XOM API for Processing XML with Java

[Elliotte Rusty Harold <elharo AT metalab.unc.edu>]

At 2:15 PM -0500 2/11/04, jcowan AT reutershealth.com wrote:


> I note that Serializer is recursive,
> and should be fixed on the same grounds:  see
> https://lists.ibiblio.org/sympa/arc/xom-interest/2002-September/000082.html
> for the non-recursive tree-walking algorithm I used in DOMParser.

That algorithm was:

Implementation note:  When I wrote DOMParser, I was careful *not* to
use recursion, because there would be no guarantee that the Java stack
wouldn't blow up.  Instead I used the incremental version of walking
the whole tree:

        1) firstChild, unless null, in which case
        2) nextSibling, unless null, in which case
        3) parent() and try again, unless root, in which case
        4) stop.

I don't think I'm going to fix this in Serializer because the API 
change would be too limiting, but I could fix it in toXML and copy if 
I could figure out how to do that. However, short of setting up my 
own non-fixed size stack to replace the Java stack, I don't see any 
way to write this non-recursively. In particular, the "try again" in 
step 3 seems to be recursion in sheep's clothing. :-) I'm not sure 
what I'm not seeing here. Is there a simple way to implement this 
algorithm non-recursively? Or do I just need to store a stack of 
parent nodes that can grow larger than the Java stack?
--

  Elliotte Rusty Harold
  elharo AT metalab.unc.edu
  Effective XML (Addison-Wesley, 2003)
  http://www.cafeconleche.org/books/effectivexml            
  http://www.amazon.com/exec/obidos/ISBN%3D0321150406/ref%3Dnosim/cafeaulaitA