(hudi) branch master updated: fix: add handling for null schema edge case (#14020)

[sivabalan]

This is an automated email from the ASF dual-hosted git repository.

sivabalan pushed a commit to branch master
in repository https://gitbox.apache.org/repos/asf/hudi.git


The following commit(s) were added to refs/heads/master by this push:
     new 82a199373298 fix: add handling for null schema edge case (#14020)
82a199373298 is described below

commit 82a1993732982220b3505cc20bb1311fa7e4714f
Author: Tim Brown <t...@onehouse.ai>
AuthorDate: Wed Oct 1 00:49:59 2025 -0400

    fix: add handling for null schema edge case (#14020)
---
 .../java/org/apache/hudi/avro/HoodieAvroUtils.java     |  3 +++
 .../java/org/apache/hudi/avro/TestHoodieAvroUtils.java | 18 ++++++++++++++----
 2 files changed, 17 insertions(+), 4 deletions(-)

diff --git 
a/hudi-common/src/main/java/org/apache/hudi/avro/HoodieAvroUtils.java 
b/hudi-common/src/main/java/org/apache/hudi/avro/HoodieAvroUtils.java
index 61792e15f7b9..2dd40387c1d0 100644
--- a/hudi-common/src/main/java/org/apache/hudi/avro/HoodieAvroUtils.java
+++ b/hudi-common/src/main/java/org/apache/hudi/avro/HoodieAvroUtils.java
@@ -282,6 +282,9 @@ public class HoodieAvroUtils {
    * @param withOperationField Whether to include the '_hoodie_operation' field
    */
   public static Schema addMetadataFields(Schema schema, boolean 
withOperationField) {
+    if (isSchemaNull(schema)) {
+      return schema;
+    }
     int newFieldsSize = HoodieRecord.HOODIE_META_COLUMNS.size() + 
(withOperationField ? 1 : 0);
     List<Schema.Field> parentFields = new 
ArrayList<>(schema.getFields().size() + newFieldsSize);
 
diff --git 
a/hudi-common/src/test/java/org/apache/hudi/avro/TestHoodieAvroUtils.java 
b/hudi-common/src/test/java/org/apache/hudi/avro/TestHoodieAvroUtils.java
index 2371c5b2caed..b655f9bfdc8e 100644
--- a/hudi-common/src/test/java/org/apache/hudi/avro/TestHoodieAvroUtils.java
+++ b/hudi-common/src/test/java/org/apache/hudi/avro/TestHoodieAvroUtils.java
@@ -91,8 +91,6 @@ import org.junit.jupiter.params.ParameterizedTest;
 import org.junit.jupiter.params.provider.Arguments;
 import org.junit.jupiter.params.provider.MethodSource;
 import org.junit.jupiter.params.provider.ValueSource;
-import org.slf4j.Logger;
-import org.slf4j.LoggerFactory;
 
 import java.io.ByteArrayOutputStream;
 import java.io.IOException;
@@ -134,8 +132,6 @@ import static org.junit.jupiter.api.Assertions.assertTrue;
  */
 public class TestHoodieAvroUtils {
 
-  private static final Logger LOG = 
LoggerFactory.getLogger(TestHoodieAvroUtils.class);
-
   private static final String EVOLVED_SCHEMA = "{\"type\": 
\"record\",\"name\": \"testrec1\",\"fields\": [ "
       + "{\"name\": \"timestamp\",\"type\": \"double\"},{\"name\": 
\"_row_key\", \"type\": \"string\"},"
       + "{\"name\": \"non_pii_col\", \"type\": \"string\"},"
@@ -269,6 +265,20 @@ public class TestHoodieAvroUtils {
 
   public static Schema SCHEMA_WITH_NESTED_FIELD_LARGE = new 
Schema.Parser().parse(SCHEMA_WITH_NESTED_FIELD_LARGE_STR);
 
+  @Test
+  void testAddMetaFields() {
+    // validate null schema does not throw errors
+    assertNull(HoodieAvroUtils.addMetadataFields(null));
+    Schema nullSchema = Schema.create(Schema.Type.NULL);
+    assertEquals(nullSchema, HoodieAvroUtils.addMetadataFields(nullSchema));
+    // test with non-null schema
+    Schema schema = HoodieAvroUtils.addMetadataFields(new 
Schema.Parser().parse(EXAMPLE_SCHEMA));
+    assertEquals(NUM_FIELDS_IN_EXAMPLE_SCHEMA + 
HoodieRecord.HOODIE_META_COLUMNS.size(), schema.getFields().size());
+    for (String metaCol : HoodieRecord.HOODIE_META_COLUMNS) {
+      assertNotNull(schema.getField(metaCol));
+    }
+  }
+
   @Test
   public void testPropsPresent() {
     Schema schema = HoodieAvroUtils.addMetadataFields(new 
Schema.Parser().parse(EXAMPLE_SCHEMA));