- From: Boris Zbarsky <bzbarsky@MIT.EDU>
- Date: Fri, 08 Jul 2011 15:42:23 -0400
- To: www-style@w3.org
On 7/8/11 8:28 AM, Erik Dahlstrom wrote: > On Thu, 07 Jul 2011 17:25:06 +0200, Boris Zbarsky <bzbarsky@mit.edu> wrote: > >> On 7/7/11 8:28 AM, Erik Dahlstrom wrote: >>> There are at least three different implementations (webkit, gecko and >>> presto) that support the CSS properties added by SVG 1.1 on the >>> CSSStyleDeclaration interface[1] >> >> Though not interoperably (c.f. "filter").... >> >> -Boris > > True, that property has been implemented un-prefixed in Internet > Explorer in a different way than what SVG specifies. Is the non-svg > syntax in 'filter' still supported in IE10 without the '-ms-' prefix? I have no idea, but I wasn't talking about IE (which you didn't even mention), and I wasn't talking about syntax. For example, this testcase: <!DOCTYPE html> <pre><script> document.writeln(typeof(document.documentElement.style.filter)) </script></pre> writes out "string" in Gecko and Presto but "undefined" in WebKit. > For all the other properties I'd be surprised if there were any issues > like that though. And I'd rather see those properties included in the > draft than having them all left out because of 'filter'. That's fine, and I have no objection to adding the properties (heck, we added them in Gecko!). Just saying that there are some landmines here. > The elm.style.filter detection is currently being used on some sites to determine that the > browser used is Internet Explorer (not a very good way of detecting that > IMHO, but there we have it). Note that we have been convincing such sites to fix their browser-detection, somewhat. So maybe the problem will go away. -Boris
Received on Friday, 8 July 2011 19:42:54 UTC