org.apache.iceberg.spark

Class SparkTableUtil

java.lang.Object

org.apache.iceberg.spark.SparkTableUtil

public class SparkTableUtil
extends java.lang.Object

Java version of the original SparkTableUtil.scala https://github.com/apache/iceberg/blob/apache-iceberg-0.8.0-incubating/spark/src/main/scala/org/apache/iceberg/spark/SparkTableUtil.scala

Nested Class Summary

Nested Classes

Modifier and Type	Class and Description
static class	SparkTableUtil.SparkPartition Class representing a table partition.

Method Summary

Modifier and Type	Method and Description
static java.util.List<SparkTableUtil.SparkPartition>	getPartitions(org.apache.spark.sql.SparkSession spark, java.lang.String table) Returns all partitions in the table.
static java.util.List<SparkTableUtil.SparkPartition>	getPartitions(org.apache.spark.sql.SparkSession spark, org.apache.spark.sql.catalyst.TableIdentifier tableIdent) Returns all partitions in the table.
static java.util.List<SparkTableUtil.SparkPartition>	getPartitionsByFilter(org.apache.spark.sql.SparkSession spark, java.lang.String table, java.lang.String predicate) Returns partitions that match the specified 'predicate'.
static java.util.List<SparkTableUtil.SparkPartition>	getPartitionsByFilter(org.apache.spark.sql.SparkSession spark, org.apache.spark.sql.catalyst.TableIdentifier tableIdent, org.apache.spark.sql.catalyst.expressions.Expression predicateExpr) Returns partitions that match the specified 'predicate'.
static void	importSparkPartitions(org.apache.spark.sql.SparkSession spark, java.util.List<SparkTableUtil.SparkPartition> partitions, Table targetTable, PartitionSpec spec, java.lang.String stagingDir) Import files from given partitions to an Iceberg table.
static void	importSparkTable(org.apache.spark.sql.SparkSession spark, org.apache.spark.sql.catalyst.TableIdentifier sourceTableIdent, Table targetTable, java.lang.String stagingDir) Import files from an existing Spark table to an Iceberg table.
static java.util.List<DataFile>	listPartition(java.util.Map<java.lang.String,java.lang.String> partition, java.lang.String uri, java.lang.String format, PartitionSpec spec, org.apache.hadoop.conf.Configuration conf, MetricsConfig metricsConfig) Returns the data files in a partition by listing the partition location.
static java.util.List<DataFile>	listPartition(SparkTableUtil.SparkPartition partition, PartitionSpec spec, SerializableConfiguration conf, MetricsConfig metricsConfig) Returns the data files in a partition by listing the partition location.
static org.apache.spark.sql.Dataset<org.apache.spark.sql.Row>	partitionDF(org.apache.spark.sql.SparkSession spark, java.lang.String table) Returns a DataFrame with a row for each partition in the table.
static org.apache.spark.sql.Dataset<org.apache.spark.sql.Row>	partitionDFByFilter(org.apache.spark.sql.SparkSession spark, java.lang.String table, java.lang.String expression) Returns a DataFrame with a row for each partition that matches the specified 'expression'.

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Method Detail

partitionDF

public static org.apache.spark.sql.Dataset<org.apache.spark.sql.Row> partitionDF(org.apache.spark.sql.SparkSession spark,
                                                                                 java.lang.String table)

Returns a DataFrame with a row for each partition in the table. The DataFrame has 3 columns, partition key (a=1/b=2), partition location, and format (avro or parquet).

Parameters:

spark - a Spark session

table - a table name and (optional) database

Returns:

a DataFrame of the table's partitions

partitionDFByFilter

public static org.apache.spark.sql.Dataset<org.apache.spark.sql.Row> partitionDFByFilter(org.apache.spark.sql.SparkSession spark,
                                                                                         java.lang.String table,
                                                                                         java.lang.String expression)

Returns a DataFrame with a row for each partition that matches the specified 'expression'.

Parameters:

spark - a Spark session.

table - name of the table.

expression - The expression whose matching partitions are returned.

Returns:

a DataFrame of the table partitions.

getPartitions

public static java.util.List<SparkTableUtil.SparkPartition> getPartitions(org.apache.spark.sql.SparkSession spark,
                                                                          java.lang.String table)

Returns all partitions in the table.

Parameters:

spark - a Spark session

table - a table name and (optional) database

Returns:

all table's partitions

getPartitions

public static java.util.List<SparkTableUtil.SparkPartition> getPartitions(org.apache.spark.sql.SparkSession spark,
                                                                          org.apache.spark.sql.catalyst.TableIdentifier tableIdent)

Returns all partitions in the table.

Parameters:

spark - a Spark session

tableIdent - a table identifier

Returns:

all table's partitions

getPartitionsByFilter

public static java.util.List<SparkTableUtil.SparkPartition> getPartitionsByFilter(org.apache.spark.sql.SparkSession spark,
                                                                                  java.lang.String table,
                                                                                  java.lang.String predicate)

Returns partitions that match the specified 'predicate'.

Parameters:

spark - a Spark session

table - a table name and (optional) database

predicate - a predicate on partition columns

Returns:

matching table's partitions

getPartitionsByFilter

public static java.util.List<SparkTableUtil.SparkPartition> getPartitionsByFilter(org.apache.spark.sql.SparkSession spark,
                                                                                  org.apache.spark.sql.catalyst.TableIdentifier tableIdent,
                                                                                  org.apache.spark.sql.catalyst.expressions.Expression predicateExpr)

Returns partitions that match the specified 'predicate'.

Parameters:

spark - a Spark session

tableIdent - a table identifier

predicateExpr - a predicate expression on partition columns

Returns:

matching table's partitions

listPartition

public static java.util.List<DataFile> listPartition(SparkTableUtil.SparkPartition partition,
                                                     PartitionSpec spec,
                                                     SerializableConfiguration conf,
                                                     MetricsConfig metricsConfig)

Returns the data files in a partition by listing the partition location. For Parquet and ORC partitions, this will read metrics from the file footer. For Avro partitions, metrics are set to null.

Parameters:

partition - a partition

conf - a serializable Hadoop conf

metricsConfig - a metrics conf

Returns:

a List of DataFile

listPartition

public static java.util.List<DataFile> listPartition(java.util.Map<java.lang.String,java.lang.String> partition,
                                                     java.lang.String uri,
                                                     java.lang.String format,
                                                     PartitionSpec spec,
                                                     org.apache.hadoop.conf.Configuration conf,
                                                     MetricsConfig metricsConfig)

Returns the data files in a partition by listing the partition location. For Parquet and ORC partitions, this will read metrics from the file footer. For Avro partitions, metrics are set to null.

Parameters:

partition - partition key, e.g., "a=1/b=2"

uri - partition location URI

format - partition format, avro or parquet

conf - a Hadoop conf

metricsConfig - a metrics conf

Returns:

a List of DataFile

importSparkTable

public static void importSparkTable(org.apache.spark.sql.SparkSession spark,
                                    org.apache.spark.sql.catalyst.TableIdentifier sourceTableIdent,
                                    Table targetTable,
                                    java.lang.String stagingDir)

Import files from an existing Spark table to an Iceberg table. The import uses the Spark session to get table metadata. It assumes no operation is going on the original and target table and thus is not thread-safe.

Parameters:

spark - a Spark session

sourceTableIdent - an identifier of the source Spark table

targetTable - an Iceberg table where to import the data

stagingDir - a staging directory to store temporary manifest files

importSparkPartitions

public static void importSparkPartitions(org.apache.spark.sql.SparkSession spark,
                                         java.util.List<SparkTableUtil.SparkPartition> partitions,
                                         Table targetTable,
                                         PartitionSpec spec,
                                         java.lang.String stagingDir)

Import files from given partitions to an Iceberg table.

Parameters:

spark - a Spark session

partitions - partitions to import

targetTable - an Iceberg table where to import the data

spec - a partition spec

stagingDir - a staging directory to store temporary manifest files