svn commit: r1458949 - in /lucene/dev/branches/branch_4x: ./ lucene/ lucene/highlighter/ lucene/highlighter/src/java/org/apache/lucene/search/postingshighlight/ lucene/highlighter/src/test/org/apache/lucene/search/postingshighlight/

[mi...@apache.org]

Author: mikemccand
Date: Wed Mar 20 16:58:40 2013
New Revision: 1458949

URL: http://svn.apache.org/r1458949
Log:
LUCENE-4853: fix sort order bug with returned snippets

Modified:
    lucene/dev/branches/branch_4x/   (props changed)
    lucene/dev/branches/branch_4x/lucene/   (props changed)
    lucene/dev/branches/branch_4x/lucene/highlighter/   (props changed)
    lucene/dev/branches/branch_4x/lucene/highlighter/src/java/org/apache/lucene/search/postingshighlight/PostingsHighlighter.java
    lucene/dev/branches/branch_4x/lucene/highlighter/src/test/org/apache/lucene/search/postingshighlight/TestPostingsHighlighter.java

Modified: lucene/dev/branches/branch_4x/lucene/highlighter/src/java/org/apache/lucene/search/postingshighlight/PostingsHighlighter.java
URL: http://svn.apache.org/viewvc/lucene/dev/branches/branch_4x/lucene/highlighter/src/java/org/apache/lucene/search/postingshighlight/PostingsHighlighter.java?rev=1458949&r1=1458948&r2=1458949&view=diff
==============================================================================
--- lucene/dev/branches/branch_4x/lucene/highlighter/src/java/org/apache/lucene/search/postingshighlight/PostingsHighlighter.java (original)
+++ lucene/dev/branches/branch_4x/lucene/highlighter/src/java/org/apache/lucene/search/postingshighlight/PostingsHighlighter.java Wed Mar 20 16:58:40 2013
@@ -273,7 +273,7 @@ public class PostingsHighlighter {
    * @throws IllegalArgumentException if <code>field</code> was indexed without 
    *         {@link IndexOptions#DOCS_AND_FREQS_AND_POSITIONS_AND_OFFSETS}
    */
-  public Map<String,String[]> highlightFields(String fields[], Query query, IndexSearcher searcher, int[] docids, int maxPassages) throws IOException {
+  public Map<String,String[]> highlightFields(String fields[], Query query, IndexSearcher searcher, int[] docidsIn, int maxPassages) throws IOException {
     final IndexReader reader = searcher.getIndexReader();
     query = rewrite(query);
     SortedSet<Term> queryTerms = new TreeSet<Term>();
@@ -284,6 +284,10 @@ public class PostingsHighlighter {
 
     BreakIterator bi = (BreakIterator)breakIterator.clone();
 
+    // Make our own copy because we sort in-place:
+    int[] docids = new int[docidsIn.length];
+    System.arraycopy(docidsIn, 0, docids, 0, docidsIn.length);
+
     // sort for sequential io
     Arrays.sort(docids);
     Arrays.sort(fields);
@@ -302,8 +306,8 @@ public class PostingsHighlighter {
       Map<Integer,String> fieldHighlights = highlightField(field, contents[i], bi, terms, docids, leaves, maxPassages);
         
       String[] result = new String[docids.length];
-      for (int j = 0; j < docids.length; j++) {
-        result[j] = fieldHighlights.get(docids[j]);
+      for (int j = 0; j < docidsIn.length; j++) {
+        result[j] = fieldHighlights.get(docidsIn[j]);
       }
       highlights.put(field, result);
     }

Modified: lucene/dev/branches/branch_4x/lucene/highlighter/src/test/org/apache/lucene/search/postingshighlight/TestPostingsHighlighter.java
URL: http://svn.apache.org/viewvc/lucene/dev/branches/branch_4x/lucene/highlighter/src/test/org/apache/lucene/search/postingshighlight/TestPostingsHighlighter.java?rev=1458949&r1=1458948&r2=1458949&view=diff
==============================================================================
--- lucene/dev/branches/branch_4x/lucene/highlighter/src/test/org/apache/lucene/search/postingshighlight/TestPostingsHighlighter.java (original)
+++ lucene/dev/branches/branch_4x/lucene/highlighter/src/test/org/apache/lucene/search/postingshighlight/TestPostingsHighlighter.java Wed Mar 20 16:58:40 2013
@@ -748,4 +748,54 @@ public class TestPostingsHighlighter ext
     ir.close();
     dir.close();
   }
+
+  public void testMultipleDocs() throws Exception {
+    Directory dir = newDirectory();
+    IndexWriterConfig iwc = newIndexWriterConfig(TEST_VERSION_CURRENT, new MockAnalyzer(random()));
+    iwc.setMergePolicy(newLogMergePolicy());
+    RandomIndexWriter iw = new RandomIndexWriter(random(), dir, iwc);
+    
+    FieldType offsetsType = new FieldType(TextField.TYPE_STORED);
+    offsetsType.setIndexOptions(IndexOptions.DOCS_AND_FREQS_AND_POSITIONS_AND_OFFSETS);
+
+    int numDocs = atLeast(100);
+    for(int i=0;i<numDocs;i++) {
+      Document doc = new Document();
+      String content = "the answer is " + i;
+      if ((i & 1) == 0) {
+        content += " some more terms";
+      }
+      doc.add(new Field("body", content, offsetsType));
+      doc.add(newStringField("id", ""+i, Field.Store.YES));
+      iw.addDocument(doc);
+
+      if (random().nextInt(10) == 2) {
+        iw.commit();
+      }
+    }
+
+    IndexReader ir = iw.getReader();
+    iw.close();
+    
+    IndexSearcher searcher = newSearcher(ir);
+    PostingsHighlighter highlighter = new PostingsHighlighter();
+    Query query = new TermQuery(new Term("body", "answer"));
+    TopDocs hits = searcher.search(query, numDocs);
+    assertEquals(numDocs, hits.totalHits);
+
+    String snippets[] = highlighter.highlight("body", query, searcher, hits);
+    assertEquals(numDocs, snippets.length);
+    for(int hit=0;hit<numDocs;hit++) {
+      Document doc = searcher.doc(hits.scoreDocs[hit].doc);
+      int id = Integer.parseInt(doc.get("id"));
+      String expected = "the <b>answer</b> is " + id;
+      if ((id  & 1) == 0) {
+        expected += " some more terms";
+      }
+      assertEquals(expected, snippets[hit]);
+    }
+
+    ir.close();
+    dir.close();
+  }
 }