Hello!
I've buld a Portlet using the apache lucene API to search our website.
Indexing will be performed as a job in java (works fine).
My portlet is using the following method to do a simple search:
| private static List sucheSchlagwort(String queries)throws Exception{
| String index = "index";
| String field = "contents";
|
| int repeat = 0;
| boolean raw = false;
| String normsField = null;
| //
| //
| IndexReader reader = IndexReader.open(index);
| if (normsField != null)
| reader = new OneNormsReader(reader, normsField);
|
| Searcher searcher = new IndexSearcher(reader);
| Analyzer analyzer = new StandardAnalyzer();
| BufferedReader in = null;
| // in = new BufferedReader(new FileReader(queries));
| QueryParser parser = new QueryParser(field, analyzer);
| List alleErgebnisse = new ArrayList();
|
| String line = queries; // in.readLine();
|
| Query query = parser.parse(line);
|
|
| Hits hits = searcher.search(query);
|
|
| // ANFANG
| // FIXME: Ausgabe ins Ausgabeportlet, nicht println
| for (int i = 0; i < hits.length(); i++) {
| Ergebnis ergebnis = new Ergebnis();
| Document doc = hits.doc(i);
| String path = doc.get("path");
| if (path != null) {
| ergebnis.setLink(doc.get("path"));
| System.out.println(doc.get("path"));
| String title = doc.get("title");
| if (title != null) {
| ergebnis.setTitel(doc.get("title"));
| System.out.println(doc.get("title"));
| }
| } else {
| ergebnis.setLink("Kein Link zu diesem Dokument
vorhanden");
| }
| alleErgebnisse.add(ergebnis);
|
|
| if (queries != null) //
non-interactive
| break;
| }
| // ENDE
|
| reader.close();
| return alleErgebnisse;
|
| }
|
Implementet into a simple java class it works fine on shell. Started as part of
my portlet, it sends me the following error:
| 2006-12-08 12:34:42,393 ERROR [STDERR]
java.lang.ArrayIndexOutOfBoundsException: -1
| 2006-12-08 12:34:42,393 ERROR [STDERR] at
java.util.ArrayList.get(ArrayList.java:323)
| .
| .
| .
| 2006-12-08 12:34:42,413 ERROR [STDERR] at
org.apache.lucene.index.IndexReader.open(IndexReader.java:95)
| 2006-12-08 12:34:42,413 ERROR [STDERR] at
SuchportletPortlet.sucheSchlagwort(SuchportletPortlet.java:54)
| .
| .
| .
|
Looks like it comes from here:
| IndexReader reader = IndexReader.open(index);
|
Changing the index to a non existing path sends an error that the path is not
existing. So there must be a problem with reading the index. Don't you think?
What about the read/write permissions under the JBOSS directory?
And remember... in a standard java program it works...
Greetings,
Jan
View the original post :
http://www.jboss.com/index.html?module=bb&op=viewtopic&p=3992226#3992226
Reply to the post :
http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=3992226
_______________________________________________
jboss-user mailing list
[email protected]
https://lists.jboss.org/mailman/listinfo/jboss-user
