You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@nutch.apache.org by je...@apache.org on 2006/03/12 18:46:08 UTC
svn commit: r385322 - in /lucene/nutch/trunk/src/java/org/apache/nutch:
analysis/AnalyzerFactory.java indexer/Indexer.java
Author: jerome
Date: Sun Mar 12 09:46:07 2006
New Revision: 385322
URL: http://svn.apache.org/viewcvs?rev=385322&view=rev
Log:
Reactivate usage of AnalyzerFactory
Modified:
lucene/nutch/trunk/src/java/org/apache/nutch/analysis/AnalyzerFactory.java
lucene/nutch/trunk/src/java/org/apache/nutch/indexer/Indexer.java
Modified: lucene/nutch/trunk/src/java/org/apache/nutch/analysis/AnalyzerFactory.java
URL: http://svn.apache.org/viewcvs/lucene/nutch/trunk/src/java/org/apache/nutch/analysis/AnalyzerFactory.java?rev=385322&r1=385321&r2=385322&view=diff
==============================================================================
--- lucene/nutch/trunk/src/java/org/apache/nutch/analysis/AnalyzerFactory.java (original)
+++ lucene/nutch/trunk/src/java/org/apache/nutch/analysis/AnalyzerFactory.java Sun Mar 12 09:46:07 2006
@@ -81,7 +81,9 @@
Extension extension = (Extension) this.conf.getObject(lang);
if (extension == null) {
extension = findExtension(lang);
- this.conf.setObject(lang, extension);
+ if (extension != null) {
+ this.conf.setObject(lang, extension);
+ }
}
return extension;
}
Modified: lucene/nutch/trunk/src/java/org/apache/nutch/indexer/Indexer.java
URL: http://svn.apache.org/viewcvs/lucene/nutch/trunk/src/java/org/apache/nutch/indexer/Indexer.java?rev=385322&r1=385321&r2=385322&view=diff
==============================================================================
--- lucene/nutch/trunk/src/java/org/apache/nutch/indexer/Indexer.java (original)
+++ lucene/nutch/trunk/src/java/org/apache/nutch/indexer/Indexer.java Sun Mar 12 09:46:07 2006
@@ -84,6 +84,7 @@
fs.delete(perm); // delete old, if any
+ final AnalyzerFactory factory = new AnalyzerFactory(job);
final IndexWriter writer = // build locally first
new IndexWriter(fs.startLocalOutput(perm, temp),
new NutchDocumentAnalyzer(job), true);
@@ -103,7 +104,12 @@
public void write(WritableComparable key, Writable value)
throws IOException { // unwrap & index doc
- writer.addDocument((Document)((ObjectWritable)value).get());
+ Document doc = (Document)((ObjectWritable)value).get();
+ NutchAnalyzer analyzer = factory.get(doc.get("lang"));
+ LOG.info(" Indexing [" + doc.getField("url").stringValue() + "]" +
+ " with analyzer " + analyzer +
+ " (" + doc.get("lang") + ")");
+ writer.addDocument(doc, analyzer);
}
public void close(final Reporter reporter) throws IOException {