W3C home > Mailing lists > Public > public-tt@w3.org > July 2017

inlineAreaBreak cloneMostNested CSS issue and question

From: Nigel Megitt <nigel.megitt@bbc.co.uk>
Date: Fri, 21 Jul 2017 08:09:38 +0000
To: Timed Text Working Group <public-tt@w3.org>
Message-ID: <5941EAB8802D6745A7D363D7B37BD1F7813F6387@bgb01xud1012>
While analysing the TTML/CSS differences I spotted that there's an unsupported CSS feature and raised https://github.com/w3c/csswg-drafts/issues/1633

Elika returned an interesting question about whether cloneMostNested should clone the most nested style properties or the most nested ones that have something to clone. My instinct is just to go with most nested for now for simplicity, but my instinct may be wrong... Any other views?

Just to check my understanding, I think the result of having it mean "most nested" is that in the following example:

<p><span tts:backgroundColor="black" tts:padding="0em 0.5em" tts:inlineAreaBreak="cloneMostNested">Some words <span tts:textDecoration="underline">are underlined</span> in this sentence.</span></p>

if the break occurs within the inner span, say between "are" and "underlined" then the most nested span, which sets no tts:backgroundColor or tts:padding (and they are not inherited), is used to set the cloned values therefore there is no effective "linePadding". To get that effect, the tts:inlineAreaBreak should be set to "clone" on the outer span.

But here's a counter-example:

<p><span tts:backgroundColor="black" tts:padding="0em 0.5em" tts:inlineAreaBreak="cloneMostNested">Some words <span tts:backgroundColor="green">have a green background</span> in this sentence.</span></p>

In this case, at the line break we want the padding from the outer span to be cloned but the backgroundColor from the inner span to be cloned. What are the correct settings to achieve this?

I think I've just convinced myself that cloneMostNested should really be "clone the most nested thing that there is to clone"! It doesn't look like the current spec text for inlineAreaBreak says this though - do we need an issue on TTML2 about this?

Nigel
Received on Friday, 21 July 2017 08:10:02 UTC

This archive was generated by hypermail 2.3.1 : Thursday, 5 October 2017 18:24:41 UTC