On Tuesday 13 May 2003 18:53, Kelvin Tan wrote:
> I confess that I've never quite understood the use of Field.UnStored. So
> the field is indexed but the original content is not saved?
Actually, that makes lots of sense in many cases. For example, original
content is likely to be already stored in a database, to be accessed by
non-search functionality, using normal SQL (etC) access. But searches are
indexed using Lucene, because of its greater flexibility (OracleText,
formerly known as Intermedia has similar features, but its proprietary...
other DBs may have similar proprietary extensions). This way content won't be
stored twice. Similarly, for web pages, pages are likely to already exist in
the file system. Why store a duplicate in the index?
In fact I think more often than not storing content in Lucene index doesn't
make much sense. As long as content is stored somewhere from where it can
easily be accessed (using unique id or so), chances are that searches (not to
mention indexing) are also faster when content is not stored, just indexed?
Just my 2 cents,
-+ Tatu +-
---------------------------------------------------------------------
To unsubscribe, e-mail: lucene-user-unsubscribe@jakarta.apache.org
For additional commands, e-mail: lucene-user-help@jakarta.apache.org
|