FAQ

Question: If i understand correctly, the &= version is using the lucene index. The = version is using the structural index.

Answer: &= is not using the lucene index, but eXist's own internal full text index (which is not as fast as the lucene implementation). The lucene index is a new feature and provides its own extension functions.

Question: What do you mean by value index?

Answer: A "value" or "range" index supports standard comparisons with =, >, < and the like. In short, the structural index helps with the expression

collection("/db")/doc/attribute/rut_afiliado

but you will also need a range index to speed up the value lookup on rut_afiliado:

rut_afiliado. ='8859471-1'

Folow reading: http://exist-db.org/indexing.html#rangeidx

Question: I am doing a performance test of query operation in eXist，and the result following showed time consumption for query about 10 records increases when the total number of all records stored in database increases. There is no indexes.

Answer: Without an index, if your query involves a comparison, eXist has to do a full scan over all the relevant nodes in the database. The more data is in the db, the more nodes have to be scanned.

Please check http://www.exist-db.org/tuning.html and create the proper indexes.

Question: When I insert a collection with new XML documents, the total size of files stored at "eXist.webapp.WEB-INF.data" increases (obviously). But, when I delete a collection or a set of collection the size of the same files does not decrease?? Why this happens?

Answer: The size of the .dbx files will never decrease. If you delete a document or collection, the pages used by that resource will be freed or rather: marked as deleted. They will be reused for the next document or collection you store. Maybe it is easier to understand if you view the .dbx files as "partitions" rather than "files". The db files will never shrink, though they can be empty.