[MINOR] Fix NPE while parsing partition values in file index (#12177)

hudi-spark-datasource/hudi-spark-common/src/main/scala/org/apache/hudi/SparkHoodieTableFileIndex.scala

@@ -373,10 +373,14 @@ class SparkHoodieTableFileIndex(spark: SparkSession,
       staticPartitionColumnValues.map(_._1): _*)
   }
 
-  protected def doParsePartitionColumnValues(partitionColumns: Array[String], partitionPath: String): Array[Object] = {
+  /**
+   * @VisibleForTesting
+   */
+  def doParsePartitionColumnValues(partitionColumns: Array[String], partitionPath: String): Array[Object] = {
     val tableConfig = metaClient.getTableConfig
     if (null != tableConfig.getKeyGeneratorClassName
       && tableConfig.getKeyGeneratorClassName.equals(KeyGeneratorType.TIMESTAMP.getClassName)
+      && null != tableConfig.propsMap.get(TimestampKeyGeneratorConfig.TIMESTAMP_TYPE_FIELD.key())
       && tableConfig.propsMap.get(TimestampKeyGeneratorConfig.TIMESTAMP_TYPE_FIELD.key())
       .matches("SCALAR|UNIX_TIMESTAMP|EPOCHMILLISECONDS|EPOCHMICROSECONDS")) {
       // For TIMESTAMP key generator when TYPE is SCALAR, UNIX_TIMESTAMP,

hudi-spark-datasource/hudi-spark/src/test/scala/org/apache/hudi/TestHoodieFileIndex.scala

@@ -49,6 +49,7 @@ import org.apache.spark.sql.execution.datasources.{NoopCache, PartitionDirectory
 import org.apache.spark.sql.functions.{lit, struct}
 import org.apache.spark.sql.hudi.HoodieSparkSessionExtension
 import org.apache.spark.sql.types._
+import org.apache.spark.unsafe.types.UTF8String
 import org.junit.jupiter.api.Assertions.{assertEquals, assertTrue}
 import org.junit.jupiter.api.{BeforeEach, Test}
 import org.junit.jupiter.params.ParameterizedTest
@@ -110,6 +111,36 @@ class TestHoodieFileIndex extends HoodieSparkClientTestBase with ScalaAssertionS
     assertEquals("partition", fileIndex.partitionSchema.fields.map(_.name).mkString(","))
   }
 
+  /**
+   * Unit test for `doParsePartitionColumnValues` method in `HoodieFileIndex`.
+   *
+   * This test verifies that the `doParsePartitionColumnValues` method correctly returns
+   * partition values when the `propsMap` in the table configuration does not contain the
+   * expected timestamp configuration key, simulating a `null` scenario. Specifically,
+   * this test validates the behavior for the `TIMESTAMP` key generator type, ensuring
+   * that the partition path string is passed as `UTF8String` in the result array.
+   */
+  @Test
+  def testParsePartitionValues(): Unit = {
+    // Set up table configuration and schema to use TIMESTAMP key generator
+    val tableConfig = metaClient.getTableConfig
+    tableConfig.setValue(HoodieTableConfig.KEY_GENERATOR_TYPE, KeyGeneratorType.TIMESTAMP.name())
+    tableConfig.setValue(HoodieTableConfig.PARTITION_FIELDS, "col1")
+    // Define schema with one partition column (col1)
+    val fields = List(
+      StructField.apply("f1", DataTypes.DoubleType, nullable = true),
+      StructField.apply("col1", DataTypes.LongType, nullable = true))
+    val schema = StructType.apply(fields)
+    // Set partition column and partition path for testing
+    val partitionColumns = Array("col1")
+    val partitionPath = "2023/10/28"
+    val fileIndex = HoodieFileIndex(spark, metaClient, Some(schema), queryOpts)
+    // Create file index and validate the result
+    val result = fileIndex.doParsePartitionColumnValues(partitionColumns, partitionPath)
+    assertEquals(1, result.length)
+    assertEquals(UTF8String.fromString(partitionPath), result(0))
+  }
+
   @ParameterizedTest
   @MethodSource(Array("keyGeneratorParameters"))
   def testPartitionSchemaForBuiltInKeyGenerator(keyGenerator: String): Unit = {