You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
But math functions are accepted in any context, unless otherwise specified, and are resolved at computed value time or later. So what is the expected output below?
[]: disallow math functions in font feature value declarations (by defining them with <number-token> whose type should be integer, instead of <integer>)
[1]: resolve math functions at parse time in font feature value declarations (assuming relative length can be resolved, #10853)
calc(1em / 1em): deal with CSSOMStrings instead of unsigned longs
I have a preference for 3. Even if math functions might not be useful in this context, 1 is restrictive and 2 is static. Besides, with @view-transition/types, they are the only descriptors whose attribute getter do not return a CSSOMString.
The text was updated successfully, but these errors were encountered:
The value of a font feature value declaration is internally stored and exposed as unsigned integer(s) (spec).
But math functions are accepted in any context, unless otherwise specified, and are resolved at computed value time or later. So what is the expected output below?
I see 3 options:
[]
: disallow math functions in font feature value declarations (by defining them with<number-token>
whose type should beinteger
, instead of<integer>
)[1]
: resolve math functions at parse time in font feature value declarations (assuming relative length can be resolved, #10853)calc(1em / 1em)
: deal withCSSOMString
s instead ofunsigned long
sI have a preference for 3. Even if math functions might not be useful in this context, 1 is restrictive and 2 is static. Besides, with
@view-transition/types
, they are the only descriptors whose attribute getter do not return aCSSOMString
.The text was updated successfully, but these errors were encountered: