W3C home > Mailing lists > Public > www-style@w3.org > October 2011

Re: [css-regions] The region-overflow property

From: Alan Stearns <stearns@adobe.com>
Date: Tue, 4 Oct 2011 19:03:25 -0700
To: David Hyatt <hyatt@apple.com>
CC: "robert@ocallahan.org" <robert@ocallahan.org>, "www-style@w3.org list" <www-style@w3.org>
Message-ID: <CAB108FD.801C%stearns@adobe.com>
On 10/4/11 6:04 PM, "David Hyatt" <hyatt@apple.com> wrote:

I guess I'm wondering why we wouldn't just build the behavior into the overflow values and eliminate the property then.

(1) overflow:hidden on last region = Honor breaks as though the content spilled out of the region in the pagination direction but the content gets clipped.
(2) overflow:visible/scroll/auto on last region = Don't honor breaks and just let all the content overflow and either be unclipped (visible) or reachable via a scrolling mechanism (scroll/auto).

The only reason to have region-overflow is if we think there is some value in supporting overflow:hidden, region-overflow:auto. I don't really see the point of this, because as you mentioned, it just looks like a rendering mistake.

So let's just kill the property completely and specify that overflow:hidden has the behavior you want.

dave
(hyatt@apple.com)

That seems reasonable to me, but I don’t know enough about overflow:hidden to determine whether it would be useful to have that behavior for regions. Using overflow:hidden on other elements looks like a rendering mistake to me too, but I assume there’s a reason it exists.

So I think we have three (or four) options:

1. Special case overflow:hidden for regions to clip at a pagination break. In this option we lose the normal ‘hidden’ behavior for regions.

2. Add a ‘break’ value for the overflow property to clip at a break. In this option we preserve the normal ‘hidden’ behavior for regions, but have to define what ‘break’ means for non-region elements. It could be defined as (a) the same as ‘hidden’ for non-region elements, or (b) we could add the clip-at-break behavior to everything (which might be useful).

3. Keep the region-overflow property as it is in the spec now. This allows both behaviors for regions, and lets us avoid defining what a ‘break’ means for non-region elements.

My preference is 2b, but I could accept any of these.

Alan
Received on Wednesday, 5 October 2011 02:03:57 GMT

This archive was generated by hypermail 2.3.1 : Tuesday, 26 March 2013 17:20:45 GMT