Lazy field loading in

[jafarim <ja...@gmail.com>]

Hi
I am seeking for making use of the latest lazy field loading in lucene 2.1.
I store the orignal bytes of a document, say a PDF file for example, in a
special untokenized field in the index. Though there is enough facilities in
IndexReader class for lazy field loading, the search API in IndexSearcher
does not contain such facilities. Hence, the Documents I get from the
Hits.doc() would not benefit from the mentioned feature.
Am I missing an important point or this is a desired feature to go on the
todo list?
--Jafarim

Re: Lazy field loading in

[Chris Hostetter <ho...@fucit.org>]

: IndexReader class for lazy field loading, the search API in IndexSearcher
: does not contain such facilities. Hence, the Documents I get from the
: Hits.doc() would not benefit from the mentioned feature.

Lazy loading stored fields is really about perfermance tweaking ... if
yoiu are that concerned baout performance, you shouldn't be using Hits at
all.

there was a change made to the trunk recently to support
Searchable.doc(int,FieldSelector) but i dont' think you'll ever see a
method returning Hits which takes in a FieldSelector .. use TopDocs or
TopFieldDocs and fetch the docs with calls to doc().



-Hoss


---------------------------------------------------------------------
To unsubscribe, e-mail: java-user-unsubscribe@lucene.apache.org
For additional commands, e-mail: java-user-help@lucene.apache.org