AVRO-1695. Java: Fix GenericData#deepCopy() to support logical types.

CHANGES.txt

@@ -15,6 +15,9 @@ Trunk (not yet released)
     AVRO-1889: Update maven-shade-plugin to enable building in Java 8 on Mac.
     (Sachin Goyal via blue)
 
+    AVRO-1695. Java: Fix GenericData#deepCopy() to support logical types.
+    (cutting)
+
   BUG FIXES
 
     AVRO-1741: Python3: Fix error when codec is not in the header.

lang/java/avro/src/main/java/org/apache/avro/generic/GenericData.java

@@ -34,6 +34,7 @@
 import org.apache.avro.AvroRuntimeException;
 import org.apache.avro.AvroTypeException;
 import org.apache.avro.Conversion;
+import org.apache.avro.Conversions;
 import org.apache.avro.LogicalType;
 import org.apache.avro.Schema;
 import org.apache.avro.Schema.Field;
@@ -1018,15 +1019,31 @@ public Object getDefaultValue(Field field) {
 
   /**
    * Makes a deep copy of a value given its schema.
+   * <P>Logical types are converted to raw types, copied, then converted back.
    * @param schema the schema of the value to deep copy.
    * @param value the value to deep copy.
    * @return a deep copy of the given value.
    */
   @SuppressWarnings({ "rawtypes", "unchecked" })
   public <T> T deepCopy(Schema schema, T value) {
+    if (value == null) return null;
+    LogicalType logicalType = schema.getLogicalType();
+    if (logicalType == null)           // not a logical type -- use raw copy
+      return (T)deepCopyRaw(schema, value);
+    Conversion conversion = getConversionByClass(value.getClass());
+    if (conversion == null)            // no conversion defined -- try raw copy
+      return (T)deepCopyRaw(schema, value);
+    // logical type with conversion: convert to raw, copy, then convert back to logical
+    Object raw = Conversions.convertToRawType(value, schema, logicalType, conversion);
+    Object copy = deepCopyRaw(schema, raw);       // copy raw
+    return (T)Conversions.convertToLogicalType(copy, schema, logicalType, conversion);
+  }
+
+  private Object deepCopyRaw(Schema schema, Object value) {
     if (value == null) {
       return null;
     }
+
     switch (schema.getType()) {
       case ARRAY:
         List<Object> arrayValue = (List) value;
@@ -1035,7 +1052,7 @@ public <T> T deepCopy(Schema schema, T value) {
         for (Object obj : arrayValue) {
           arrayCopy.add(deepCopy(schema.getElementType(), obj));
         }
-        return (T)arrayCopy;
+        return arrayCopy;
       case BOOLEAN:
         return value; // immutable
       case BYTES:
@@ -1045,13 +1062,13 @@ public <T> T deepCopy(Schema schema, T value) {
         byte[] bytesCopy = new byte[length];
         byteBufferValue.get(bytesCopy, 0, length);
         byteBufferValue.position(start);
-        return (T)ByteBuffer.wrap(bytesCopy, 0, length);
+        return ByteBuffer.wrap(bytesCopy, 0, length);
       case DOUBLE:
         return value; // immutable
       case ENUM:
-        return (T)createEnum(value.toString(), schema);
+        return createEnum(value.toString(), schema);
       case FIXED:
-        return (T)createFixed(null, ((GenericFixed) value).bytes(), schema);
+        return createFixed(null, ((GenericFixed) value).bytes(), schema);
       case FLOAT:
         return value; // immutable
       case INT:
@@ -1066,7 +1083,7 @@ public <T> T deepCopy(Schema schema, T value) {
           mapCopy.put((CharSequence)(deepCopy(STRINGS, entry.getKey())),
               deepCopy(schema.getValueType(), entry.getValue()));
         }
-        return (T)mapCopy;
+        return mapCopy;
       case NULL:
         return null;
       case RECORD:
@@ -1080,21 +1097,21 @@ public <T> T deepCopy(Schema schema, T value) {
                                      getField(value, name, pos, oldState));
           setField(newRecord, name, pos, newValue, newState);
         }
-        return (T)newRecord;
+        return newRecord;
       case STRING:
         // Strings are immutable
         if (value instanceof String) {
-          return (T)value;
+          return value;
         }
 
         // Some CharSequence subclasses are mutable, so we still need to make
         // a copy
         else if (value instanceof Utf8) {
           // Utf8 copy constructor is more efficient than converting
           // to string and then back to Utf8
-          return (T)new Utf8((Utf8)value);
+          return new Utf8((Utf8)value);
         }
-        return (T)new Utf8(value.toString());
+        return new Utf8(value.toString());
       case UNION:
         return deepCopy(
             schema.getTypes().get(resolveUnion(schema, value)), value);

lang/java/avro/src/test/java/org/apache/avro/generic/TestGenericLogicalTypes.java

@@ -232,4 +232,65 @@ private <D> File write(GenericData model, Schema schema, D... data) throws IOExc
 
     return file;
   }
+
+  @Test
+  public void testCopyUuid() {
+    testCopy(LogicalTypes.uuid().addToSchema(Schema.create(Schema.Type.STRING)),
+             UUID.randomUUID(),
+             GENERIC);
+  }
+
+  @Test
+  public void testCopyUuidRaw() {
+    testCopy(LogicalTypes.uuid().addToSchema(Schema.create(Schema.Type.STRING)),
+             UUID.randomUUID().toString(),        // use raw type
+             GenericData.get());                  // with no conversions
+  }
+
+  @Test
+  public void testCopyDecimal() {
+    testCopy(LogicalTypes.decimal(9, 2).addToSchema(Schema.create(Schema.Type.BYTES)),
+             new BigDecimal("-34.34"),
+             GENERIC);
+  }
+
+  @Test
+  public void testCopyDecimalRaw() {
+    testCopy(LogicalTypes.decimal(9, 2).addToSchema(Schema.create(Schema.Type.BYTES)),
+             ByteBuffer.wrap(new BigDecimal("-34.34").unscaledValue().toByteArray()),
+             GenericData.get());                  // no conversions
+  }
+
+  private void testCopy(Schema schema, Object value, GenericData model) {
+    // test direct copy of instance
+    checkCopy(value, model.deepCopy(schema, value), false);
+
+    // test nested in a record
+    Schema recordSchema = Schema.createRecord("X", "", "test", false);
+    List<Schema.Field> fields = new ArrayList<Schema.Field>();
+    fields.add(new Schema.Field("x", schema, "", null));
+    recordSchema.setFields(fields);
+
+    GenericRecordBuilder builder = new GenericRecordBuilder(recordSchema);
+    builder.set("x", value);
+    GenericData.Record record = builder.build();
+    checkCopy(record, model.deepCopy(recordSchema, record), true);
+
+    // test nested in array
+    Schema arraySchema = Schema.createArray(schema);
+    ArrayList array = new ArrayList(Arrays.asList(value));
+    checkCopy(array, model.deepCopy(arraySchema, array), true);
+
+    // test record nested in array
+    Schema recordArraySchema = Schema.createArray(recordSchema);
+    ArrayList recordArray = new ArrayList(Arrays.asList(record));
+    checkCopy(recordArray, model.deepCopy(recordArraySchema, recordArray), true);
+  }
+
+  private void checkCopy(Object original, Object copy, boolean notSame) {
+    if (notSame) 
+      Assert.assertNotSame(original, copy);
+    Assert.assertEquals(original, copy);
+  }
+
 }