- From: Sylvain Galineau <sylvaing@microsoft.com>
- Date: Thu, 19 Jan 2012 21:15:18 +0000
- To: Tab Atkins Jr. <jackalmage@gmail.com>, Boris Zbarsky <bzbarsky@mit.edu>
- CC: "www-style@w3.org" <www-style@w3.org>
[Tab Atkins Jr.:] > > On Tue, Jan 17, 2012 at 9:26 AM, Boris Zbarsky <bzbarsky@mit.edu> wrote: > > https://bugzilla.mozilla.org/attachment.cgi?id=589081 shows 210 in all > > modern browsers. Per the definition currently in the spec I _think_ > > it should say 220, because that definition talks about adding up the > > paddings and the "content width", but the content can (and in this > > case does) overlap the paddings. > > > > I believe in this instance the spec is what's wrong, not the > > implementations. > > I know I tested overflow situations when I was creating my explanatory > diagram <http://www.xanthir.com/diagrams/scrollwidth-clientwidth.html>, > but I guess I somehow missed the case of "overflow + padding-right". > > You're right - the definition doesn't match reality. > Fwiw I've always been surprised by the overflow:auto rendering in this case. With overflow:visible you'd get a 200px wide yellow box sticking out of a green 100px square. But as soon as overflow is auto (or scroll) the parent's green background is visible all the way to the end of the scrolling area i.e. the background area appears to be extended to the scrollable area. What's the rationale behind that difference?
Received on Thursday, 19 January 2012 21:16:04 UTC