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
A rather unique case came up today, where I had to nest a font-face directive inside a library mixin, which in turn is mixed into a basic class selector.
My real-world case involves scaffolding a quite complex grid layout where I have to include a font with a zero horizontal advance for white-space characters, which effectively makes the inline-block grid column elements line up snugly together.
I'd expect directives that have block context but cannot have nested rulesets, to not propagate scope selectors inside when the directive is bubbled outwards to the global scope.
I think I have an idea on how I can get the compiler to support this and I can look at authoring a pull request for it, if you agree that this is indeed something to fix.
The text was updated successfully, but these errors were encountered:
rjgotten
changed the title
Directives and selector propagation
Nested directives bubbling and selector propagation
Mar 19, 2015
It looks like it's a regression introduced with #2473. Nested @keyframes, @fontface and similar directives are semi-broken by it (they behave like @supports and @document now while they should not).
A rather unique case came up today, where I had to nest a
font-face
directive inside a library mixin, which in turn is mixed into a basic class selector.My real-world case involves scaffolding a quite complex grid layout where I have to include a font with a zero horizontal advance for white-space characters, which effectively makes the inline-block grid column elements line up snugly together.
A reduced test case of the structure is:
Currently, this test case produces the output:
I'd expect directives that have block context but cannot have nested rulesets, to not propagate scope selectors inside when the directive is bubbled outwards to the global scope.
I.e. I'd expect the above to produce:
I think I have an idea on how I can get the compiler to support this and I can look at authoring a pull request for it, if you agree that this is indeed something to fix.
The text was updated successfully, but these errors were encountered: