Examples with HoodieFileGroupId org.apache.hudi.common.model.HoodieFileGroupId used on opensource projects

Example 1 with HoodieFileGroupId

use of org.apache.hudi.common.model.HoodieFileGroupId in project hudi by apache.

the class CompactionTestBase method runNextDeltaCommits.

protected List<HoodieRecord> runNextDeltaCommits(SparkRDDWriteClient client, final HoodieReadClient readClient, List<String> deltaInstants, List<HoodieRecord> records, HoodieWriteConfig cfg, boolean insertFirst, List<String> expPendingCompactionInstants) throws Exception {
    HoodieTableMetaClient metaClient = HoodieTableMetaClient.builder().setConf(hadoopConf).setBasePath(cfg.getBasePath()).build();
    List<Pair<String, HoodieCompactionPlan>> pendingCompactions = readClient.getPendingCompactions();
    List<String> gotPendingCompactionInstants = pendingCompactions.stream().map(pc -> pc.getKey()).sorted().collect(Collectors.toList());
    assertEquals(expPendingCompactionInstants, gotPendingCompactionInstants);
    Map<HoodieFileGroupId, Pair<String, HoodieCompactionOperation>> fgIdToCompactionOperation = CompactionUtils.getAllPendingCompactionOperations(metaClient);
    if (insertFirst) {
        // Use first instant for inserting records
        String firstInstant = deltaInstants.get(0);
        deltaInstants = deltaInstants.subList(1, deltaInstants.size());
        JavaRDD<HoodieRecord> writeRecords = jsc.parallelize(records, 1);
        client.startCommitWithTime(firstInstant);
        JavaRDD<WriteStatus> statuses = client.upsert(writeRecords, firstInstant);
        List<WriteStatus> statusList = statuses.collect();
        if (!cfg.shouldAutoCommit()) {
            client.commit(firstInstant, statuses);
        }
        assertNoWriteErrors(statusList);
        metaClient = HoodieTableMetaClient.builder().setConf(hadoopConf).setBasePath(cfg.getBasePath()).build();
        HoodieTable hoodieTable = getHoodieTable(metaClient, cfg);
        List<HoodieBaseFile> dataFilesToRead = getCurrentLatestBaseFiles(hoodieTable);
        assertTrue(dataFilesToRead.stream().findAny().isPresent(), "should list the base files we wrote in the delta commit");
        validateDeltaCommit(firstInstant, fgIdToCompactionOperation, cfg);
    }
    int numRecords = records.size();
    for (String instantTime : deltaInstants) {
        records = dataGen.generateUpdates(instantTime, numRecords);
        metaClient = HoodieTableMetaClient.builder().setConf(hadoopConf).setBasePath(cfg.getBasePath()).build();
        createNextDeltaCommit(instantTime, records, client, metaClient, cfg, false);
        validateDeltaCommit(instantTime, fgIdToCompactionOperation, cfg);
    }
    return records;
}

Example 2 with HoodieFileGroupId

use of org.apache.hudi.common.model.HoodieFileGroupId in project hudi by apache.

the class CompactionTestBase method executeCompactionWithReplacedFiles.

protected void executeCompactionWithReplacedFiles(String compactionInstantTime, SparkRDDWriteClient client, HoodieTable table, HoodieWriteConfig cfg, String[] partitions, Set<HoodieFileGroupId> replacedFileIds) throws IOException {
    client.compact(compactionInstantTime);
    List<FileSlice> fileSliceList = getCurrentLatestFileSlices(table);
    assertTrue(fileSliceList.stream().findAny().isPresent(), "Ensure latest file-slices are not empty");
    assertFalse(fileSliceList.stream().anyMatch(fs -> replacedFileIds.contains(fs.getFileGroupId())), "Compacted files should not show up in latest slices");
    // verify that there is a commit
    table = getHoodieTable(HoodieTableMetaClient.builder().setConf(hadoopConf).setBasePath(cfg.getBasePath()).setLoadActiveTimelineOnLoad(true).build(), cfg);
    HoodieTimeline timeline = table.getMetaClient().getCommitTimeline().filterCompletedInstants();
    // verify compaction commit is visible in timeline
    assertTrue(timeline.filterCompletedInstants().getInstants().filter(instant -> compactionInstantTime.equals(instant.getTimestamp())).findFirst().isPresent());
    for (String partition : partitions) {
        table.getSliceView().getLatestFileSlicesBeforeOrOn(partition, compactionInstantTime, true).forEach(fs -> {
            // verify that all log files are merged
            assertEquals(0, fs.getLogFiles().count());
            assertTrue(fs.getBaseFile().isPresent());
        });
    }
}

Example 3 with HoodieFileGroupId

use of org.apache.hudi.common.model.HoodieFileGroupId in project hudi by apache.

the class TestAsyncCompaction method testCompactionOnReplacedFiles.

@Test
public void testCompactionOnReplacedFiles() throws Exception {
    // Schedule a compaction. Replace those file groups and ensure compaction completes successfully.
    HoodieWriteConfig cfg = getConfig(true);
    try (SparkRDDWriteClient client = getHoodieWriteClient(cfg)) {
        HoodieReadClient readClient = getHoodieReadClient(cfg.getBasePath());
        String firstInstantTime = "001";
        String secondInstantTime = "004";
        String compactionInstantTime = "005";
        String replaceInstantTime = "006";
        String fourthInstantTime = "007";
        int numRecs = 2000;
        List<HoodieRecord> records = dataGen.generateInserts(firstInstantTime, numRecs);
        runNextDeltaCommits(client, readClient, Arrays.asList(firstInstantTime, secondInstantTime), records, cfg, true, new ArrayList<>());
        HoodieTableMetaClient metaClient = HoodieTableMetaClient.builder().setConf(hadoopConf).setBasePath(cfg.getBasePath()).build();
        HoodieTable hoodieTable = getHoodieTable(metaClient, cfg);
        scheduleCompaction(compactionInstantTime, client, cfg);
        metaClient.reloadActiveTimeline();
        HoodieInstant pendingCompactionInstant = metaClient.getActiveTimeline().filterPendingCompactionTimeline().firstInstant().get();
        assertEquals(compactionInstantTime, pendingCompactionInstant.getTimestamp(), "Pending Compaction instant has expected instant time");
        Set<HoodieFileGroupId> fileGroupsBeforeReplace = getAllFileGroups(hoodieTable, dataGen.getPartitionPaths());
        // replace by using insertOverwrite
        JavaRDD<HoodieRecord> replaceRecords = jsc.parallelize(dataGen.generateInserts(replaceInstantTime, numRecs), 1);
        client.startCommitWithTime(replaceInstantTime, HoodieTimeline.REPLACE_COMMIT_ACTION);
        client.insertOverwrite(replaceRecords, replaceInstantTime);
        metaClient.reloadActiveTimeline();
        hoodieTable = getHoodieTable(metaClient, cfg);
        Set<HoodieFileGroupId> newFileGroups = getAllFileGroups(hoodieTable, dataGen.getPartitionPaths());
        // make sure earlier file groups are not visible
        assertEquals(0, newFileGroups.stream().filter(fg -> fileGroupsBeforeReplace.contains(fg)).count());
        // compaction should run with associated file groups are replaced
        executeCompactionWithReplacedFiles(compactionInstantTime, client, hoodieTable, cfg, dataGen.getPartitionPaths(), fileGroupsBeforeReplace);
    }
}

Example 4 with HoodieFileGroupId

use of org.apache.hudi.common.model.HoodieFileGroupId in project hudi by apache.

the class TestHoodieCompactionStrategy method createCompactionOperations.

private List<HoodieCompactionOperation> createCompactionOperations(HoodieWriteConfig config, Map<Long, List<Long>> sizesMap, Map<Long, String> keyToPartitionMap) {
    List<HoodieCompactionOperation> operations = new ArrayList<>(sizesMap.size());
    sizesMap.forEach((k, v) -> {
        HoodieBaseFile df = TestHoodieBaseFile.newDataFile(k);
        String partitionPath = keyToPartitionMap.get(k);
        List<HoodieLogFile> logFiles = v.stream().map(TestHoodieLogFile::newLogFile).collect(Collectors.toList());
        FileSlice slice = new FileSlice(new HoodieFileGroupId(partitionPath, df.getFileId()), df.getCommitTime());
        slice.setBaseFile(df);
        logFiles.stream().forEach(f -> slice.addLogFile(f));
        operations.add(new HoodieCompactionOperation(df.getCommitTime(), logFiles.stream().map(s -> s.getPath().toString()).collect(Collectors.toList()), df.getPath(), df.getFileId(), partitionPath, config.getCompactionStrategy().captureMetrics(config, slice), df.getBootstrapBaseFile().map(BaseFile::getPath).orElse(null)));
    });
    return operations;
}

Example 5 with HoodieFileGroupId

use of org.apache.hudi.common.model.HoodieFileGroupId in project hudi by apache.

the class HoodieCompactionAdminTool method run.

/**
 * Executes one of compaction admin operations.
 */
public void run(JavaSparkContext jsc) throws Exception {
    HoodieTableMetaClient metaClient = HoodieTableMetaClient.builder().setConf(jsc.hadoopConfiguration()).setBasePath(cfg.basePath).build();
    try (CompactionAdminClient admin = new CompactionAdminClient(new HoodieSparkEngineContext(jsc), cfg.basePath)) {
        final FileSystem fs = FSUtils.getFs(cfg.basePath, jsc.hadoopConfiguration());
        if (cfg.outputPath != null && fs.exists(new Path(cfg.outputPath))) {
            throw new IllegalStateException("Output File Path already exists");
        }
        switch(cfg.operation) {
            case VALIDATE:
                List<ValidationOpResult> res = admin.validateCompactionPlan(metaClient, cfg.compactionInstantTime, cfg.parallelism);
                if (cfg.printOutput) {
                    printOperationResult("Result of Validation Operation :", res);
                }
                serializeOperationResult(fs, res);
                break;
            case UNSCHEDULE_FILE:
                List<RenameOpResult> r = admin.unscheduleCompactionFileId(new HoodieFileGroupId(cfg.partitionPath, cfg.fileId), cfg.skipValidation, cfg.dryRun);
                if (cfg.printOutput) {
                    System.out.println(r);
                }
                serializeOperationResult(fs, r);
                break;
            case UNSCHEDULE_PLAN:
                List<RenameOpResult> r2 = admin.unscheduleCompactionPlan(cfg.compactionInstantTime, cfg.skipValidation, cfg.parallelism, cfg.dryRun);
                if (cfg.printOutput) {
                    printOperationResult("Result of Unscheduling Compaction Plan :", r2);
                }
                serializeOperationResult(fs, r2);
                break;
            case REPAIR:
                List<RenameOpResult> r3 = admin.repairCompaction(cfg.compactionInstantTime, cfg.parallelism, cfg.dryRun);
                if (cfg.printOutput) {
                    printOperationResult("Result of Repair Operation :", r3);
                }
                serializeOperationResult(fs, r3);
                break;
            default:
                throw new IllegalStateException("Not yet implemented !!");
        }
    }
}