On Thu, 2005-09-15 at 11:56 -0700, James Huang wrote:
> Yes, "+" is what I missed! Thanks.
>
> Suppose there is a book published by 3 publishers (I
> don't know how that works in real world):
>
> // At index time:
> doc.add( Field.Keyword("publisher", "Manning") );
> doc.add( Field.Keyword("publisher", "SAMS") );
> doc.add( Field.Keyword("publisher", "O'Reilly") );
>
> // At search time:
> queryString += " +publisher:SAMS";
> ...
>
> should find me that Document.
That may or may not work depending on your analyzer.
If you're using the query parser with the standard analyzer it will
search the 'publisher' field for 'sams' not 'SAMS', and hence get no
matches back.
If you want to use the query parser instead of building the query by
hand you can use the PerFieldAnalyzerWrapper class and write a
KeywordAnalyzer, i.e.:
package org.apache.lucene.analysis;
import java.io.IOException;
import java.io.Reader;
/** "Tokenizes" the entire stream as a single token. */
public class KeywordAnalyzer extends Analyzer {
public TokenStream tokenStream(String fieldName, final Reader reader) {
return new TokenStream() {
private boolean done;
private final char[] buffer = new char[1024];
public Token next() throws IOException {
if (!done) {
done = true;
StringBuffer sb = new StringBuffer();
int length;
while (true) {
length = reader.read(this.buffer);
if (length == -1) break;
sb.append(this.buffer, 0, length);
}
String text = sb.toString();
return new Token(text, 0, text.length());
}
return null;
}
};
}
}
--------------------
PerFieldAnalyzerWrapper result =
new PerFieldAnalyzerWrapper(new StandardAnalyzer());
result.addAnalyzer("publisher", new KeywordAnalyzer());
QueryParser parser = new QueryParser(<your regular field>, result);
--
Miles Barr <[EMAIL PROTECTED]>
Runtime Collective Ltd.
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
