W3C home > Mailing lists > Public > www-style@w3.org > March 2015

Re: [css-fonts] What does font-size-adjust for '0' look like?

From: John Daggett <jdaggett@mozilla.com>
Date: Thu, 19 Mar 2015 11:43:13 +0900
Message-ID: <CALYZoVNH4y_+Z63v1Rx_6P7Yq0RmDnY_tk8M87F+OGcn9JGrag@mail.gmail.com>
To: Gérard Talbot <www-style@gtalbot.org>
Cc: ChangSeok Oh <changseok@gnome.org>, W3C www-style mailing list <www-style@w3.org>
Gérard Talbot wrote:

> > I have a simple question on font-size-adjust. [1]
> > According to the spec. font-size-adjust can have ‘none’ or <number>
> > for its value.
> > BTW what is expected to look like where font-size-adjust is ‘0’? I
> > checked Gecko(which is the only vendor supporting for the property
> > now) treated it as ‘none’.
> > Does it make sense? As my reading, font-size-adjust represents an
> > aspect value (x-height / font-size) of a font.
> > So ‘0’ might mean x-height is 0 or equivalent very small value here.
> > Thus It should be same effect with 'font-size = 0px’
>
> Yes, it should be same effect with 'font-size = 0px’. I agree with your
> reasoning. The spec editor, John Daggett, should probably be notified of
> this. 0 would have to mean a 0px font-size.

The existing wording already covers the font-size-adjust: 0 case. The Gecko
implementation does not handle this case correctly.

> Negative values are illegal and ignored.

This is the only thing that I think requires additional spec wording:

  Negative values have the same effect as specifying 'none'.

That is to say, they are parsed and computed but will not affect the
used font size.

> Percentage values are ignored.
> 'auto' value must be ignored.

Both of these are invalid given the existing definition. They should not
parse if used.

Cheers,

John Daggett
Mozilla Japan

​
Received on Thursday, 19 March 2015 02:43:41 UTC

This archive was generated by hypermail 2.3.1 : Monday, 2 May 2016 14:39:30 UTC