- From: Manos Batsis <m.batsis@bsnet.gr>
- Date: Wed, 13 Mar 2002 13:39:04 +0200
- To: "Curt Arnold" <carnold@houston.rr.com>
- Cc: <www-dom-ts@w3.org>
True. The only way to use one could have used the same tests both for XML and HTML, would be to use HTML element and attribute names (and values) for the XML tests as well and have a stylesheet producing each version from the same source doc only to test the XML interfaces. What I'm doing now is actually quite the opposite. Unfortunately and against some of my original intentions, HTML UAs are supposed to ignore unknown elements and attributes or unknown attribute values ( at least as far as rendering goes) according to [1]. Existing tests of course *may* work as HTML tests, but they wouldn't stand a chance as formal tests even if the HTML interfaces where actually tested. Still, this work can benefit from m12n now and even more in the future. One pursuit is web accessible tests for browsers(?); both XML and HTML tests can be packed in such a document accessible on line but the only formal way to do so is to build a m12n schema. But I'm afraid that the real benefit of the m12n approach will only start showing on Level2 tests, if used. [1] http://www.w3.org/TR/html4/appendix/notes.html#h-B.1 -----Original Message----- From: Curt Arnold [mailto:carnold@houston.rr.com] Sent: Tuesday, March 12, 2002 10:00 PM To: www-dom-ts@w3.org Subject: Fw: Using existing staff.xml based tests with HTML processors Manos, Maybe I have misinterpreted your efforts, but I don't think they address our immediate objective. Moving staff.xml into modularized XHTML does not address the lack of DOM Core tests for existing (not X) HTML processors.
Received on Wednesday, 13 March 2002 06:39:31 UTC