Dear fellow users, I was wondering if anyone is using Lucene right now to index data derived from business object models. My general problem is to index data which may be the result of an expensive computation involving a graph of objects (for example computing which customer has which items in their shopping cart or computing access privileges for objects). I would like to index this so that (for example) displaying a customers shopping cart is fast and so we can also see who is currently shopping for some certain items, etc. Of course, data would be more complex than this. Basically, it is the kind of thing you would do with a "view" in SQL DBs.
The main problem is updating this indexed view in a timely fashion when the business data changes. Ideally, updates should be instant, but of course that will not happen as potentially intensive computation is required and business objects are a bit more intelligent (and thus harder to deal with) than SQL table rows. I've found some research papers on the topic, but I was wondering if anyone does this right now and if they could explain their approach. I would be especially curious about update optimizations and the question if this can be done in a generic way or if this has to be built in to the business objects (e.g. to notice that the derived data has to be updated). Thanks in advance and best regards, Max Pfingsthorn --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
