W3C home > Mailing lists > Public > public-css-archive@w3.org > February 2018

Re: [csswg-drafts] [css-fonts-4] Deduplication of computed style & inheritance for duplicate axis tags

From: jfkthame via GitHub <sysbot+gh@w3.org>
Date: Thu, 15 Feb 2018 10:18:25 +0000
To: public-css-archive@w3.org
Message-ID: <issue_comment.created-365883351-1518689904-sysbot+gh@w3.org>
Currently, I believe Gecko passes the specified list of axis-value settings through as-is, and the "last value wins" behavior occurs at rendering time. However, I agree with @upsuper that there's a good case for earlier simplification.

(Note, though, that early simplification of `font-variation-settings` -- e.g. removing duplicates at parsing time -- does not eliminate the need for further processing of the variation list at rendering time, because of how variations specified in a `@font-face` descriptor combine with those specified in an element's style.)

BTW, according to [Fonts 3](https://drafts.csswg.org/css-fonts-3/#propdef-font-feature-settings),

> The computed value of font-feature-settings is a map, so any duplicates in the specified value must not be preserved.

and IMO we should have the equivalent behavior for `font-variation-settings` and `font-feature-settings`; it will be unnecessarily confusing for authors if these two properties behave differently with respect to inheritance, computation, etc.

(I don't have Edge on hand just now to check, but a brief test indicates that neither Gecko nor Blink currently follow the spec here for `font-feature-settings`; duplicates are preserved in the computed value. However, I think we should consider that a bug, and get it fixed accordingly -- for both features and variations.)


-- 
GitHub Notification of comment by jfkthame
Please view or discuss this issue at https://github.com/w3c/csswg-drafts/issues/2288#issuecomment-365883351 using your GitHub account
Received on Thursday, 15 February 2018 10:18:27 UTC

This archive was generated by hypermail 2.4.0 : Tuesday, 5 July 2022 06:41:24 UTC