关于关键字搜索和lucene.net C＃

Question

i am just browsing article on lucene.net. i got some sample code for create index using lucene.net and few lines of code is not clear to me. here is those line

    protected void btnCreateIndex_Click(object sender, EventArgs e)
{
    IndexWriter writer = new IndexWriter(MapPath("~/searchlucene/"), new StandardAnalyzer(), false);

    IndexDocument(writer, "About Hockey", "hockey", "Hockey is a cool sport which I really like, bla bla");
    IndexDocument(writer, "Some great players", "hockey", "Some of the great players from Sweden - well Peter Forsberg, Mats Sunding, Henrik Zetterberg");
    IndexDocument(writer, "Soccer info", "soccer", "Soccer might not be as fun as hockey but it's also pretty fun");
    IndexDocument(writer, "Players", "soccer", "From Sweden we have Zlatan Ibrahimovic and Henrik Larsson. They are the most well known soccer players");
    IndexDocument(writer, "1994", "soccer", "I remember World Cup 1994 when Sweden took the bronze. we had great players. players , bla bla");
    IndexDocument(writer, "BBA-header", "BBA-321type", "Hello BBA");

    writer.Optimize();
    writer.Close();

}
private void IndexDocument(IndexWriter writer, string sHeader, string sType, string sContent)
{
    Document doc = new Document();

    doc.Add(new Field("header", sHeader, Field.Store.YES, Field.Index.TOKENIZED));
    doc.Add(new Field("type", sType, Field.Store.YES, Field.Index.TOKENIZED));
    doc.Add(new Field("content", sContent, Field.Store.YES, Field.Index.TOKENIZED));
    writer.AddDocument(doc);
}

i have couple of question

1) doc.Add(new Field("header", sHeader, Field.Store.YES, Field.Index.TOKENIZED)); what is the meaning of this line. Field.Index.TOKENIZED what is TOKENIZED & UNTOKENIZED?? when i search keyword specified in type argument then nothing is coming. just do not understand the behaviour

here is sample for search where i specify a keyword which was index as type

    ListBox1.Items.Clear();
    var searcher = new Lucene.Net.Search.IndexSearcher(MapPath("~/searchlucene/"));
    var oParser = new Lucene.Net.QueryParsers.QueryParser("content", new StandardAnalyzer());
    string sHeader = " OR (header:" + TextBox1.Text + ")";
    string sType = " OR (type:" + TextBox1.Text + ")";
    string sSearchQuery = "(" + TextBox1.Text + sHeader + sType + ")";

    var oHitColl = searcher.Search(oParser.Parse(sSearchQuery));
    for (int i = 0; i < oHitColl.Length(); i++)
    {
        Document oDoc = oHitColl.Doc(i);
        ListBox1.Items.Add(new ListItem(oDoc.Get("header") + oDoc.Get("type") +  oDoc.Get("content")));            
    }

    searcher.Close();

please someone help me to understand to drive out my confusion. thanks

Answer 1

I just tested your code, and it works fine with Lucene 2.9.4.

Field.Index.TOKENIZED means the Analyzer will break your text in tokens, meaning it will be searchable in full-text. You would use UN_TOKENIZED for fields you dont want analyzed, like product IDs.

Note: you should use Field.Index.ANALYZED and Field.Index.NOT_ANALYZED which are the replacements for their deprecrated TOKENIZED / UN_TOKENIZED counterparts.

To see differences between analyzed and not, you can try both and use Luke to inspect your indexes, that will probably give you a good idea of how it works.

http://code.google.com/p/luke/