mis parsing element nodes

[Claire Suttle <cl...@geoviz.com>]

Hi,

I've been creating a new document by importing an element node and its children from one document into another, and writing the result out to a new file.

When I then re-parse the resulting document, I find that the element directly below the one that was inserted is mis-interpreted as a deferred-text node, rather than being read as an element node. 

I've been using the same code to import an element from one document into another in a few places in my project and not had any problems. The difference is that this time, I am inserting the imported node before an existing node, rather than appending to a list of element nodes. 

The problem seems to be a missing carriage return between the end tag of the imported element and the start tag of the next element. If I go into the file on my hard drive and add that carriage return by hand, then everything parses correctly. - Which seems odd to me, as my inserted element is all inserted on one long line and it manages to read all the child elements in there. There's a couple of references to carriage return issues in the bug lists for xerces, and I wondered if this was a known issue, or is there were any work arounds for this.

Thanks for any help,
Claire