Hello sirs !

I would like to submit a problem I have concerning an output of  a DOM object
to a file.

I have made a method which write this object in a tree-view, by adding
recursively new-line characters and spaces, depending on the depth of the 
node. That works fine for objects newly created.

If I load an already existing  XML file for parsing (file which contains
already a tree-aspect), play with it and try to ‘output’ it to a file, I get
double new-line characters and spaces. So I’ve implemented a ‘boolean’ to
indicate if it’s desired a tree-view or not (to make the distinction between
the newly created DOM object and that one taken from a existing file). This
works fine as long as I keep this clear separation.

What if I’m loading a DOM object and try to add some new elements? 

If  I have:   addNewLineCharacter(false)
I’ll get something like:
       <oldelement>
         sometext
       </oldlement><newelement>another text</newelement>

If  I have:   addNewLineCharacter(true)
I’ll get something like:
       <oldelement>

           sometext

       </oldlement>
       <newelement>
         another text
       </newelement>

So my question is: does an element with one ‘child’ contains some _TextNode_s
too, for ‘\n’ characters and spaces? How can I ignore the ‘whitespaces’?

I have implemented myself the writing method and I can’t understand why if I’m
loading a XML file and having it in memory, I still have those characters used
only for ‘pretty-printing’.

I would rather prefer some ‘technical explanations’ than pointing to some
existing methods.

Sincerely, 

silviu leahu


***************************************************************************
This is xml-dev, the mailing list for XML developers.
To unsubscribe, mailto:majordomo@x...&BODY=unsubscribe%20xml-dev
List archives are available at http://xml.org/archives/xml-dev/
***************************************************************************