Hey,
Sorry for the length of time replying. It completely slipped my mind that you now have to turn PSVI info on to get the info.
setFeature(XMLUni::fgXercesDOMHasPSVIInfo, true)
Is what you need. The old code was not thread safe, so when we had full PSVI info we decided to move across to using that. It means we can lose that old code as well, which will increase performance quite a lot.
Regards,
Gareth
Steve Harrington wrote:
Gareth,
I have attached a stripped down set of files having removed most of the non-related code. This small set of files includes a main method which will illustrate the problem. You can run it against the XML file pset-testing.xml (which is an instance of the schema pset.xsd) - all included in the tar file. It works fine with xerces 2.5 but fails under 2.6. Any insight is appreciated.
After building it, to execute simply type:
testDOM -file pset-testing.xml
Let me know if you need more information.
Thanks, Steve
Gareth Reakes wrote:
Hi,
Could you provide us with a small snippet of xml and the code you use to parse the document?
Cheers,
Gareth
Steve Harrington wrote:
Hi,
I had some working code that I was running with xerces 2.5 and when I changed to xerces 2.6, it broke. Did something related to DOMTypeInfo change? Here's the code snippet:
for (unsigned int i = 0; i < paramNodes->getLength(); i++) { DOMNode * paramItem = paramNodes->item(i); if(paramItem->getNodeType() == DOMNode::ELEMENT_NODE) { DOMElement * paramElem = (DOMElement*)paramItem; const DOMTypeInfo * typeInfo = paramElem->getTypeInfo(); const XMLCh* typeName = typeInfo->getName(); // check the type of parameter and instantiate the proper class if(XMLString::equals(typeName, INT_PARAM_TYPE)) { handleIntParam(paramElem); } // etc... }
Under 2.5, this worked fine whereas now it seems that I'm getting nothing for the typeName (from DOMTypeInfo::getName method). I realize it says "experimental and subject to change" on the API documentation - perhaps I've played with fire and am now being burned :-) If there is a better way to get the name of the type then what I'm doing here, I'd be happy to change my code.
Cheers, Steve
--------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
------------------------------------------------------------------------
--------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
-- Gareth Reakes, Managing Director Parthenon Computing +44-1865-811184 http://www.parthcomp.com
--------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
