the-other-tim-brown commented on code in PR #17629: URL: https://github.com/apache/hudi/pull/17629#discussion_r2638157209
########## hudi-spark-datasource/hudi-spark/src/test/java/org/apache/hudi/io/storage/TestHoodieSparkLanceWriter.java: ########## @@ -0,0 +1,503 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one + * or more contributor license agreements. See the NOTICE file + * distributed with this work for additional information + * regarding copyright ownership. The ASF licenses this file + * to you under the Apache License, Version 2.0 (the + * "License"); you may not use this file except in compliance + * with the License. You may obtain a copy of the License at + * + * http://www.apache.org/licenses/LICENSE-2.0 + * + * Unless required by applicable law or agreed to in writing, software + * distributed under the License is distributed on an "AS IS" BASIS, + * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. + * See the License for the specific language governing permissions and + * limitations under the License. + */ + +package org.apache.hudi.io.storage; + +import com.lancedb.lance.file.LanceFileReader; +import org.apache.arrow.memory.BufferAllocator; +import org.apache.arrow.memory.RootAllocator; +import org.apache.arrow.vector.BigIntVector; +import org.apache.arrow.vector.BitVector; +import org.apache.arrow.vector.Float4Vector; +import org.apache.arrow.vector.Float8Vector; +import org.apache.arrow.vector.IntVector; +import org.apache.arrow.vector.VarBinaryVector; +import org.apache.arrow.vector.VarCharVector; +import org.apache.arrow.vector.VectorSchemaRoot; +import org.apache.arrow.vector.ipc.ArrowReader; +import org.apache.hudi.client.SparkTaskContextSupplier; +import org.apache.hudi.common.model.HoodieKey; +import org.apache.hudi.common.testutils.HoodieTestUtils; +import org.apache.hudi.storage.HoodieStorage; +import org.apache.hudi.storage.StoragePath; +import org.apache.spark.sql.catalyst.InternalRow; +import org.apache.spark.sql.catalyst.expressions.GenericInternalRow; +import org.apache.spark.sql.types.DataTypes; +import org.apache.spark.sql.types.StructType; +import org.apache.spark.unsafe.types.UTF8String; +import org.junit.jupiter.api.AfterEach; +import org.junit.jupiter.api.BeforeEach; +import org.junit.jupiter.api.Test; +import org.junit.jupiter.api.condition.DisabledIfSystemProperty; +import org.junit.jupiter.api.io.TempDir; + +import java.io.File; +import java.io.IOException; +import java.util.ArrayList; +import java.util.List; + +import static org.apache.hudi.common.model.HoodieRecord.HoodieMetadataField.COMMIT_SEQNO_METADATA_FIELD; +import static org.apache.hudi.common.model.HoodieRecord.HoodieMetadataField.COMMIT_TIME_METADATA_FIELD; +import static org.apache.hudi.common.model.HoodieRecord.HoodieMetadataField.FILENAME_METADATA_FIELD; +import static org.apache.hudi.common.model.HoodieRecord.HoodieMetadataField.PARTITION_PATH_METADATA_FIELD; +import static org.apache.hudi.common.model.HoodieRecord.HoodieMetadataField.RECORD_KEY_METADATA_FIELD; +import static org.junit.jupiter.api.Assertions.assertEquals; +import static org.junit.jupiter.api.Assertions.assertFalse; +import static org.junit.jupiter.api.Assertions.assertNotNull; +import static org.junit.jupiter.api.Assertions.assertTrue; + +/** + * Tests for {@link HoodieSparkLanceWriter}. + */ +@DisabledIfSystemProperty(named = "lance.skip.tests", matches = "true") +public class TestHoodieSparkLanceWriter { + + @TempDir + File tempDir; + + private HoodieStorage storage; + private SparkTaskContextSupplier taskContextSupplier; + private String instantTime; + + @BeforeEach + public void setUp() throws IOException { + storage = HoodieTestUtils.getStorage(tempDir.getAbsolutePath()); + taskContextSupplier = new SparkTaskContextSupplier(); + instantTime = "20251201120000000"; + } + + @AfterEach + public void tearDown() throws Exception { + if (storage != null) { + storage.close(); + } + } + + @Test + public void testWriteRowWithMetadataPopulation() throws Exception { + // Schema WITH meta fields (writer expects this when populateMetaFields=true) + StructType schema = createSchemaWithMetaFields(); + + StoragePath path = new StoragePath(tempDir.getAbsolutePath() + "/test_with_metadata.lance"); + HoodieSparkLanceWriter writer = new HoodieSparkLanceWriter( + path, schema, instantTime, taskContextSupplier, storage, true); + + // Create row with PLACEHOLDER meta fields + user data + InternalRow row = createRowWithMetaFields(1, "Alice", 30L); + HoodieKey key = new HoodieKey("key1", "partition1"); + + writer.writeRowWithMetadata(key, row); + writer.close(); + + // Verify using LanceFileReader + assertTrue(storage.exists(path), "Lance file should exist"); + + try (BufferAllocator allocator = new RootAllocator(); + LanceFileReader reader = LanceFileReader.open(path.toString(), allocator)) { + + assertEquals(1, reader.numRows(), "Should have 1 record"); + + // Schema should have 5 meta fields + 3 user fields = 8 fields + assertEquals(8, reader.schema().getFields().size(), "Should have 8 fields"); + + // Read and verify data + ArrowReader arrowReader = reader.readAll(null, null, Integer.MAX_VALUE); + assertTrue(arrowReader.loadNextBatch(), "Should load batch"); + VectorSchemaRoot root = arrowReader.getVectorSchemaRoot(); + + // Verify meta fields were populated + VarCharVector commitTimeVector = (VarCharVector) root.getVector(COMMIT_TIME_METADATA_FIELD.getFieldName()); + assertNotNull(commitTimeVector); + assertEquals(instantTime, new String(commitTimeVector.get(0)), "Commit time should match"); + + VarCharVector recordKeyVector = (VarCharVector) root.getVector(RECORD_KEY_METADATA_FIELD.getFieldName()); + assertEquals("key1", new String(recordKeyVector.get(0)), "Record key should match"); + + VarCharVector partitionPathVector = (VarCharVector) root.getVector(PARTITION_PATH_METADATA_FIELD.getFieldName()); + assertEquals("partition1", new String(partitionPathVector.get(0)), "Partition path should match"); + + VarCharVector fileNameVector = (VarCharVector) root.getVector(FILENAME_METADATA_FIELD.getFieldName()); + assertEquals(path.getName(), new String(fileNameVector.get(0)), "File name should match"); + + VarCharVector seqNoVector = (VarCharVector) root.getVector(COMMIT_SEQNO_METADATA_FIELD.getFieldName()); + String seqNo = new String(seqNoVector.get(0)); + assertTrue(seqNo.startsWith(instantTime), "Sequence number should start with instant time"); + + // Verify user data fields + IntVector idVector = (IntVector) root.getVector("id"); + assertEquals(1, idVector.get(0), "ID should match"); + + VarCharVector nameVector = (VarCharVector) root.getVector("name"); + assertEquals("Alice", new String(nameVector.get(0)), "Name should match"); + + BigIntVector ageVector = (BigIntVector) root.getVector("age"); + assertEquals(30L, ageVector.get(0), "Age should match"); + + arrowReader.close(); + } + } + + @Test + public void testWriteRowWithoutMetadataPopulation() throws Exception { + // Schema WITHOUT meta fields + StructType schema = createSchemaWithoutMetaFields(); + + StoragePath path = new StoragePath(tempDir.getAbsolutePath() + "/test_without_metadata.lance"); + HoodieSparkLanceWriter writer = new HoodieSparkLanceWriter( + path, schema, instantTime, taskContextSupplier, storage, false); + + // Create row with just user data (no meta fields) + InternalRow row = createRow(1, "Bob", 25L); + HoodieKey key = new HoodieKey("key2", "partition2"); + + writer.writeRowWithMetadata(key, row); + writer.close(); + + // Verify using LanceFileReader + assertTrue(storage.exists(path)); + + try (BufferAllocator allocator = new RootAllocator(); + LanceFileReader reader = LanceFileReader.open(path.toString(), allocator)) { + + assertEquals(1, reader.numRows()); + + // Schema should have ONLY 3 user fields (no meta fields) + assertEquals(3, reader.schema().getFields().size(), "Should have only user fields"); + + // Read and verify data + ArrowReader arrowReader = reader.readAll(null, null, Integer.MAX_VALUE); + assertTrue(arrowReader.loadNextBatch()); + VectorSchemaRoot root = arrowReader.getVectorSchemaRoot(); + + // Verify NO meta fields exist + assertFalse(hasField(root, COMMIT_TIME_METADATA_FIELD.getFieldName()), "Should not have commit time field"); + assertFalse(hasField(root, RECORD_KEY_METADATA_FIELD.getFieldName()), "Should not have record key field"); + + // Verify user data + IntVector idVector = (IntVector) root.getVector("id"); + assertEquals(1, idVector.get(0)); + + VarCharVector nameVector = (VarCharVector) root.getVector("name"); + assertEquals("Bob", new String(nameVector.get(0))); + + arrowReader.close(); + } + } + + @Test + public void testWriteRowSimple() throws Exception { + StructType schema = createSchemaWithoutMetaFields(); + + StoragePath path = new StoragePath(tempDir.getAbsolutePath() + "/test_simple_write.lance"); + HoodieSparkLanceWriter writer = new HoodieSparkLanceWriter( + path, schema, instantTime, taskContextSupplier, storage, false); + + InternalRow row = createRow(1, "Charlie", 35L); + writer.writeRow("key3", row); + writer.close(); + + // Verify file exists and has correct record count + try (BufferAllocator allocator = new RootAllocator(); + LanceFileReader reader = LanceFileReader.open(path.toString(), allocator)) { + assertEquals(1, reader.numRows()); + } + } + + @Test + public void testSequenceIdGeneration() throws Exception { + StructType schema = createSchemaWithMetaFields(); + + StoragePath path = new StoragePath(tempDir.getAbsolutePath() + "/test_sequence_id.lance"); + HoodieSparkLanceWriter writer = new HoodieSparkLanceWriter( + path, schema, instantTime, taskContextSupplier, storage, true); + + // Write multiple records + for (int i = 0; i < 3; i++) { + InternalRow row = createRowWithMetaFields(i, "User" + i, 20L + i); + HoodieKey key = new HoodieKey("key" + i, "partition1"); + writer.writeRowWithMetadata(key, row); + } + writer.close(); + + // Verify sequence IDs are unique and properly formatted + try (BufferAllocator allocator = new RootAllocator(); + LanceFileReader reader = LanceFileReader.open(path.toString(), allocator)) { + + assertEquals(3, reader.numRows()); + + ArrowReader arrowReader = reader.readAll(null, null, Integer.MAX_VALUE); + arrowReader.loadNextBatch(); + VectorSchemaRoot root = arrowReader.getVectorSchemaRoot(); + + VarCharVector seqNoVector = (VarCharVector) root.getVector(COMMIT_SEQNO_METADATA_FIELD.getFieldName()); Review Comment: Then in that case, do we really need a separate test for this or should we combine? -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
