Re: RE : Re: [css3-exclusions] CSS Floats & reusability of shapes

Hi Francois,

oh, I see. I do not have a strong opinion on this, but would just note that I think this is orthogonal to the exclusions spec., even though, as you illustrated, it would be helpful in that context.


From: François REMY <<>>
Date: Mon, 25 Jul 2011 14:52:35 -0700
To: Adobe Systems <<>>, CSS 3 W3C Group <<>>
Subject: RE : Re: [css3-exclusions] CSS Floats & reusability of shapes

Thank you for your answer. While the draft don't forbid the use of pseudo elements, the fact you can only define two of them per element does.

My proposal is to increase pseudo elements use cases coverage by allowing any arbitrary number of them, without adding further complexity to their implementation.

De : Vincent Hardy
Envoyé le : lundi 25 juillet 2011 23:38
À : François REMY; CSS 3 W3C Group
Objet : Re: [css3-exclusions] CSS Floats & reusability of shapes

Bonjour François,

While exclusions as pseudo-elements are not showing in the drafts, there is nothing that should prevent that since you could have the wrap-mode property on the pseudo-elements. So I think your example should work,


From: François REMY <<>>
Date: Mon, 25 Jul 2011 12:55:14 -0700
To: CSS 3 W3C Group <<>>
Subject: [css3-exclusions] CSS Floats & reusability of shapes

The current brain storming about "Exclusions" regions ( and the specific statement about the "reusability of shapes" in the CSS4 Floats proposal lead me to think about one of the features that somehow disappeared from any proposal but that would solve that issue perfectly : multiple pseudo-elements.

I mean, you could easily replace this markup [[

    .exclusion-region-1 { .... }
    .exclusion-region-2 { .... }

        <div class="exculsion-region-1" />
        <div class="exculsion-region-2" />

]] with this one [[

    article::before[1] { .... }
    article::before[2] { .... }


]] (in fact, in this particular case, you could have used both ::after and ::before because there's only two exclusion shapes to define)

I don’t know if there are specific UA requirements that made this impossible and explain why this idea disappeard from all the current proposals, but I wanted to bring my 50 cents to the discussion ;-)

Best regards,

PS: Maybe a ::before[new] would be useful to ask to create a new ::before pseudo-element that can’t overlap with any other CSS ::before element defined by another rule. At this time, for exemple, you can have two rules thinking they style “their” ::before pseudo-element but in fact they may be competing on the same pseudo-element. It then gives a very different result than expected. The insertion order of those “[new]” pseudo elements in the document would be in function of the predecence of style rules (please note they are always inserted after any element whose position is specified explicitly).

Received on Monday, 25 July 2011 21:58:10 UTC