- From: Boris Zbarsky <bzbarsky@MIT.EDU>
- Date: Sat, 02 Oct 2004 22:53:07 -0500
- To: Andrew Fedoniouk <news@terrainformatica.com>
- CC: www style <www-style@w3.org>
Andrew Fedoniouk wrote:
> Sorry, Boris I was not clear enough.
>
> Example:
> Let's say we have standalone paragraph with style
>
> <body>
> <p style="display:table-cell">
> </body>
>
> As per CSS [1] "display:table-cell" forces change of underlying DOM - UA
> *must* create table and row elements.
No, that's not true.
The UA must create table and table-row _boxes_. These are purely display
objects and are not attached to any DOM nodes.
> Example: for the HTML above shall following style selector be in effect for
> the paragraph?
>
> body > p { color:red }
Absolutely, since the DOM is completely unaltered. The parent of the <p> node
is still the <body>. There are four boxes around (block, table, table-row,
table-cell), of which only two are associated with a DOM node.
> (p does not have body as a direct parent anymore, right?)
Sure it does.
> body > p { display:block; }
> body > p:hover { display:table-cell; }
That's fine.
> Shall UA clear remnants of anonymous cell application (table and row
> elements) when mouse will left such paragraph?
Yes. (Mozilla gets this wrong currently, but it's on the "to fix" list.)
> And again: after switch into 'hover' state first rule will not be valid for
> the P anymore.
That's not true. Again, the DOM is not changed.
-Boris
Received on Sunday, 3 October 2004 03:53:13 UTC