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
{{ message }}
Repository navigation
[css-ruby][css-display] Are ruby containers atomic? Specs contradict #1180
Yes, the ruby formatting context appears to be an exception from the general rule that the inline-level box that establishes a new formatting context is usually atomic. Ruby containers can break between lines, so they are definitely non-atomic.
I suppose that the CSS Display spec should be changed to something like
An inline-level box that is replaced or that establishes a new block, flexbox, grid, or table formatting context
An inline-level element can't establish the inline formatting context, and establishing ruby formatting context doesn't make an inline-level box atomic. If new formatting contexts that make inline-level boxes atomic appear in the future, they should be explicitly added to that list.
Ruby containers are weird... they establish a new ruby formatting context, but their content also participates in their parent's inline formatting context. That's going to be odd to explain. :) I'd rather call out ruby as an exception than list out the things that aren't exceptional though.
Activity
SelenIT commentedon Apr 7, 2017
Yes, the ruby formatting context appears to be an exception from the general rule that the inline-level box that establishes a new formatting context is usually atomic. Ruby containers can break between lines, so they are definitely non-atomic.
I suppose that the CSS Display spec should be changed to something like
An inline-level element can't establish the inline formatting context, and establishing ruby formatting context doesn't make an inline-level box atomic. If new formatting contexts that make inline-level boxes atomic appear in the future, they should be explicitly added to that list.
fantasai commentedon Apr 7, 2017
Ruby containers are weird... they establish a new ruby formatting context, but their content also participates in their parent's inline formatting context. That's going to be odd to explain. :) I'd rather call out ruby as an exception than list out the things that aren't exceptional though.
[css-display] Clarify ruby containers' interaction with atomic inlines.
fantasai commentedon Apr 7, 2017
Okay, this should be fixed now. Let me know if you have an further improvements to suggest on this point.