TreeWalker

[Greg Matthews <gm...@flex.com.au>]

Hi,

I've been looking to convert from using the Java TR-2 xml package
to Xerces 1.0.1.

Haven't had much success in finding an equivalent for

TreeWalker.getNextElement( <tagName> );


I can create a TreeWalker and go through the nodes but...


1. wonder if there is an equivalent of the above, or i have
    to just loop through nodes myself until i get to the node
    i need.

2. would like to know if there's an easy way to skip over
    elements in a doc without drilling down to lower levels
    in the tree. e.g.

             <TOP><A><B/><A/><A><B/></A><A/></TOP>

    is there a way to create the treewalker at the <TOP> level then 
    just have it iterate over the <A> nodes, where in the example
    above all the <B> nodes happen to be child elements to the <A>'s.


Thanks
Greg.


PS. Is someone going to start putting in a few more javadoc 
      comments anytime soon? A few more helpful comments
      would be really good.

Re: TreeWalker

[Ralf Pfeiffer <rp...@apache.org>]

If you take a look at the factory member function for TreeWalker,
you can specify a nodetype or set of nodetypes which the treewalker will

walk, and ignore the rest. You can also set up a filter in which you
override
an acceptNode function, and accept some nodes, filter out others.

A UI we use for testing the TreeWalker implementation can be found in
the samples dir: dom.traversal.TreeWalkerView. It gives an example of
using a filter based on node name.

-rip

Greg Matthews wrote:

> Hi, I've been looking to convert from using the Java TR-2 xml
> packageto Xerces 1.0.1. Haven't had much success in finding an
> equivalent for TreeWalker.getNextElement( <tagName> );  I can create a
> TreeWalker and go through the nodes but...  1. wonder if there is an
> equivalent of the above, or i have    to just loop through nodes
> myself until i get to the node    i need. 2. would like to know if
> there's an easy way to skip over    elements in a doc without drilling
> down to lower levels    in the tree. e.g.
> <TOP><A><B/><A/><A><B/></A><A/></TOP>     is there a way to create the
> treewalker at the <TOP> level then    just have it iterate over the
> <A> nodes, where in the example    above all the <B> nodes happen to
> be child elements to the <A>'s.  ThanksGreg.  PS. Is someone going to
> start putting in a few more javadoc      comments anytime soon? A few
> more helpful comments      would be really good.