[HUDI-1518] Remove the logic that delete replaced file when archive (#3310)
* remove delete replaced file when archive * done * remove unsed import * remove delete replaced files when archive realted UT * code reviewed Co-authored-by: yuezhang <yuezhang@freewheel.tv>
This commit is contained in:
@@ -21,15 +21,12 @@ package org.apache.hudi.io;
|
||||
import org.apache.hudi.avro.model.HoodieActionInstant;
|
||||
import org.apache.hudi.avro.model.HoodieCleanMetadata;
|
||||
import org.apache.hudi.avro.model.HoodieCleanerPlan;
|
||||
import org.apache.hudi.avro.model.HoodieRequestedReplaceMetadata;
|
||||
import org.apache.hudi.avro.model.HoodieRollbackMetadata;
|
||||
import org.apache.hudi.client.utils.MetadataConversionUtils;
|
||||
import org.apache.hudi.common.HoodieCleanStat;
|
||||
import org.apache.hudi.common.fs.FSUtils;
|
||||
import org.apache.hudi.common.fs.HoodieWrapperFileSystem;
|
||||
import org.apache.hudi.common.model.HoodieCleaningPolicy;
|
||||
import org.apache.hudi.common.model.HoodieCommitMetadata;
|
||||
import org.apache.hudi.common.model.HoodieReplaceCommitMetadata;
|
||||
import org.apache.hudi.common.model.WriteOperationType;
|
||||
import org.apache.hudi.common.table.HoodieTableMetaClient;
|
||||
import org.apache.hudi.common.table.timeline.HoodieActiveTimeline;
|
||||
@@ -50,7 +47,6 @@ import org.apache.hudi.table.HoodieTimelineArchiveLog;
|
||||
import org.apache.hudi.testutils.HoodieClientTestHarness;
|
||||
|
||||
import org.apache.hadoop.conf.Configuration;
|
||||
import org.apache.hadoop.fs.FileStatus;
|
||||
import org.apache.hadoop.fs.Path;
|
||||
import org.junit.jupiter.api.AfterEach;
|
||||
import org.junit.jupiter.api.BeforeEach;
|
||||
@@ -233,45 +229,6 @@ public class TestHoodieTimelineArchiveLog extends HoodieClientTestHarness {
|
||||
verifyInflightInstants(metaClient, 2);
|
||||
}
|
||||
|
||||
@Test
|
||||
public void testArchiveTableWithReplacedFiles() throws Exception {
|
||||
HoodieTestUtils.init(hadoopConf, basePath);
|
||||
HoodieWriteConfig cfg = HoodieWriteConfig.newBuilder().withPath(basePath)
|
||||
.withSchema(HoodieTestDataGenerator.TRIP_EXAMPLE_SCHEMA).withParallelism(2, 2).forTable("test-trip-table")
|
||||
.withCompactionConfig(HoodieCompactionConfig.newBuilder().retainCommits(1).archiveCommitsWith(2, 3).build())
|
||||
.build();
|
||||
|
||||
// when using insert_overwrite or insert_overwrite_table
|
||||
// first commit may without replaceFileIds
|
||||
createReplaceMetadataWithoutReplaceFileId("000");
|
||||
|
||||
int numCommits = 4;
|
||||
int commitInstant = 100;
|
||||
for (int i = 0; i < numCommits; i++) {
|
||||
createReplaceMetadata(String.valueOf(commitInstant));
|
||||
commitInstant += 100;
|
||||
}
|
||||
|
||||
metaClient = HoodieTableMetaClient.reload(metaClient);
|
||||
HoodieTimeline timeline = metaClient.getActiveTimeline().getCommitsTimeline().filterCompletedInstants();
|
||||
HoodieTable table = HoodieSparkTable.create(cfg, context, metaClient);
|
||||
assertEquals(5, timeline.countInstants(), "Loaded 5 commits and the count should match");
|
||||
HoodieTimelineArchiveLog archiveLog = new HoodieTimelineArchiveLog(cfg, table);
|
||||
boolean result = archiveLog.archiveIfRequired(context);
|
||||
assertTrue(result);
|
||||
|
||||
FileStatus[] allFiles = metaClient.getFs().listStatus(new Path(basePath + "/" + HoodieTestDataGenerator.DEFAULT_FIRST_PARTITION_PATH));
|
||||
Set<String> allFileIds = Arrays.stream(allFiles).map(fs -> FSUtils.getFileIdFromFilePath(fs.getPath())).collect(Collectors.toSet());
|
||||
|
||||
// verify 100-1,200-1 are deleted by archival
|
||||
assertFalse(allFileIds.contains("file-100-1"));
|
||||
assertFalse(allFileIds.contains("file-200-1"));
|
||||
assertTrue(allFileIds.contains("file-100-2"));
|
||||
assertTrue(allFileIds.contains("file-200-2"));
|
||||
assertTrue(allFileIds.contains("file-300-1"));
|
||||
assertTrue(allFileIds.contains("file-400-1"));
|
||||
}
|
||||
|
||||
@Test
|
||||
public void testArchiveTableWithNoArchival() throws IOException {
|
||||
HoodieWriteConfig cfg = HoodieWriteConfig.newBuilder().withPath(basePath)
|
||||
@@ -679,42 +636,6 @@ public class TestHoodieTimelineArchiveLog extends HoodieClientTestHarness {
|
||||
assertEquals(notArchivedInstants, Arrays.asList(notArchivedInstant1, notArchivedInstant2, notArchivedInstant3), "");
|
||||
}
|
||||
|
||||
private void createReplaceMetadataWithoutReplaceFileId(String instantTime) throws Exception {
|
||||
|
||||
// create replace instant without a previous replace commit
|
||||
HoodieRequestedReplaceMetadata requestedReplaceMetadata = HoodieRequestedReplaceMetadata.newBuilder()
|
||||
.setOperationType(WriteOperationType.INSERT_OVERWRITE_TABLE.toString())
|
||||
.setVersion(1)
|
||||
.setExtraMetadata(Collections.emptyMap())
|
||||
.build();
|
||||
HoodieReplaceCommitMetadata completeReplaceMetadata = new HoodieReplaceCommitMetadata();
|
||||
HoodieCommitMetadata inflightReplaceMetadata = new HoodieCommitMetadata();
|
||||
completeReplaceMetadata.setOperationType(WriteOperationType.INSERT_OVERWRITE_TABLE);
|
||||
inflightReplaceMetadata.setOperationType(WriteOperationType.INSERT_OVERWRITE_TABLE);
|
||||
HoodieTestTable.of(metaClient)
|
||||
.addReplaceCommit(instantTime, Option.of(requestedReplaceMetadata), Option.of(inflightReplaceMetadata), completeReplaceMetadata);
|
||||
}
|
||||
|
||||
private void createReplaceMetadata(String instantTime) throws Exception {
|
||||
String fileId1 = "file-" + instantTime + "-1";
|
||||
String fileId2 = "file-" + instantTime + "-2";
|
||||
|
||||
// create replace instant to mark fileId1 as deleted
|
||||
HoodieRequestedReplaceMetadata requestedReplaceMetadata = HoodieRequestedReplaceMetadata.newBuilder()
|
||||
.setOperationType(WriteOperationType.INSERT_OVERWRITE.toString())
|
||||
.setVersion(1)
|
||||
.setExtraMetadata(Collections.emptyMap())
|
||||
.build();
|
||||
HoodieReplaceCommitMetadata completeReplaceMetadata = new HoodieReplaceCommitMetadata();
|
||||
HoodieCommitMetadata inflightReplaceMetadata = new HoodieCommitMetadata();
|
||||
completeReplaceMetadata.addReplaceFileId(HoodieTestDataGenerator.DEFAULT_FIRST_PARTITION_PATH, fileId1);
|
||||
completeReplaceMetadata.setOperationType(WriteOperationType.INSERT_OVERWRITE);
|
||||
inflightReplaceMetadata.setOperationType(WriteOperationType.INSERT_OVERWRITE);
|
||||
HoodieTestTable.of(metaClient)
|
||||
.addReplaceCommit(instantTime, Option.of(requestedReplaceMetadata), Option.of(inflightReplaceMetadata), completeReplaceMetadata)
|
||||
.withBaseFilesInPartition(HoodieTestDataGenerator.DEFAULT_FIRST_PARTITION_PATH, fileId1, fileId2);
|
||||
}
|
||||
|
||||
private HoodieInstant createCleanMetadata(String instantTime, boolean inflightOnly) throws IOException {
|
||||
HoodieCleanerPlan cleanerPlan = new HoodieCleanerPlan(new HoodieActionInstant("", "", ""), "", new HashMap<>(),
|
||||
CleanPlanV2MigrationHandler.VERSION, new HashMap<>());
|
||||
|
||||
Reference in New Issue
Block a user