I want read the single term in my index with Lucene

Question

I want to read every single index. I want to read and print to console the single term in my index. (I don't want to view the contents with Luke). Must I use the class IndexReader?

Can someone help me?

I tried to do:

    iReader = IndexReader.open(directory);

    int num = iReader.numDocs();
    for ( int i = 0; i < num; i++)
    {
        if ( ! iReader.isDeleted( i))
        {
            org.apache.lucene.document.Document d = iReader.document(i);
            System.out.println( "d=" +d.getField("title").tokenStreamValue());

        }
    }


    org.apache.lucene.document.Document doc = new org.apache.lucene.document.Document();

    //aggiungo tutti i documenti


    Field title = new Field(
              "title",
              testDoc.title,
              Field.Store.YES,
              Field.Index.ANALYZED,
              Field.TermVector.WITH_POSITIONS_OFFSETS);

    doc.add(title);


    Field content = new Field(
              "content",
              testDoc.content,
              Field.Store.YES,
              Field.Index.ANALYZED,
              Field.TermVector.WITH_POSITIONS_OFFSETS);
    doc.add(content);


    iWriter.addDocument(doc);

but d = null; Where did I go wrong? I want to retrieve the term to the Field title that I indexed...

Thanks very much.

score 1 · Accepted Answer · edited May 23 '17 at 09:58

1

Again, I use Java, but the principle will be the same.

What you want to do is similar to enumerating term frequencies, but you just care about distinct fields.

This example and this example on how to count term frequencies in a Lucene index should get you going.

edited May 23 '17 at 09:58

Community

1
1

answered Jan 14 '11 at 15:15

Joel

29,538
35
110
138

Thanks Joel, but I want only retrieve the term in a index...the term frequencies I do not care... – JackDaniels Jan 14 '11 at 15:37
@user568720: The code will answer your question. Just look at it. – Aaron Digulla Jan 14 '11 at 19:21

Aaron Digulla · Answer 2 · 2011-01-14T19:23:46.477

0

To examine the index, use IndexReader. The class has a method document(int) which you can use to find the individual documents that the index contains. The document then offers you all the fields that were created for that document.

With the field, you can either get it's value or the stream of tokens (i.e. the strings that end up in the index).

[EDIT] If you delete documents, the index will have holes. So you must add a check:

org.apache.lucene.document.Document d = iReader.document(i);
if( d == null ) continue; // <<-- You need this check

System.out.println( "d=" +d.getField("title").tokenStreamValue());

edited Jan 14 '11 at 19:23

answered Jan 14 '11 at 13:42

Aaron Digulla

321,842
108
597
820

I Aaron thanks for your response. I modify my post above. Where is the mistake?? – JackDaniels Jan 14 '11 at 14:05

score 0 · Answer 3 · answered Jan 14 '11 at 14:24

0

I'm using Lucene.Net, but I presume the logic is identical.

There must be exactly one of StringValue(), ReaderValue() and BinaryValue() set. Those not in use will either return null, or throw an exception. In your case, try reading StringValue() instead.

answered Jan 14 '11 at 14:24

sisve

19,501
3
53
95

Simon, if I use StringValue(), will return the value of doc.title .. I want the token... – JackDaniels Jan 14 '11 at 14:32
Just pass the document title through your analyzer and you'll get the resulting tokens. – sisve Jan 14 '11 at 14:42
Another solution would be to use IndexReader.GetTermFreqVector(documentId, fieldName), but that requires that you index your field with TermVector.YES. – sisve Jan 14 '11 at 14:46
I did not understand. Where do I do this operation? and what should I use? Thanks for your patience – JackDaniels Jan 14 '11 at 15:06
See answer from Joel, he's already provided some links to get you started. – sisve Jan 14 '11 at 15:20

I want read the single term in my index with Lucene

3 Answers3