Hello, using PatternAnalyzer solved the problem as well, using a whitespace pattern. There's lowercase support as well (see http://lucene.apache.org/java/2_2_0/api/org/apache/lucene/index/memory/PatternAnalyzer.html)
Many thanks. Regards, Ueli Kistler -----Ursprüngliche Nachricht----- Von: AHMET ARSLAN [mailto:iori...@yahoo.com] Gesendet: Freitag, 14. August 2009 12:56 An: java-user@lucene.apache.org Betreff: Re: AW: AW: Wildcard search fails > Noticed that in Luke... is there any existing analyzer around that > supports case-insensitive search and recognizes "RZ/G/17" as one token? As far as I know there is no built-in analyzer that uses whitespace tokenizer and lowercase filter together. But it is easy to cast tokenizer and token filters to create a new analyzer. You said that you were using SnowballAnalyzer, right? Just replace tokenizer from standard to whitespace. Edit SnowballAnalyzer.java : public TokenStream tokenStream(String fieldName, Reader reader) { TokenStream result = new WhitespaceTokenizer(reader); result = new LowerCaseFilter(result); if (stopSet != null) result = new StopFilter(result, stopSet); result = new SnowballFilter(result, name); return result; } By changing this method you can create custom analyzers. If you do not want stemming just erase the line that contains SnowballFilter, etc. Since you have lowercasefilter in your analyzer, your wildcard queries should be lowercased. Alternatively you can set setLowercaseExpandedTerms(true) of your QueryParser. --------------------------------------------------------------------- To unsubscribe, e-mail: java-user-unsubscr...@lucene.apache.org For additional commands, e-mail: java-user-h...@lucene.apache.org --------------------------------------------------------------------- To unsubscribe, e-mail: java-user-unsubscr...@lucene.apache.org For additional commands, e-mail: java-user-h...@lucene.apache.org
