p2p.wrox.com Forums - <?javax.xml.transform.disable-output-escaping ?>

Subject:	<?javax.xml.transform.disable-output-escaping ?>
Posted By:	robbert	Post Date:	8/24/2006 1:10:05 PM
Hi there, I'm using javax.xml.transform.Transformer.TransformerFactory to transform an XML document by an XSLT document. In the XSLT document, there are commands like, <xsl:text disable-output-escaping="yes"><![CDATA[ <script>...</script></xsl:text> The resulting document will look like, <?javax.xml.transform.disable-output-escaping ?><script>...</script><?javax.xml.transform.disable-output-escaping ?> I've already Google'd for that, extensively - and found several related posts - but not a solution... Thanks Robert
Reply By:	mhkay	Reply Date:	8/25/2006 6:15:59 PM
The JAXP interface specifies that disable-output-escaping requests are passed in this way from the XSLT transformer to a user-written ContentHandler. This will only happen if you don't serialize the output directly. It's best not to use d-o-e unless you are serializing the output (actually, it's best not to use it at all). Michael Kay http://www.saxonica.com/ Author, XSLT Programmer's Reference and XPath 2.0 Programmer's Reference
Reply By:	robbert	Reply Date:	8/27/2006 6:51:32 AM
Thanks Michael. The "ContentHandler" is a DOM4J DocumentResult object in my case. quote: This will only happen if you don't serialize the output directly. I didn't understand that, sorry. quote: It's best not to use d-o-e unless you are serializing the output (actually, it's best not to use it at all). In my understanding, using d-o-e breaks the rule that the XSL transformer should be able to produce the result tree in one pass. A second pass will be necessary. Nevertheless, it's obvious that I need to integrate the <script> tags into the document, at any rate... Are common techniques to accomplish that?
Reply By:	mhkay	Reply Date:	8/27/2006 7:11:22 AM
disable-output-escaping is an instruction to the serializer, that is the component that turns the result tree into lexical XML. If your result is not lexical XML but a DOM4J DocumentResult, then d-o-e has no meaning. The XSLT 2.0 spec says that in such cases it should be ignored. But the JAXP spec takes into account that when you send to a ContentHandler, the ContentHandler might be acting as a serializer and might want to know about the request to disable-output-escaping. So it defines this processing instruction to convey the request. Why are you generating the script element as text, rather than as a regular element node? This is the root of your trouble: you're thinking in terms of generating lexical XML rather than generating a result tree. Michael Kay http://www.saxonica.com/ Author, XSLT Programmer's Reference and XPath 2.0 Programmer's Reference
Reply By:	robbert	Reply Date:	8/27/2006 8:51:44 AM
quote: Why are you generating the script element as text, rather than as a regular element node? This is the root of your trouble Muhahaha! That one's cured! - Thanks! Also thanks for the background information. I was coming from the Microsoft world - that, I think, says it all.

Go to topic 48912

Return to index page 192
Return to index page 191
Return to index page 190
Return to index page 189
Return to index page 188
Return to index page 187
Return to index page 186
Return to index page 185
Return to index page 184
Return to index page 183