W3C home > Mailing lists > Public > www-style@w3.org > August 2013

Re: [css3-multicol] Nested multicolumn elements rendering

From: Morten Stenshorne <mstensho@opera.com>
Date: Wed, 14 Aug 2013 11:36:20 +0200
To: Håkon Wium Lie <howcome@opera.com>
Cc: www-style@gtalbot.org, "www-style mailing list" <www-style@w3.org>
Message-ID: <87txisaa23.fsf@aeneas.oslo.osa>
Håkon Wium Lie <howcome@opera.com> writes:

> Morten Stenshorne wrote:
>  > > I've asked Hakom a similar question on July 21st and he answered:
>  > >
>  > > {
>  > > If height [of a multi-column element] is auto, [then column-fill:] 'auto'
>  > > and [column-fill:] 'balance' should produce the same result.
>  > 
>  > That's only for continuous media, I think. The inner multicol is kind of
>  > in paged media, and my main point is that column-fill:auto is meaningful
>  > here. column-fill:auto should only be dishonored if the column heights
>  > are truly unconstrained (and even then it can be argued that it's a
>  > dubious thing to do; you could still achieve interesting effects with
>  > only explicit column breaks).
> Yes, it would make sense to honor 'column-fill: auto' even in
> unconstrained environments. One can achieve interesting effects, and
> one honors specified values (instead of ignoring them).
> If so, we would have, I believe:
>                                          auto                  balance
> continous media, constrained columns     fill                  balance        
> continous media, unconstrained columns   honor exlicit breaks  balance        
> continous media, overflow columns        -- property has no effect --
> paged media, all-but-last-pages          balance               balance
> paged media, last page                   fill                  balance
> Does this look right?

Why force-balance in paged media?

Anyway, I don't think we should change the behavior of
"column-fill:auto; height:auto;", since the current version of the spec
is rather detailed on how to behave in this case (except that "if the
length of columns has been constrained" [1] perhaps is a bit
vague). Most engines (WebKit/Blink doesn't count, since it doesn't
implement column-fill yet) have implemented this correctly already. If
we want to make changes in this area, I think it is better to introduce
a new value to 'column-fill' (it could be called 'fill', meaning "never

[1] http://www.w3.org/TR/2011/CR-css3-multicol-20110412/#cf

---- Morten Stenshorne, developer, Opera Software ASA ----
------------------ http://www.opera.com/ -----------------
Received on Wednesday, 14 August 2013 09:36:36 UTC

This archive was generated by hypermail 2.4.0 : Friday, 25 March 2022 10:08:33 UTC