best practice for reusing documents with multi-valued fields

[Christopher Condit <co...@sdsc.edu>]

I know that it's best practice to reuse the Document object when
indexing, but I'm curious how multi-valued fields affect this. I tried
this before indexing each document:

doc.removeFields(myMultiValuedField);
for (String fieldName: fieldNames) {
  Field field= doc.getField(field);
  if (null != field) {
    field.setValue("");
  } else {
    System.out.println("null field");
  }
}

And it seems that the null fields (which I had removed) are just
accumulating. Is there a better way to reuse a document with
multi-valued fields?

Thanks,
-Chris

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

Re: best practice for reusing documents with multi-valued fields

[Anshum <an...@gmail.com>]

Hi Chris,
So doc.removeFields works fine, just tried it again. You could try using an
approach on the lines of the one mentioned below.

*--snip--*
IndexWriter iw = new IndexWriter(indexDir, new
StandardAnalyzer(Version.LUCENE_30),true, MaxFieldLength.UNLIMITED);
 Document doc = new Document();
doc.add(getField(FIELD1, "aa"));
 doc.add(getField(FIELD1, "bb"));
doc.add(getField(FIELD2, "cc"));
 iw.addDocument(doc);
doc.removeFields("field1");
doc.add(getField(FIELD1, "aaa"));
        doc.add(getField(FIELD1, "bbb"));
        doc.add(getField(FIELD2, "ccc"));
        iw.addDocument(doc);
        IndexReader ir = iw.getReader();
        IndexSearcher is = new IndexSearcher(ir);

        BooleanQuery query = new BooleanQuery();
        query.add(new TermQuery(new Term("field1", "aa")), Occur.MUST);
        ScoreDoc[] sd = is.search(query, 10).scoreDocs;
        for(ScoreDoc scoreDoc:sd){
              System.out.println(ir.document(scoreDoc.doc));
        }
        is.close();
        ir.close();
        iw.close();

*--Snip--*



--
Anshum Gupta
http://ai-cafe.blogspot.com


On Fri, Apr 15, 2011 at 6:32 AM, Christopher Condit <co...@sdsc.edu> wrote:

> I know that it's best practice to reuse the Document object when
> indexing, but I'm curious how multi-valued fields affect this. I tried
> this before indexing each document:
>
> doc.removeFields(myMultiValuedField);
> for (String fieldName: fieldNames) {
>  Field field= doc.getField(field);
>  if (null != field) {
>    field.setValue("");
>  } else {
>    System.out.println("null field");
>  }
> }
>
> And it seems that the null fields (which I had removed) are just
> accumulating. Is there a better way to reuse a document with
> multi-valued fields?
>
> Thanks,
> -Chris
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: java-user-unsubscribe@lucene.apache.org
> For additional commands, e-mail: java-user-help@lucene.apache.org
>
>