Use original order by default for same-score items rather than sortin…

…g by docId
apache · Dec 22, 2024 · 2e4dec9 · 2e4dec9
1 parent 8c7050b
commit 2e4dec9
Show file tree

Hide file tree

Showing 3 changed files with 24 additions and 5 deletions.
diff --git a/lucene/core/src/java/org/apache/lucene/search/QueryRescorer.java b/lucene/core/src/java/org/apache/lucene/search/QueryRescorer.java
@@ -50,6 +50,12 @@ public TopDocs rescore(IndexSearcher searcher, TopDocs firstPassTopDocs, int top
       throws IOException {
     ScoreDoc[] hits = firstPassTopDocs.scoreDocs.clone();
 
+    // record original index of each hit in its originalIndex
+    // NB code below relies on hits being in docId order so the order is about to be lost
+    for (int i = 0; i < hits.length; i++) {
+      hits[i].originalIndex = i;
+    }
+
     Arrays.sort(hits, (a, b) -> a.doc - b.doc);
 
     List<LeafReaderContext> leaves = searcher.getIndexReader().leaves();
@@ -110,6 +116,10 @@ public TopDocs rescore(IndexSearcher searcher, TopDocs firstPassTopDocs, int top
             return -1;
           } else if (a.score < b.score) {
             return 1;
+          } else if (a.originalIndex > b.originalIndex) {
+            return 1;
+          } else if (a.originalIndex < b.originalIndex) {
+            return -1;
           } else {
             // This subtraction can't overflow int
             // because docIDs are >= 0:

diff --git a/lucene/core/src/java/org/apache/lucene/search/ScoreDoc.java b/lucene/core/src/java/org/apache/lucene/search/ScoreDoc.java
@@ -34,6 +34,9 @@ public class ScoreDoc {
   /** Only set by {@link TopDocs#merge} */
   public int shardIndex;
 
+  /** Only set by {@link QueryRescorer#rescore} */
+  public int originalIndex;
+
   /** Constructs a ScoreDoc. */
   public ScoreDoc(int doc, float score) {
     this(doc, score, -1);
@@ -49,6 +52,13 @@ public ScoreDoc(int doc, float score, int shardIndex) {
   // A convenience method for debugging.
   @Override
   public String toString() {
-    return "doc=" + doc + " score=" + score + " shardIndex=" + shardIndex;
+    return "doc="
+        + doc
+        + " score="
+        + score
+        + " shardIndex="
+        + shardIndex
+        + " originalIndex="
+        + originalIndex;
   }
 }
diff --git a/lucene/core/src/test/org/apache/lucene/search/TestQueryRescorer.java b/lucene/core/src/test/org/apache/lucene/search/TestQueryRescorer.java
@@ -453,8 +453,8 @@ public int compare(Integer a, Integer b) {
               } else if (bv < av) {
                 return reverseInt;
               } else {
-                // Tie break by docID, ascending
-                return a - b;
+                // use original order if scores are the same
+                return 0;
               }
             } catch (IOException ioe) {
               throw new RuntimeException(ioe);
@@ -464,8 +464,7 @@ public int compare(Integer a, Integer b) {
 
     boolean fail = false;
     for (int i = 0; i < numHits; i++) {
-      // System.out.println("expected=" + expected[i] + " vs " + hits2.scoreDocs[i].doc + " v=" +
-      // idToNum[Integer.parseInt(r.storedFields().document(expected[i]).get("id"))]);
+      // System.out.println("expected=" + expected[i] + " vs " + hits2.scoreDocs[i].toString());
       if (expected[i].intValue() != hits2.scoreDocs[i].doc) {
         // System.out.println("  diff!");
         fail = true;