- ensure tmpIndex is reset when the tmpSet is merged with the sparseSet

- only sort values in the encodedSet up to a specified index to prevent uninitialized values from polluting result - fix failing tests and add a new test of a single element
addthis · Jul 15, 2013 · e4466d8 · e4466d8
1 parent c5dd394
commit e4466d8
Show file tree

Hide file tree

Showing 2 changed files with 15 additions and 6 deletions.
diff --git a/src/main/java/com/clearspring/analytics/stream/cardinality/HyperLogLogPlus.java b/src/main/java/com/clearspring/analytics/stream/cardinality/HyperLogLogPlus.java
@@ -287,7 +287,6 @@ public boolean offer(Object o)
                 if (tmpIndex > sortThreshold)
                 {
                     mergeTempList();
-                    tmpIndex = 0;
                     if (sparseSet != null && sparseSet.length > sparseSetThreshold)
                     {
                         convertToNormal();
@@ -850,19 +849,21 @@ protected void mergeTempList()
         int[] retSet = sparseSet;
         if (tmpIndex > 0)
         {
-            tmpSet = sortEncodedSet(tmpSet);
+            tmpSet = sortEncodedSet(tmpSet, tmpIndex);
             retSet = merge(sparseSet, tmpSet);
             tmpSet = new int[sortThreshold+1];
+            tmpIndex = 0;
         }
         sparseSet = retSet == null ? new int[0] : retSet;
     }
 
     // exposed for testing
-    public int[] sortEncodedSet(int[] encodedSet)
+    public int[] sortEncodedSet(int[] encodedSet, int validIndex)
     {
         List<Integer> sortedList = new ArrayList<Integer>();
-        for (int k : encodedSet)
+        for (int i = 0; i < validIndex; i++)
         {
+            int k = encodedSet[i];
             sortedList.add(k);
         }
 

diff --git a/src/test/java/com/clearspring/analytics/stream/cardinality/TestHyperLogLogPlus.java b/src/test/java/com/clearspring/analytics/stream/cardinality/TestHyperLogLogPlus.java
@@ -182,7 +182,7 @@ public void testSortEncodedSet()
         testSet[1] = 655416;
         testSet[2] = 655425;
         HyperLogLogPlus hyperLogLogPlus = new HyperLogLogPlus(14, 25);
-        testSet = hyperLogLogPlus.sortEncodedSet(testSet);
+        testSet = hyperLogLogPlus.sortEncodedSet(testSet, 3);
         assertEquals(655403, testSet[0]);
         assertEquals(655425, testSet[1]);
         assertEquals(655416, testSet[2]);
@@ -222,6 +222,14 @@ public void testMergeSelf() throws CardinalityMergeException, IOException
 
     }
 
+    @Test
+    public void testOne() throws IOException
+    {
+        HyperLogLogPlus one = new HyperLogLogPlus(8,25);
+        one.offer("a");
+        assertEquals(1, one.cardinality());
+    }
+
     @Test
     public void testSparseSpace() throws IOException
     {
@@ -374,7 +382,7 @@ public void testLegacyCodec_sparse() throws IOException
 
         byte[] legacyBytes = baos.toByteArray();
 
-        // decode legacy
+        //  decode legacy
         HyperLogLogPlus decoded = HyperLogLogPlus.Builder.build(legacyBytes);
         assertEquals(baseline.cardinality(), decoded.cardinality());
         byte[] newBytes = baseline.getBytes();