Re: [csswg-drafts] [css-multicol-2] `column-wrap`, `column-height` and column balancing (#11976)

![Image](https://github.com/user-attachments/assets/0372cc6d-0324-4f5d-83e8-0377211ee81e)

Excuse my terrible diagramming, it's slightly better than a photo of my notepad scrawl. 

So, a multicol only has one row if it's overflowing inline (the default behavior). If spanners are introduced, that creates multiple lines within that row. I tried to be careful with the terminology to avoid confusion with where row gaps and lines go.

If a multicol wraps, then we end up with multiple **rows**. Each row could have multiple lines, if a spanner is introduced.

I see a row as being the same as a page in most cases. So, what happens when we have a multicol nested inside paged media and spanners is I think what happens here (I think this is why in my mind this is a nested fragmentation flow, but maybe that's an incorrect description). I do think it makes things more straightforward if we assume block direction wrapping is generally equivalent a paged multicol. Though if we can find enough examples where that makes no sense, that's all good.

What also needs to be specced is the [behavior of `column-span: all`](https://www.w3.org/TR/css-multicol-2/#column-span) when we are wrapping. I'm assuming what we want is that `all` doesn't mean span all columns by wrapping with them, that seems unusual, but that `all` would instead be something like:

> The element forces a column break and is taken out of flow to span across all columns of the **nearest multicol row** in the same block formatting context. 

That said, this draft also includes some early work on `column-span: <integer>` so if we're intending to go to that place we should consider that too.

-- 
GitHub Notification of comment by rachelandrew
Please view or discuss this issue at https://github.com/w3c/csswg-drafts/issues/11976#issuecomment-2753981259 using your GitHub account


-- 
Sent via github-notify-ml as configured in https://github.com/w3c/github-notify-ml-config

Received on Wednesday, 26 March 2025 10:46:04 UTC