[Webkit-unassigned] [Bug 261293] New: Errors with oblique/slnt angle values for variable fonts
bugzilla-daemon at webkit.org
bugzilla-daemon at webkit.org
Thu Sep 7 14:35:12 PDT 2023
https://bugs.webkit.org/show_bug.cgi?id=261293
Bug ID: 261293
Summary: Errors with oblique/slnt angle values for variable
fonts
Product: WebKit
Version: Safari 16
Hardware: Mac (Apple Silicon)
OS: macOS 13
Status: NEW
Severity: Normal
Priority: P2
Component: Text
Assignee: webkit-unassigned at lists.webkit.org
Reporter: nick at nicksherman.com
CC: mmaxfield at apple.com
Created attachment 467590
--> https://bugs.webkit.org/attachment.cgi?id=467590&action=review
Oblique angle test with a variable font shown in Safari 16.6 on the left and Firefox 117.0 on the right
There are currently several issues related to the handling of oblique angles and the corresponding `slnt` axis in variable fonts.
1) When an oblique angle is specified via the `font-style` property, the text is slanted in the opposite direction it should be. This is likely the result of a 1:1 mapping between the angle in degrees and the resulting value of the `slnt` axis, instead of an inverted mapping. The `slnt` axis in variable fonts is set so positive numbers correspond to counter-clockwise slanting, and negative numbers to clockwise slanting, but the opposite is true for how oblique angles are set with degrees in CSS. In other words, `15deg` in CSS should correspond to a `slnt` value of -15.
There are a couple strange exceptions to this behavior that are wrong in other ways but I haven't figured out the logic (see the 45deg and -45deg lines in the sample page linked below and/or attached screenshots).
2) This may be related to the above, but setting a default `oblique` value (without an angle) results in no visible angle being applied. The default should be 14deg (i.e. a `slnt` value of -14).
3) It seems that if a `slnt` axis value is set directly via the `font-variation-settings` property, and it exceeds the supported range of the variable font, the value is ignored altogether. It should instead be clamped to the closest value the font supports.
Here is a test page on CodePen to see these issues:
https://codepen.io/NickSherman/pen/QWzdaZM
There is also a comparison screenshot attached here of Safari 16.6 vs Firefox 117.0.
References:
https://drafts.csswg.org/css-fonts-4/#font-style-prop
https://learn.microsoft.com/en-us/typography/opentype/spec/dvaraxistag_slnt
--
You are receiving this mail because:
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.webkit.org/pipermail/webkit-unassigned/attachments/20230907/2379f6c5/attachment-0001.htm>
More information about the webkit-unassigned
mailing list