12.09.2010, в 06:36, Gregoire написал(а): > I've created a test-case so that it is easier to answer my question.
This is a combination of a WebKit bug, and a mistake in xsltforms.xsl. The WebKit bug has been fixed last month as part of HTML5 parser compliance effort, you can see that your test case works as expected in WebKit nightlies <http://nightly.webkit.org>. The xsltforms.xsl problem is that it produces an HTML document with lots of pseudo-XML cruft in it, which used to confuse WebKit. Notably, on line 84 it outputs '<html xmlns="http://www.w3.org/1999/xhtml";>' - removing the default namespace fixes the problem in Safari 5.0.2. I'm attaching the result an XSL processor gives for test-chrome-xslt.xml for your reference. You can see garbage xmlns attributes on <html>, <script> and <dataisland> - these serve no purpose when output="html" (as it is in xsltforms.xsl).Title: WebKit's XSLTProcessor transformToFragment test case
instance-announces
| Id | price |
|---|---|
| ? | ? |
Sort by 'price' : AscendingDescending
... Loading ...
- WBR, Alexey Proskuryakov
_______________________________________________ webkit-help mailing list [email protected] http://lists.webkit.org/mailman/listinfo.cgi/webkit-help
