- From: Xidorn Quan <quanxunzhen@gmail.com>
- Date: Mon, 15 Dec 2014 08:37:08 +1100
- To: Koji Ishii <kojiishi@gmail.com>
- Cc: www-style list <www-style@w3.org>
- Message-ID: <CAMdq69_tzRNnsqYuFW7Wvo+eXz1HVdn0Z-=UfEEWXW2k5E+ivQ@mail.gmail.com>
On Mon, Dec 15, 2014 at 1:16 AM, Koji Ishii <kojiishi@gmail.com> wrote: > > > > On Tue, Dec 2, 2014 at 10:49 AM, Xidorn Quan <quanxunzhen@gmail.com> > wrote: > >> > >> Then line-height. As in the spec, we encourage authors to specify > >> sufficient line-height to put annotations, the line-height of <ruby> > should > >> not be "normal" in most cases. But line-height is a property which > inherits > >> by default, which means ruby annotations will inherit the large > line-height > >> from its parent (ruby container). This must not be what authors expect. > I > >> suggest that we break the inheritance for annotations in the default > style > >> sheet, hence: > >> > >> rt, rtc { line-height: normal; } > >> > >> or maybe 1em. I'm not sure about what value is better in practice, but > >> anyway it should not inherit. > > I can't imagine any use cases of using leadings for ruby text, so that > makes sense to me too. However, I'm not sure whether it should be done > through the default stylesheet or by code. The second paragraph of the > section 3. Ruby Layout[1] says: > > | Each ruby annotation container is sized and positioned to contain > | exactly the full height of its ruby annotations. > > so leadings should be ignored IIUC. I confirmed that WebKit ignores > line-height of rt, so if you want to achieve the same effect in the > default stylesheet than the code, probably: > > rt, rtc { line-height: 1 !important; } > > Not sure if there were any cases where having this in the default > stylesheet differs from what the spec says. > WebKit also defines "line-height: normal" in its UA style sheet. Defining this would make sense if there are inline-block inside the annotation. I'd prefer not make rtc ignore the leadings, but use only style to control. It would give authors more flexibility with reasonable default behavior. Also, I don't think we need "!important" here. - Xidorn
Received on Sunday, 14 December 2014 21:38:15 UTC