W3C home > Mailing lists > Public > www-style@w3.org > November 2011

Re: [css3-lists] Should some properties not inherit into markers?

From: Peter Moulder <peter.moulder@monash.edu>
Date: Thu, 24 Nov 2011 11:38:06 +1100
To: www-style@w3.org
Message-id: <20111124003806.GA21738@bowman.infotech.monash.edu.au>
On Wed, Nov 23, 2011 at 09:44:19AM -0800, Tab Atkins Jr. wrote:

> So, I propose adding "::marker { text-transform: initial; }" to the
> Lists 3 UA stylesheet.  Thoughts?

That approach won't have any effect on 'content' use.  A use case where
this might matter is for page headings, when one might grab a section
heading and its number for an <h2> element and uppercase it for purposes
of the page heading (while the h2 use displays it in mixed case).

The question remaining for this use case is how common it would be to
want the section number to be left untransformed: perhaps in the

If we did want to allow for this case, then one way to accomplish it
would be as follows.  There are now already one or two cases in CSS
where content items need to be preserved as abstract expressions rather
than filling in immediately: examples are page number references
(target-counter) and leaders.  Conveniently, I believe another example
(if we follow the principle that computed values shouldn't depend on
network operations[*1], and if I correctly understand that "If the
computed value of 'list-style-image' is a valid image" requires a network
operation to determine) is the computed value of 'content' for ::marker.

We could then add counter() and counters() to this set, and specify that
text-transform not affect the text produced by any of the members of this
set.  In terms of implementation, this would correspond to delaying the
evaluation of these expressions until after text-transform had been
applied.  (This also means that text-transform can be applied before the
network operation finishes, without having to remember to re-apply
text-transform once the network operation has finished and we've
determined that the image isn't in fact valid.)

This is actually looking like it might be easier to implement than not
implement, so I'll also pose the opposite question: how often are there
cases where we *would* prefer text-transform to apply to counters?


[*1]: See http://lists.w3.org/Archives/Public/www-style/2011Oct/0840.html
Received on Thursday, 24 November 2011 00:38:34 UTC

This archive was generated by hypermail 2.3.1 : Monday, 2 May 2016 14:38:52 UTC