Getting a iterator over a huge xml-file

Question

Hi !
I have to persist data from some huge xml-files.
The structure of the xml is simple, like in:
<items>
<item>
</item>
<item>
</item>
</items>
The best way for me, is to get an iterator over the item-elements represented as a DOM strcture for easy access.
Because the files are so huge (80-150 MBytes) and I have no control over the creation of them, the underlying parser should be SAX oriented.
So I need a solution which will parse until one item-element (maybe spezified through a XPATH query) is completed and build a DOM-tree from it.
Then I can pass this DOM-element to my persistence layer. When this item is persisted, and the Iterator hasNext(), then I could do a simple .next() to get the next DOM representation of the item.
Is there such a solution out ? Or do I have to 'invent' such kind of parsing ?
Would by great to get some hints ... this problem is driving me crazy :)
Thank you in advance,
Gerald

Answer 1

Vaguely familiar, http://lists.xml.org/archives/xml-dev/200201/msg00032.html
I can't think of a DOM builder that would work off a fragment; you would probably have to create one yourself.
There's also http://drizzle.stanford.edu/~peastman/pax.html which might solve your problem.
Pete