James Tauber wrote:
>> Now about that file system...
>Of course, this is the very thing I had in mind when I raised the idea a
>little while ago under the rubric of uberdocuments and XML/DOM shells. Now
>that we have XPath, the idea is even more attractive to me.
>What I would like to see is an implementation of the DOM and XPath with
There is the PDOM (persistent DOM) in both the free and
commercial versions, and its exactly what you ask for. It's a
pure Java implementation of a persistent DOM, with indexing and
caching for speedup.
The query processor on top uses XQL98 is instead of XPath, because
there was no XPath when the PDOM was implemented. XQL98 is, however,
very close to XPath and even more powerful.
As the PDOM can work with any DOM complient implementation (it
implements org.w3c.dom.Document) other query processors can be
put on top of it, not only the XQL98 engine that is there
by default. We are working on an XQL99 engine( XPath
will be a subset of XQL99) on top of the PDOM.
Right now XML-QLs are much in flow right now, so we'll wait at least
until the QL'99 workshop in Santa Teresa (Sept 15/16) before further
decisions are taken.
[GMDEngine] http://xml.darmstadt.gmd.de/xql (free)
[Infonyte] http://www.globit.com/infonyte.htm (commercial)