Changeset 12408 for trunk/indexers/lucene-gs

Timestamp:

2006-08-07T15:24:47+12:00 (18 years ago)

Author:

mdewsnip

Message:

Added a "-filter" option which can currently be used for specifying range filters (eg. we're going to use it for dates). Many thanks to Me and DL Consulting Ltd.

File:

• trunk/indexers/lucene-gs/src/org/greenstone/LuceneWrapper/GS2LuceneQuery.java (modified) (6 diffs)

trunk/indexers/lucene-gs/src/org/greenstone/LuceneWrapper/GS2LuceneQuery.java

@@ -24 +24 @@
 import org.apache.lucene.index.TermFreqVector;
 import org.apache.lucene.queryParser.QueryParser;
+import org.apache.lucene.search.Filter;
 import org.apache.lucene.search.Hit;
 import org.apache.lucene.search.Hits;
 import org.apache.lucene.search.IndexSearcher;
 import org.apache.lucene.search.Query;
+import org.apache.lucene.search.RangeFilter;
 import org.apache.lucene.search.Searcher;
 import org.apache.lucene.search.Sort;
@@ -46 +48 @@
 
         Sort sorter = new Sort();
+        Filter filter = null;
             boolean fuzzy = false;
 
@@ -58 +61 @@
                             ///ystem.err.println("**** sort by = " + args[i]);
                             sorter = new Sort(args[i]);
+                        }
+                    if (args[i].equals("-filter"))
+                        {
+                            i++;
+                filter = parseFilterString(args[i]);
                         }
                     if (args[i].equals("-dco"))
@@ -211 +219 @@
 
         // Perform the query
-        Hits hits = searcher.search(query, sorter);
+        Hits hits;
+        if (filter != null) {
+            hits = searcher.search(query, filter, sorter);
+        }
+        else {
+            hits = searcher.search(query, sorter);
+        }
         System.out.println("<ResultSet>");
         System.out.println("  <QueryString>" + query_string + "</QueryString>");
@@ -244 +258 @@
                         // TX field
                         TermFreqVector term_freq_vector = reader.getTermFreqVector(hit.getId(), "TX");
-                        if (term_freq_vector.size() > 0)
+                        if (term_freq_vector != null && term_freq_vector.size() > 0)
                             {
                                 int[] term_frequencies = term_freq_vector.getTermFrequencies();
@@ -316 +330 @@
         }
     }
+
+
+    private static Filter parseFilterString(String filter_string)
+    {
+    // Range filters
+    if (filter_string.matches("(.*):[\\{\\[](.+) TO (.+)[\\}\\]]")) {
+        String field_name = filter_string.substring(0, filter_string.indexOf(":"));
+        boolean include_lower = (filter_string.charAt(filter_string.indexOf(":") + 1) == '[');
+        String lower_term = filter_string.substring(filter_string.indexOf(":") + 2, filter_string.indexOf(" TO "));
+        String upper_term = filter_string.substring(filter_string.indexOf(" TO ") + " TO ".length(), filter_string.length() - 1);
+        boolean include_upper = (filter_string.charAt(filter_string.length() - 1) == ']');
+        return new RangeFilter(field_name, lower_term, upper_term, include_lower, include_upper);
+    }
+
+    System.err.println("Error: Could not understand filter string \"" + filter_string + "\"");
+    return null;
+    }
 }