AFAIU, this is normal mode of operation for XML parsers and DOM / SAX. They can always split text into multiple TextNodes or character() events as they want it. I'm not sure there is way around this; I think you should collect all consecutive TextNodes and combine them. Vadim
There's a normalize() method in org.w3c.dom.Node that solves this problem, putting all Text nodes into one. Description:
Thanks all for your answers.
I changed to dom4j and found the getText() method returns the whole content
Martin
