- From: Chris Lilley <chris@w3.org>
- Date: Fri, 14 Oct 2011 00:37:19 +0200
- To: "Linss, Peter" <peter.linss@hp.com>
- CC: fantasai <fantasai.lists@inkedblade.net>, "public-css-testsuite@w3.org" <public-css-testsuite@w3.org>
On Thursday, October 13, 2011, 2:22:29 AM, Peter wrote: LP> On Oct 12, 2011, at 1:55 PM, fantasai wrote: >> Some of our tests are not XHTML, so running validation on them doesn't make >> much sense. Generally these are the ones in the 'other-formats' folder. >> They should be considered tests, but we can't automatically process them >> the way we do other tests. >> Ex: http://test.csswg.org/shepherd/testcase/background-18/ >> Not sure how to handle these, but the system shouldn't be marking them >> invalid for missing XHTML bits... LP> The system isn't trying to validate XML files as XHTML (it is LP> validating the XML though). The error it's reporting is that it's not finding any metadata. LP> Of course, for now, it's looking for the metadata inside an XHTML LP> <head> element, which theoretically could be inside a generic XML LP> file (just use the XHTML namespace for the <head> and its LP> children). The system also has hooks to look for metadata inside a LP> parallel XML file for those files that can't contain an XHTML LP> <head>, but that's not quite connected yet. I've been debating the LP> format for the parallel metadata files. It would be handy if the system also allowed for XML metadata inside the resource (for example in SVG, which does not have a head element but does have a metadata element). Since SVG2 tests will use this system, not being able to do this would be an issue as would having to put it in a parallel xml file. LP> We also need to extend the metadata to be able to define multiple LP> tests per file (for script tests like MQ). LP> Peter -- Chris Lilley Technical Director, Interaction Domain W3C Graphics Activity Lead, Fonts Activity Lead Co-Chair, W3C Hypertext CG Member, CSS, WebFonts, SVG Working Groups
Received on Thursday, 13 October 2011 22:37:23 UTC