Examples with HiveWrittenPartitions io.trino.plugin.hive.HiveWrittenPartitions used on opensource projects

Example 1 with HiveWrittenPartitions

use of io.trino.plugin.hive.HiveWrittenPartitions in project trino by trinodb.

the class IcebergMetadata method finishInsert.

@Override
public Optional<ConnectorOutputMetadata> finishInsert(ConnectorSession session, ConnectorInsertTableHandle insertHandle, Collection<Slice> fragments, Collection<ComputedStatistics> computedStatistics) {
    IcebergWritableTableHandle table = (IcebergWritableTableHandle) insertHandle;
    Table icebergTable = transaction.table();
    List<CommitTaskData> commitTasks = fragments.stream().map(slice -> commitTaskCodec.fromJson(slice.getBytes())).collect(toImmutableList());
    Type[] partitionColumnTypes = icebergTable.spec().fields().stream().map(field -> field.transform().getResultType(icebergTable.schema().findType(field.sourceId()))).toArray(Type[]::new);
    AppendFiles appendFiles = transaction.newAppend();
    ImmutableSet.Builder<String> writtenFiles = ImmutableSet.builder();
    for (CommitTaskData task : commitTasks) {
        DataFiles.Builder builder = DataFiles.builder(icebergTable.spec()).withPath(task.getPath()).withFileSizeInBytes(task.getFileSizeInBytes()).withFormat(table.getFileFormat().toIceberg()).withMetrics(task.getMetrics().metrics());
        if (!icebergTable.spec().fields().isEmpty()) {
            String partitionDataJson = task.getPartitionDataJson().orElseThrow(() -> new VerifyException("No partition data for partitioned table"));
            builder.withPartition(PartitionData.fromJson(partitionDataJson, partitionColumnTypes));
        }
        appendFiles.appendFile(builder.build());
        writtenFiles.add(task.getPath());
    }
    // try to leave as little garbage as possible behind
    if (table.getRetryMode() != NO_RETRIES) {
        cleanExtraOutputFiles(session, writtenFiles.build());
    }
    appendFiles.commit();
    transaction.commitTransaction();
    transaction = null;
    return Optional.of(new HiveWrittenPartitions(commitTasks.stream().map(CommitTaskData::getPath).collect(toImmutableList())));
}

Example 2 with HiveWrittenPartitions

use of io.trino.plugin.hive.HiveWrittenPartitions in project trino by trinodb.

the class IcebergMetadata method finishRefreshMaterializedView.

@Override
public Optional<ConnectorOutputMetadata> finishRefreshMaterializedView(ConnectorSession session, ConnectorTableHandle tableHandle, ConnectorInsertTableHandle insertHandle, Collection<Slice> fragments, Collection<ComputedStatistics> computedStatistics, List<ConnectorTableHandle> sourceTableHandles) {
    // delete before insert .. simulating overwrite
    executeDelete(session, tableHandle);
    IcebergWritableTableHandle table = (IcebergWritableTableHandle) insertHandle;
    Table icebergTable = transaction.table();
    List<CommitTaskData> commitTasks = fragments.stream().map(slice -> commitTaskCodec.fromJson(slice.getBytes())).collect(toImmutableList());
    Type[] partitionColumnTypes = icebergTable.spec().fields().stream().map(field -> field.transform().getResultType(icebergTable.schema().findType(field.sourceId()))).toArray(Type[]::new);
    AppendFiles appendFiles = transaction.newFastAppend();
    ImmutableSet.Builder<String> writtenFiles = ImmutableSet.builder();
    for (CommitTaskData task : commitTasks) {
        DataFiles.Builder builder = DataFiles.builder(icebergTable.spec()).withPath(task.getPath()).withFileSizeInBytes(task.getFileSizeInBytes()).withFormat(table.getFileFormat().toIceberg()).withMetrics(task.getMetrics().metrics());
        if (!icebergTable.spec().fields().isEmpty()) {
            String partitionDataJson = task.getPartitionDataJson().orElseThrow(() -> new VerifyException("No partition data for partitioned table"));
            builder.withPartition(PartitionData.fromJson(partitionDataJson, partitionColumnTypes));
        }
        appendFiles.appendFile(builder.build());
        writtenFiles.add(task.getPath());
    }
    String dependencies = sourceTableHandles.stream().map(handle -> (IcebergTableHandle) handle).filter(handle -> handle.getSnapshotId().isPresent()).map(handle -> handle.getSchemaTableName() + "=" + handle.getSnapshotId().get()).distinct().collect(joining(","));
    // try to leave as little garbage as possible behind
    if (table.getRetryMode() != NO_RETRIES) {
        cleanExtraOutputFiles(session, writtenFiles.build());
    }
    // Update the 'dependsOnTables' property that tracks tables on which the materialized view depends and the corresponding snapshot ids of the tables
    appendFiles.set(DEPENDS_ON_TABLES, dependencies);
    appendFiles.commit();
    transaction.commitTransaction();
    transaction = null;
    return Optional.of(new HiveWrittenPartitions(commitTasks.stream().map(CommitTaskData::getPath).collect(toImmutableList())));
}