- From: fantasai via GitHub <sysbot+gh@w3.org>
- Date: Tue, 04 Dec 2018 23:01:15 +0000
- To: public-css-archive@w3.org
> I am also interested in which ligatures are "optional". >From CSS3 Text's perspective, that would be any ligatures which are not required. Per @litherum's https://github.com/w3c/csswg-drafts/issues/2644#issuecomment-422075536 that means anything except `rlig` in OpenType. And maybe there needs to be some special handling for Khmer because of its broken relationship with OpenType ligature tags. It's technically out-of-scope for CSS to define what OpenType considers required, though. I tried to add some clarifications on this point in https://github.com/w3c/csswg-drafts/commit/caf1e4747f2bd906bb9f671af8c463b28b97deda though. Wrt Fraktur, it seems that using Fraktur changes the definition of “typographic grapheme cluster”, and pages tagged with `de-Latf` should have their segmentation adjusted accordingly. There's hundreds of examples of this type of thing, and it might be nice to have a registry where we collect info on “what's a typographic grapheme cluster” in all the various writing systems and variants thereof, but that's definitely too detailed for the CSS Specification. However the i18nWG is collecting such info in its *REQ reports, and it would be a good idea to have it noted in LatinReq and add corresponding WPT tests. https://www.w3.org/International/layout I can add this one as an example of how segmentation can vary by lang tag, though, since I think I don't have such an example yet in the spec. Wrt updating the font feature precedence rules to be clearer about the various types of ligatures, your wording changes for rule 4 seem clearly correct to me; I'm not convinced about the others though. Deferring to @litherum and @svgeesus on that. -- GitHub Notification of comment by fantasai Please view or discuss this issue at https://github.com/w3c/csswg-drafts/issues/2644#issuecomment-444292691 using your GitHub account
Received on Tuesday, 4 December 2018 23:01:18 UTC