2

[John Keiser (jkeiser)]

Note to self: this selector /depends/ on bug 115462 as much as it is part of it.
 We gain two things from that bug:

- a per-node repository for this kind of state
- the ability to (hopefully efficiently) notify the stylesheet of changes in a
per-node selector and recalculate style.  hyatt tells me that the way I'm trying
to do to implement this (with ContentStylesChanged) is a bad idea.

This selector is necessary for XBL checkboxes not to suck though; currently
bryner are going to have to implement a Moz-specific attribute that he can style
off of.  Still, I am not sure if the CSS changes are going to get done or land
before 1.0.

Comment 3

[John Keiser (jkeiser)]

bryner is implementing this now.  Yaaaay :)

Comment 4

[Hixie (not reading bugmail)]

(note: unless the fix for this uses the technique described in bug 115462, I'd
rather this bug not be marked FIXED when the patch is checked in.)

Comment 5

[John Keiser (jkeiser)]

I don't see a need to muddy the exit criteria for this bug.  When :checked is
working, this bug is fixed, by definition--it doesn't matter how performant the
implementation is.  If it does, we can change this bug to "make :checked more
extensible and performant" I suppose, or open a new bug, but I think bug 115462
adequately covers that.

I do think bug 115462 is a very good idea, and will help us out a lot--just it's
not covered by this bug.

Comment 6

[Hixie (not reading bugmail)]

fair enough

Comment 7

[Brian Ryner (not reading)]

:checked is now working for the elements explicitly mentioned in the CSS3 spec
(namely selects, checkboxes, and radio buttons).  Please file any additional
RFE's as separate bugs (not on me, please file on the default style system
assignee).

Comment 8

[Anne (:annevk)]

Attachment: test case

This test case shows it it not dynamically updated. Problably there is another
bug for that, or nobody has noticed this yet, but shouln't bug be reopened
then?

Comment 9

[Anne (:annevk)]

-> VERIFIED

Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.7a) Gecko/20040114

Comment 10

[Boris Zbarsky [:bzbarsky]]

Attachment: Regression tests

Comment 11

[Boris Zbarsky [:bzbarsky]]

Checked in the tests.