hi all,

I have a object like this

public class Link {
  private long id;
  private string url;
  // other 20 property
  private int vote; //summary of vote, for sorting used
}

so when I index document, my Lucene document also contain all field from my
Link object. e.g

doc_id = 1
url = "solr.org
vote = 23

because the vote is change frequently then other property, every time the
vote is increase, I need to reindex the whole document so I can use vote
field for sorting.

is that anyway just to index partial Lucene document (in this case, only
vote field) instead of index again whole document ? any performance concern
if is index whole doc (I think is not look like a good solution).

any idea ?

kiwi

Reply via email to