Examples with OVERWRITE com.facebook.presto.hive.PartitionUpdate.UpdateMode.OVERWRITE used on opensource projects

Example 1 with OVERWRITE

use of com.facebook.presto.hive.PartitionUpdate.UpdateMode.OVERWRITE in project presto by prestodb.

the class HiveMetadata method finishInsertInternal.

private Optional<ConnectorOutputMetadata> finishInsertInternal(ConnectorSession session, ConnectorInsertTableHandle insertHandle, Collection<Slice> fragments, Collection<ComputedStatistics> computedStatistics) {
    HiveInsertTableHandle handle = (HiveInsertTableHandle) insertHandle;
    List<PartitionUpdate> partitionUpdates = getPartitionUpdates(session, fragments);
    HiveStorageFormat tableStorageFormat = handle.getTableStorageFormat();
    partitionUpdates = PartitionUpdate.mergePartitionUpdates(partitionUpdates);
    MetastoreContext metastoreContext = getMetastoreContext(session);
    Table table = metastore.getTable(metastoreContext, handle.getSchemaName(), handle.getTableName()).orElseThrow(() -> new TableNotFoundException(handle.getSchemaTableName()));
    if (!table.getStorage().getStorageFormat().getInputFormat().equals(tableStorageFormat.getInputFormat()) && isRespectTableFormat(session)) {
        throw new PrestoException(HIVE_CONCURRENT_MODIFICATION_DETECTED, "Table format changed during insert");
    }
    if (handle.getBucketProperty().isPresent() && isCreateEmptyBucketFiles(session)) {
        List<PartitionUpdate> partitionUpdatesForMissingBuckets = computePartitionUpdatesForMissingBuckets(session, handle, table, partitionUpdates);
        // replace partitionUpdates before creating the zero-row files so that those files will be cleaned up if we end up rollback
        partitionUpdates = PartitionUpdate.mergePartitionUpdates(concat(partitionUpdates, partitionUpdatesForMissingBuckets));
        HdfsContext hdfsContext = new HdfsContext(session, table.getDatabaseName(), table.getTableName(), table.getStorage().getLocation(), false);
        for (PartitionUpdate partitionUpdate : partitionUpdatesForMissingBuckets) {
            Optional<Partition> partition = table.getPartitionColumns().isEmpty() ? Optional.empty() : Optional.of(partitionObjectBuilder.buildPartitionObject(session, table, partitionUpdate, prestoVersion, handle.getEncryptionInformation().map(encryptionInfo -> encryptionInfo.getDwrfEncryptionMetadata().map(DwrfEncryptionMetadata::getExtraMetadata).orElseGet(ImmutableMap::of)).orElseGet(ImmutableMap::of)));
            zeroRowFileCreator.createFiles(session, hdfsContext, partitionUpdate.getWritePath(), getTargetFileNames(partitionUpdate.getFileWriteInfos()), getStorageFormat(partition, table), handle.getCompressionCodec(), getSchema(partition, table));
        }
    }
    List<String> partitionedBy = table.getPartitionColumns().stream().map(Column::getName).collect(toImmutableList());
    Map<String, Type> columnTypes = handle.getInputColumns().stream().collect(toImmutableMap(HiveColumnHandle::getName, column -> column.getHiveType().getType(typeManager)));
    Map<List<String>, ComputedStatistics> partitionComputedStatistics = createComputedStatisticsToPartitionMap(computedStatistics, partitionedBy, columnTypes);
    Set<String> existingPartitions = getExistingPartitionNames(session.getIdentity(), metastoreContext, handle.getSchemaName(), handle.getTableName(), partitionUpdates);
    for (PartitionUpdate partitionUpdate : partitionUpdates) {
        if (partitionUpdate.getName().isEmpty()) {
            if (handle.getEncryptionInformation().isPresent()) {
                throw new PrestoException(HIVE_UNSUPPORTED_ENCRYPTION_OPERATION, "Inserting into an existing table with encryption enabled is not supported yet");
            }
            // insert into unpartitioned table
            PartitionStatistics partitionStatistics = createPartitionStatistics(session, partitionUpdate.getStatistics(), columnTypes, getColumnStatistics(partitionComputedStatistics, ImmutableList.of()));
            metastore.finishInsertIntoExistingTable(session, handle.getSchemaName(), handle.getTableName(), partitionUpdate.getWritePath(), getTargetFileNames(partitionUpdate.getFileWriteInfos()), partitionStatistics);
        } else if (partitionUpdate.getUpdateMode() == APPEND) {
            if (handle.getEncryptionInformation().isPresent()) {
                throw new PrestoException(HIVE_UNSUPPORTED_ENCRYPTION_OPERATION, "Inserting into an existing partition with encryption enabled is not supported yet");
            }
            // insert into existing partition
            List<String> partitionValues = toPartitionValues(partitionUpdate.getName());
            PartitionStatistics partitionStatistics = createPartitionStatistics(session, partitionUpdate.getStatistics(), columnTypes, getColumnStatistics(partitionComputedStatistics, partitionValues));
            metastore.finishInsertIntoExistingPartition(session, handle.getSchemaName(), handle.getTableName(), handle.getLocationHandle().getTargetPath().toString(), partitionValues, partitionUpdate.getWritePath(), getTargetFileNames(partitionUpdate.getFileWriteInfos()), partitionStatistics);
        } else if (partitionUpdate.getUpdateMode() == NEW || partitionUpdate.getUpdateMode() == OVERWRITE) {
            Map<String, String> extraPartitionMetadata = handle.getEncryptionInformation().map(encryptionInfo -> encryptionInfo.getDwrfEncryptionMetadata().map(DwrfEncryptionMetadata::getExtraMetadata).orElseGet(ImmutableMap::of)).orElseGet(ImmutableMap::of);
            if (isPreferManifestsToListFiles(session) && isFileRenamingEnabled(session)) {
                // Store list of file names and sizes in partition metadata when prefer_manifests_to_list_files and file_renaming_enabled are set to true
                extraPartitionMetadata = updatePartitionMetadataWithFileNamesAndSizes(partitionUpdate, extraPartitionMetadata);
            }
            // Track the manifest blob size
            getManifestSizeInBytes(session, partitionUpdate, extraPartitionMetadata).ifPresent(hivePartitionStats::addManifestSizeInBytes);
            // insert into new partition or overwrite existing partition
            Partition partition = partitionObjectBuilder.buildPartitionObject(session, table, partitionUpdate, prestoVersion, extraPartitionMetadata);
            if (!partition.getStorage().getStorageFormat().getInputFormat().equals(handle.getPartitionStorageFormat().getInputFormat()) && isRespectTableFormat(session)) {
                throw new PrestoException(HIVE_CONCURRENT_MODIFICATION_DETECTED, "Partition format changed during insert");
            }
            if (existingPartitions.contains(partitionUpdate.getName())) {
                if (partitionUpdate.getUpdateMode() == OVERWRITE) {
                    metastore.dropPartition(session, handle.getSchemaName(), handle.getTableName(), handle.getLocationHandle().getTargetPath().toString(), partition.getValues());
                } else {
                    throw new PrestoException(HIVE_PARTITION_READ_ONLY, "Cannot insert into an existing partition of Hive table: " + partitionUpdate.getName());
                }
            }
            PartitionStatistics partitionStatistics = createPartitionStatistics(session, partitionUpdate.getStatistics(), columnTypes, getColumnStatistics(partitionComputedStatistics, partition.getValues()));
            metastore.addPartition(session, handle.getSchemaName(), handle.getTableName(), table.getStorage().getLocation(), false, partition, partitionUpdate.getWritePath(), partitionStatistics);
        } else {
            throw new IllegalArgumentException(format("Unsupported update mode: %s", partitionUpdate.getUpdateMode()));
        }
    }
    return Optional.of(new HiveWrittenPartitions(partitionUpdates.stream().map(PartitionUpdate::getName).map(name -> name.isEmpty() ? UNPARTITIONED_ID : name).collect(toList())));
}