lucene-java-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Tom Howe" <>
Subject RE: Hierarchical document
Date Mon, 20 Oct 2003 15:06:28 GMT
>> Hi,
>> I have a very hierarchical document structure where each level of the
>> hierarchy contains indexable information.  It looks like this:  
>> 		Study -> 
>> 			Section -> 
>> 				DataFile -> 
>> 					Variable.
>> The goal is to create a situation where a user can execute a search
>> any level and the search would include all of the information below
>> in the hierarchy and retrieve the proper aggregated document.

>Say, you're on the level of Study/Section, then in indexing add the 
>fields "study" and "section" and set them to, say, "true". When 
>searching, just search where those two fields are "true".


Right, but my concern is that each of these levels are really different
documents.  So, I guess the question should have been, Do I need to
create documents for the lowest common denominator and then aggregate
them into higher level documents by hand or make a several document
types with redundant information and search by document type or create
multiple indices (one for each level) with redundant information?   In
other words, should I just add a bunch of DataFile documents that
contain Section and Study information and then, if a user wants a set of
Study documents, just aggregate them after the search by hand or is
there a more "lucene" way of doing this?  I'm trying to avoid storing
too much redundant information to implement this kind of hierarchical
structure, but that may not be possible.  I hope I'm being somewhat
clear with my question.

Thanks again,

To unsubscribe, e-mail:
For additional commands, e-mail:

To unsubscribe, e-mail:
For additional commands, e-mail:

View raw message