AVRO-3789: [java] fix map comparison in GenericData (#2318)

* AVRO-3789: add more tests for GenericData map comparison * AVRO-3789: fix GenericData map comparison for empty maps and different-sized maps * AVRO-3789: use isEmpty to check for empty maps
apache · Jul 12, 2023 · a88cc9e · a88cc9e
1 parent 8c3ee16
commit a88cc9e
Show file tree

Hide file tree

Showing 2 changed files with 98 additions and 1 deletion.
diff --git a/lang/java/avro/src/main/java/org/apache/avro/generic/GenericData.java b/lang/java/avro/src/main/java/org/apache/avro/generic/GenericData.java
@@ -1178,7 +1178,11 @@ protected int compareMaps(final Map<?, ?> m1, final Map<?, ?> m2) {
       return 0;
     }
 
-    if (m2.size() != m2.size()) {
+    if (m1.isEmpty() && m2.isEmpty()) {
+      return 0;
+    }
+
+    if (m1.size() != m2.size()) {
       return 1;
     }
 

diff --git a/lang/java/avro/src/test/java/org/apache/avro/generic/TestGenericData.java b/lang/java/avro/src/test/java/org/apache/avro/generic/TestGenericData.java
@@ -196,6 +196,99 @@ public void testMapValuesEqualsStringAndUtf8Compatibility() {
     assertEquals(r1, r0);
   }
 
+  @Test
+  public void testEqualsEmptyMaps() {
+    Field myMapField = new Field("my_map", Schema.createMap(Schema.create(Schema.Type.STRING)), null, null);
+    Schema schema = Schema.createRecord("my_record", "doc", "mytest", false);
+    schema.setFields(Arrays.asList(myMapField));
+
+    GenericRecord r0 = new GenericData.Record(schema);
+    r0.put("my_map", new HashMap<>());
+    GenericRecord r1 = new GenericData.Record(schema);
+    r1.put("my_map", new HashMap<>());
+
+    assertEquals(r0, r1);
+    assertEquals(r1, r0);
+  }
+
+  @Test
+  public void testEqualsEmptyMapAndNonEmptyMap() {
+    Field myMapField = new Field("my_map", Schema.createMap(Schema.create(Schema.Type.STRING)), null, null);
+    Schema schema = Schema.createRecord("my_record", "doc", "mytest", false);
+    schema.setFields(Arrays.asList(myMapField));
+
+    GenericRecord r0 = new GenericData.Record(schema);
+    r0.put("my_map", new HashMap<>());
+    GenericRecord r1 = new GenericData.Record(schema);
+    HashMap<CharSequence, CharSequence> pair1 = new HashMap<>();
+    pair1.put("keyOne", "valueOne");
+    r1.put("my_map", pair1);
+
+    assertNotEquals(r0, r1);
+    assertNotEquals(r1, r0);
+  }
+
+  @Test
+  public void testEqualsMapAndSubset() {
+    Field myMapField = new Field("my_map", Schema.createMap(Schema.create(Schema.Type.STRING)), null, null);
+    Schema schema = Schema.createRecord("my_record", "doc", "mytest", false);
+    schema.setFields(Arrays.asList(myMapField));
+
+    GenericRecord r0 = new GenericData.Record(schema);
+    HashMap<CharSequence, String> m1 = new HashMap<>();
+    m1.put("keyOne", "valueOne");
+    m1.put("keyTwo", "valueTwo");
+    r0.put("my_map", m1);
+
+    GenericRecord r1 = new GenericData.Record(schema);
+    HashMap<CharSequence, String> m2 = new HashMap<>();
+    m2.put("keyOne", "valueOne");
+    r1.put("my_map", m2);
+
+    assertNotEquals(r0, r1);
+    assertNotEquals(r1, r0);
+  }
+
+  @Test
+  public void testEqualsMapAndSameSizeMapWithDifferentKeys() {
+    Field myMapField = new Field("my_map", Schema.createMap(Schema.create(Schema.Type.STRING)), null, null);
+    Schema schema = Schema.createRecord("my_record", "doc", "mytest", false);
+    schema.setFields(Arrays.asList(myMapField));
+
+    GenericRecord r0 = new GenericData.Record(schema);
+    HashMap<CharSequence, String> m1 = new HashMap<>();
+    m1.put("keyOne", "valueOne");
+    r0.put("my_map", m1);
+
+    GenericRecord r1 = new GenericData.Record(schema);
+    HashMap<CharSequence, String> m2 = new HashMap<>();
+    m2.put("keyTwo", "valueTwo");
+    r1.put("my_map", m2);
+
+    assertNotEquals(r0, r1);
+    assertNotEquals(r1, r0);
+  }
+
+  @Test
+  public void testEqualsMapAndSameSizeMapWithDifferentValues() {
+    Field myMapField = new Field("my_map", Schema.createMap(Schema.create(Schema.Type.STRING)), null, null);
+    Schema schema = Schema.createRecord("my_record", "doc", "mytest", false);
+    schema.setFields(Arrays.asList(myMapField));
+
+    GenericRecord r0 = new GenericData.Record(schema);
+    HashMap<CharSequence, String> m1 = new HashMap<>();
+    m1.put("keyOne", "valueOne");
+    r0.put("my_map", m1);
+
+    GenericRecord r1 = new GenericData.Record(schema);
+    HashMap<CharSequence, String> m2 = new HashMap<>();
+    m2.put("keyOne", "valueTwo");
+    r1.put("my_map", m2);
+
+    assertNotEquals(r0, r1);
+    assertNotEquals(r1, r0);
+  }
+
   @Test
   public void testArrayValuesEqualsStringAndUtf8Compatibility() {
     Field myArrayField = new Field("my_array", Schema.createArray(Schema.create(Schema.Type.STRING)), null, null);