Changeset 24725 for main/trunk/greenstone2/common-src/indexers/lucene-gs/src/org/greenstone/LuceneWrapper/GS2LuceneIndexer.java

Timestamp:

2011-10-05T15:44:19+13:00 (13 years ago)

Author:

davidb

Message:

Restruturing of Lucene version 2.x and 3.x to make it easier to control which one is used

File:

• main/trunk/greenstone2/common-src/indexers/lucene-gs/src/org/greenstone/LuceneWrapper/GS2LuceneIndexer.java (modified) (7 diffs)

main/trunk/greenstone2/common-src/indexers/lucene-gs/src/org/greenstone/LuceneWrapper/GS2LuceneIndexer.java

@@ -45 +45 @@
 import org.apache.lucene.index.Term;
 import org.apache.lucene.analysis.Analyzer;
+
+import org.apache.lucene.store.SimpleFSDirectory;
+import org.apache.lucene.index.IndexWriter.MaxFieldLength;
 
 import java.util.Stack;
@@ -190 +193 @@
     protected String file_id_ = null;
 
-    static private String[] stop_words = GS2Analyzer.STOP_WORDS;
-
-
     /** pass in true if want to create a new index, false if want to use the existing one */
     public Indexer (String doc_tag_level, File index_dir, boolean create) 
@@ -206 +206 @@
         reader.setFeature("http://xml.org/sax/features/validation", false);
 
-        analyzer_ = new GS2Analyzer(stop_words);
-
-        writer_ = new IndexWriter(index_dir.getPath(), analyzer_, create);
+        SimpleFSDirectory index_dir_dir = new SimpleFSDirectory(new File(index_dir.getPath()));
+
+        analyzer_ = new GS2Analyzer(); // uses build in stop_word_set
+
+        writer_ = new IndexWriter(index_dir_dir, analyzer_, create, MaxFieldLength.UNLIMITED);
+                       
         // by default, will only index 10,000 words per document
         // Can throw out_of_memory errors
@@ -318 +321 @@
         //String node_id = atts.getValue("gs2:id");
         //print(" " + qName + ": " + node_id + " (" + mode_ + ")" );
-        //current_doc_.add(new Field("nodeID", node_id, Field.Store.YES, Field.Index.UN_TOKENIZED));
+        //current_doc_.add(new Field("nodeID", node_id, Field.Store.YES, Field.Index.NOT_ANALYZED));
         
         current_doc_oid_ = atts.getValue("gs2:docOID");
         print(" " + qName + ": " + current_doc_oid_ + " (" + mode_ + ")" );
-        current_doc_.add(new Field("docOID", current_doc_oid_, Field.Store.YES, Field.Index.UN_TOKENIZED));
+        current_doc_.add(new Field("docOID", current_doc_oid_, Field.Store.YES, Field.Index.NOT_ANALYZED));
         }
         
@@ -359 +362 @@
         if (qName.equals(indexable_current_node_))
             {
-            current_doc_.add(new Field(qName, current_contents_, Field.Store.NO, Field.Index.TOKENIZED, Field.TermVector.YES));
+            current_doc_.add(new Field(qName, current_contents_, Field.Store.NO, Field.Index.ANALYZED, Field.TermVector.YES));
             // The byXX fields are used for sorting search results
             // We don't want to do that for Text or AllFields fields
@@ -365 +368 @@
             if (!qName.equals("TX") && !qName.equals("ZZ"))
                 {
-                current_doc_.add(new Field("by" + qName, current_contents_, Field.Store.NO, Field.Index.UN_TOKENIZED, Field.TermVector.NO));
+                current_doc_.add(new Field("by" + qName, current_contents_, Field.Store.NO, Field.Index.NOT_ANALYZED, Field.TermVector.NO));
                 }
             
@@ -472 +475 @@
     {
         debug("GS2LuceneDelete.deleteDocument(" + doc_id + ")");
-        debug("- Initial number of documents in index: " + writer_.docCount());
+        debug("- Initial number of documents in index: " + writer_.numDocs());
         writer_.deleteDocuments(new Term("docOID", doc_id));
-        debug("- Final number of documents in index: " + writer_.docCount());
+        debug("- Final number of documents in index: " + writer_.numDocs());
     }