Re: [csswg-drafts] [css-sizing] cyclic percentage concept should not exist

If I understand the spec correctly, I think the spec already has applied the "intrinsication" to all the percentage-sized values of all descendant boxes in the "very first" step of an intrinsic size calculation, no matter how deep a descendant box is, that is, it's not possible to be non-cyclic percentage for an intrinsic size, because _all_ of descendant percentage-sized values are turned to cyclic for an intrinsic size calculation per spec, no matter how deep a descendant box is.

I didn't immediately realise that to begin with, because it's not obviously stated in the spec.

I happened to be aggressively learning and reading through css-size-3 these two days, and once had a hard time conclusing a simplified evocative "outline" in my brain, as to in which order various complicated situations that have cylic dependency would be resolved, and how would they work. After trying to figure out the haze for some time, I was finally enlightened and inferred that when caculating an intrinsic size calculation for a box, _all_ of its descendant boxes' percentage-sized values are converted to either auto or 0, no matter how deep the descendant box is. I think I have a clear understanding of the overall machenism now, but it's not easy for me to evocatively express it to others with just a few words.

Since the "intrinsicate" is applied to all the percentage-sized values of all descendant boxes in the "very first" step of an intrinsic size calculation. IMHO, the spec might have more explicit statements instead of the concept of cyclic percentage, so people don't have to infer to get such a conclusion.

-- 
GitHub Notification of comment by Zhang-Junzhi
Please view or discuss this issue at https://github.com/w3c/csswg-drafts/issues/3010#issuecomment-412323879 using your GitHub account

Received on Sunday, 12 August 2018 07:18:12 UTC