processing html tags inside xml with fop

[jamieg99]

Hello,

I've got an object that's being converted into SAXSource and then converted into a pdf with FOP. Some of the data however is in HTML format inside the xml tags and is being escaped (>, etc) before it transformed.

I'd like to have this these html tags parsed as actual elements by the stylesheet, which would mean reading it in it's unescaped format, but I can't figure out how to do this. I have "disable-output-escaping" set in the stylesheet, but I think the data in the xml has already been parsed as "escaped" before it gets to the stylesheet and processed.

Here's the code for converting.


FOUserAgent foUserAgent = getUserAgent();

PDFRenderer pdfrenderer = new PDFRenderer();
pdfrenderer.set UserAgent(foUse rAgent);
foUserAgent.set RendererOverrid e(pdfrenderer);

URIResolver resolver = myWebContext.ge tResolver();
foUserAgent.set URIResolver(res olver);

ByteArrayOutput Stream out = new ByteArrayOutput Stream();

byte[] b = null;
try {

TransformerFact ory factory = TransformerFact ory.newInstance ();


Transformer transformer = factory.newTran sformer();

//transformer.set OutputProperty( "disable-output-escaping", "yes");
// kicks back error --- invalid property


Fop fop = fopFactory.newF op(MimeConstant s.MIME_PDF, foUserAgent, out);

Source xsl = resolver.resolv e(xslFilename, null);

transformer = factory.newTran sformer(xsl);

res = new SAXResult(fop.g etDefaultHandle r());

transformer.tra nsform(xmlSrc, res);

b = out.toByteArray ();


Any help would be greatly appreciated.

Thanks!

-Jamie

[jamieg99]

After much futzing around with this, I ended up using JAXB to generate the xml I needed and enclosing the HTML data inside CDATA sections.

I then changed TransformerFact oryImpl to use saxon and process the stylesheet:

System.setPrope rty("javax.xml. transform.Trans formerFactory", "net.sf.saxon.T ransformerFacto ryImpl");

And used saxon's saxon:parse() function to further process the html-tag data inside the CDATA section since it's well-formed html. That gave me an xml result for fop to handle which included the html I needed to create the correct pdf formatting.

It now works, although probably not the greatest of solutions.

I couldn't figure out how to generate CDATA sections with SAX and LexicalHandlers , so this was the next best alternative.

Cheers


-Jamie