On Wed, 2003-07-23 at 21:54, Hugo Burm wrote: > I dump my DOM node with org.apache.cocoon.xml.dom.DOMUtil.node2String to a > string. > Then I try to parse this string, hoping to create a copy of the original > node, but I get a number of parser errors. > > The problem is that node2String expands entities like "&" into "&". And > when I parse the resulting string, I get the errors I could wait for. > > I can do a workaround myself (search and replace). But I am wondering: is > this behaviour by intention or is it a bug? If not, how can I do this round > trip more gracefully? > > I have the feeling that I did miss something elementary. But I did all the > obvious things (checking the archives and trying Google). >
Nope, you didn't miss anything. If you have a look at the sources you'll see that no escaping is done on the data. However, you're better of using an identity TransformerHandler transformation. There's also an utility function in Cocoon which helps with this: org.apache.cocoon.xml.XMLUtils.serializeNodeToXML(Node node) -- Bruno Dumon http://outerthought.org/ Outerthought - Open Source, Java & XML Competence Support Center [EMAIL PROTECTED] [EMAIL PROTECTED]
