[patch] inlines

[Germain Garand]

Le Lundi 16 Février 2004 12:14, Dirk Mueller a écrit :
> On Monday 16 February 2004 12:36, Germain Garand wrote:
> > for 008.html, I think the testcase is actually invalid because it doesn't
> > set the display to block explicitly.
>
> CSS 2.1 12.1:
>
> User agents must ignore the following properties with :before and :after
> pseudo-elements: 'position', 'float', list properties, and table
> properties.

yes I know, that's why I kept that bit in my original patch:

    // From the CSS2 specification:
    // User agents must ignore the following properties with :before and 
:after
    // pseudo-elements: 'position', 'float', list properties, and table 
properties.
    // Basically we need to ensure that no RenderLayer gets made for generated
    // content.
    pseudo->setPosition(STATIC);
    pseudo->setFloating(FNONE);
    pseudo->setOverflow(OVISIBLE); // FIXME: Glazman's blog does this. Wacky.

it was removed in WebCore-125, probably because 
CSS2.1's current draft doesn't mention this limitation anymore, AFAICS.

Now, if it's wise to align on that or not, I don't now.

>
> > It worked with your patch because up to know, we wrongly set the initial
> > display of :before/:after to block, which caused quite a few bugs.
>
> CSS 2.1 12.1:
>
> If the subject of the selector is a block-level element, allowed values are
> 'none', 'inline' and 'block'. If the value of the pseudo-element's
> 'display' property has any other value, the pseudo-element will behave as
> if its value were 'block'.

_allowed_ values are  'none', 'inline' and 'block', but the initial value is 
inline nevertheless:

"So, for example, because the initial value of the 'display' property is 
'inline', the quote in the previous example is inserted as an inline box 
(i.e., on the same line as the element's initial text content). The next 
example explicitly sets the 'display' property to 'block', so that the 
inserted text becomes a block:"