I just discovered that this problem does not exist in Xerces 1.4.4, only in 2.0.0. But I don't know, if I can use the old version temporarily in our product. I downloaded the current Xerces development sources to see, if maybe we could use this version, but there is no xercesImpl.jar inside. Do I have to compile and build this jar-file myself from the current sources? Couldn't this be integrated into the nightly build process (just wondering)?
> -----Original Message----- > From: Thomas B�rkel > Sent: Dienstag, 26. Februar 2002 12:54 > To: Xerces Mailinglist > Subject: Bad bug in Xerces? > > > HI! > > Consider the following XML, which is being built with > dbf = DocumentBuilderFactory.newInstance(); > db = dbf.newDocumentBuilder(); > doc = db.parse(new InputSource(new StringReader(text))): > > <convert><p0><![CDATA[100200]]></p0><p1><![CDATA[<Auswahl>\r\n > <pos nr='1' > menge='100'/>\r\n</Auswahl>]]></p1><p2><![CDATA[Auftrag]]></p2 > ></convert> > > > If you write out the document after this with > streamResult = new StreamResult(new StringWriter()); > transformer = > TransformerFactory.newInstance().newTransformer(); > transformer.transform(new DOMSource(node), streamResult), > then you get this: > > <convert><p0><![CDATA[100200]]></p0><p1><![CDATA[100200<Auswah > l>\r\n<pos nr='1' > menge='100'/>\r\n</Auswahl>]]></p1><p2><![CDATA[Auftrag]]></p2 > ></convert> > > > Clearly, the XML is damaged. And it's not the Transformer's > fault, because if you iterate through the DOM, you see that > the 2nd CDATA node really has the wrong data. > > > Anyone any idea, please? Am I doing something wrong? > > > Regards, > Thomas > > > --------------------------------------------------------------------- > 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]
