Randy, I can't say for sure, what you do looks right to me. I am experiencing sometimes that Xindice (dbxml 1.0b4) looses the connection to the indexes. They are not being used in queries and not listed in a dbxmladmin list_indexers command. I suggest that you try dbxmladmin list_indexers to see if Xindice is aware of the indexes in the first place. Otherwise you will have to rebuild them and try again. This usually works for me, at least for a while.
Anders ----- Original Message ----- From: "Randy George" <[EMAIL PROTECTED]> To: <[email protected]> Sent: Thursday, January 24, 2002 10:24 PM Subject: Re: newbie question on Xindice indexing > Hello, > > I hope this is not an extraordinarily dumb question but I am having some > difficulty with an experiment using Xindice. > > I have loaded a collection with about 59000 extremely small xml docs that > look like this: > > <?xml version="1.0"?> > <quad> > <name>Rising Sun</name> > <index>39076F1</index> > <state>MD</state> > <state>PA</state> > </quad> > > I then create an index : > C:\dbxml-test>dbxmladmin ai -c /db/usgs -n quadndx -p index > CREATED : quadndx > > and perform a query: > C:\dbxml-test>dbxml xpath -c /db/usgs -q "/quad[index='32088F4']" > > with this result after a very long wait: > <?xml version="1.0"?> > <quad xmlns:src="http://www.dbxml.org/Query"; src:col="/db/usgs" > src:key="32088F4"> > <name>Porterville</name> > <index>32088F4</index> > <state>MS</state> > <state>AL</state> > </quad> > > If I do a retrieve document on the key there is no appreciable wait on the > retrieval: > C:\dbxml-test>dbxml rd -c /db/usgs -n 32088F4 > > Am I creating the Xindice index with the correct pattern for speeding access > to the <index> tag? > Sorry for the "index" confusion - the tag <index> is the USGS index book id > for a given quad > > Thanks > Randy George >
