> -----Original Message----- > From: Dave Hoffer > Sent: Saturday, August 23, 2003 10:17 AM > To: '[EMAIL PROTECTED]' > Subject: Xerces-C v2.2 CR LF bug/question > > I have a question about how Xerces handles reading and writing strings that contain > CR and LF characters. > > I have an application that allows the user to enter a multiline text string. The > string may have multiple CR LF pairs. When I save the string to the XML file, all > is well. I can look at the file and find the CR LF pairs. (I am saving in UTF-8 > format but my application is UTF-16.) > > However, when my application is shut down and restarted and I load the same file, > when I read the string the CR LF pair is replaced with a single LF. Why is this > happening? My code structure for reading the file is something like... > > // This part is used when creating a new document and when opening an existing > document... > DOMImplementation* pDOMImplementation = > DOMImplementationRegistry::getDOMImplementation(L"Core"); > m_pDomDocument = pDOMImplementation->createDocument(NULL, L"Target", 0); > m_pDomDocument->setEncoding(WSZ_ENCODING_UTF_8); > ... > // This part is only used when opening an existing document... > m_pDomDocument->release(); > m_pDomDocument = NULL; > DOMImplementation* pDOMImplementation = > DOMImplementationRegistry::getDOMImplementation(L"LS"); > m_DOMBuilder = > ((DOMImplementationLS*)pDOMImplementation)->createDOMBuilder(DOMImplementationLS::MODE_SYNCHRONOUS, > 0); > m_DOMBuilder->setFeature(L"namespaces", true); > DOMDocument* pDomDocument = m_DOMBuilder->parseURI( m_strTargetSpec ); > m_pDomDocument = (DOMDocument*)pDomDocument->cloneNode(true); > m_pDomDocument->setEncoding(pDomDocument->getEncoding()); > > > Thanks for any help! > > -dh > >
--------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
