Hi Dominik,
I recognized the conflict between the two adjusting CSS properties in Blink
from this mail thread. :P
I don’t have a concrete plan to address it yet. If I, however, drew a rough
solution, it would look like the following:
1. Handle the two CSS properties separately with two distinct placeholders
in FontDescription.
2. When parsing, we assign values corresponding to their placeholders.
3. When resolving styles (e.g., font builder, style builder),
FontDescription::SizeAdjust() returns a font-size-adjust value first if it is
available; otherwise returns the font-face: size-adjust.
This rough sketch can change later. Please let me know if you have a different
approach in mind.
Thanks,
--
ChangSeok
> On Apr 4, 2023, at 3:59 AM, Dominik Röttsches <[email protected]> wrote:
>
> Hi ChangSeok,
>
> Thanks for confirming the expected behavior and testing in FF.
>
> I plan to implement the same behavior of Firefox, following the spec document.
>
> Can you share a few more details how you plan to achieve that? Have you been
> able to verify that you can achieve this behavior in a local development
> branch? If so, what changes are needed to the interaction between the two
> features in Blink?
>
> Thanks,
>
> Dominik
>
>
> On Fri, Mar 31, 2023 at 12:56 AM ChangSeok Oh <[email protected]> wrote:
> Hi Dominik,
>
> Thanks for informing me about the history and conflict between @font-face:
> size-adjust and font-size-adjust.
>
> Per your comment, I investigated the latest spec document and how Gecko
> handles the conflict.
> In a nutshell, the draft spec prioritizes the font-size-adjust over the
> size-adjust (i.e., overriding) [1], and Gecko follows the rule.
>
> The following is the related part of the document.
>
> > The font-size-adjust property is applied after the size-adjust descriptor.
> > NOTE: The consequence of applying font-size-adjust after size-adjust is
> > that size-adjust appears to have no effect.
>
> In addition, I wrote a simple test on jsfiddle.net [2]. Please feel free to
> test it with your Firefox.
>
> I plan to implement the same behavior of Firefox, following the spec document.
>
> Best,
>
> [1]
> https://w3c.github.io/csswg-drafts/css-fonts-5/#valdef-ascent-overridedescriptor-percentage
> [2] https://jsfiddle.net/wfsdagc7/
>
> --
> ChangSeok
>
> > On Mar 30, 2023, at 12:13 AM, Dominik Röttsches <[email protected]> wrote:
> >
> > Hi ChangSeok,
> >
> > Thanks for working on this. During the development and shipping of the
> > @font-face size-adjust descriptor we found that at least in our code the
> > font-size-adjust implementation was not compatible with this change. As
> > size-adjust was a higher priority then, the font-size-adjust
> > RuntimeEnabledFeature was downgraded from experimental to test, discussion
> > during review in this change.
> >
> > We have a related TODO in the code that when the size-adjust descriptor is
> > present in a @font-face, the font-size-adjust property is ignored. Could
> > you describe how you plan to address this? Could you also investigate how
> > FireFox handles cases where the size-adjust descriptor and font-size-adjust
> > interact. Are there any problems arising from the use of both at the same
> > time?
> >
> > Dominik
> >
> > On Tue, Mar 28, 2023 at 9:35 PM ChangSeok Oh <[email protected]> wrote:
> >
> >
> > > On Mar 28, 2023, at 6:29 AM, David Baron <[email protected]> wrote:
> > >
> > > On Mon, Mar 27, 2023 at 6:33 PM ChangSeok Oh <[email protected]> wrote:
> > > Interoperability and Compatibility
> > >
> > > Gecko: Publicly support
> > >
> > > WebKit: Positive and in development
> > > https://bugs.webkit.org/show_bug.cgi?id=254191
> > > https://github.com/WebKit/WebKit/pull/11744 (I am the author)
> > >
> > > Web developers: No signals
> > >
> > > Other signals:
> > >
> > > WebView application risks
> > > No known risks
> > >
> > > It might be useful to separately describe (for some or all of these
> > > sources) support for the 1-value syntax (CSS 2.1, css-fonts-3,
> > > css-fonts-4) and support for the 2-value syntax (css-fonts-5). For
> > > example, according to MDN, Gecko has been shipping the 1-value syntax
> > > since Firefox 3 (June 2008) and the 2-value syntax since Firefox 92
> > > (September 2021).
> >
> > We already have a separate feature flag and ChromeStatus page [1].
> > Unfortunately, it has been pending for years. I plan to ship it with this
> > two-value syntax once this new feature is done. But I don’t have a strong
> > opinion on the shipping plan. We can consider shipping separately.
> >
> > [1] https://chromestatus.com/feature/5720910061371392
> >
> > >
> > > I'll also add that I'm a big fan of this feature and am happy to see it
> > > implemented.
> >
> > Thanks!
> >
> > >
> > > -David
> >
> > --
> > You received this message because you are subscribed to the Google Groups
> > "blink-dev" group.
> > To unsubscribe from this group and stop receiving emails from it, send an
> > email to [email protected].
> > To view this discussion on the web visit
> > https://groups.google.com/a/chromium.org/d/msgid/blink-dev/0E07FF35-2FC4-43BA-A55E-F269997FDD0D%40gmail.com.
>
--
You received this message because you are subscribed to the Google Groups
"blink-dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to [email protected].
To view this discussion on the web visit
https://groups.google.com/a/chromium.org/d/msgid/blink-dev/81521824-0EED-47DA-B55B-2FA39EC75D07%40gmail.com.
