Package org.apache.spark.sql.streaming

Class DataStreamWriter<T>

Object
org.apache.spark.sql.streaming.DataStreamWriter<T>
public final class DataStreamWriter<T> extends Object
Interface used to write a streaming Dataset to external storage systems (e.g. file systems, key-value stores, etc). Use Dataset.writeStream to access this.

Since:
2.0.0

  • Method Details

    • SOURCE_NAME_MEMORY

      public static String SOURCE_NAME_MEMORY()

    • SOURCE_NAME_FOREACH

      public static String SOURCE_NAME_FOREACH()

    • SOURCE_NAME_FOREACH_BATCH

      public static String SOURCE_NAME_FOREACH_BATCH()

    • SOURCE_NAME_CONSOLE

      public static String SOURCE_NAME_CONSOLE()

    • SOURCE_NAME_TABLE

      public static String SOURCE_NAME_TABLE()

    • SOURCE_NAME_NOOP

      public static String SOURCE_NAME_NOOP()

    • SOURCES_ALLOW_ONE_TIME_QUERY

      public static scala.collection.immutable.Seq<String> SOURCES_ALLOW_ONE_TIME_QUERY()

    • partitionBy

      public DataStreamWriter<T> partitionBy(String... colNames)
      Partitions the output by the given columns on the file system. If specified, the output is laid out on the file system similar to Hive's partitioning scheme. As an example, when we partition a dataset by year and then month, the directory layout would look like:

      • year=2016/month=01/
      • year=2016/month=02/

      Partitioning is one of the most widely used techniques to optimize physical data layout. It provides a coarse-grained index for skipping unnecessary data reads when queries have predicates on the partitioned columns. In order for partitioning to work well, the number of distinct values in each column should typically be less than tens of thousands.

      Parameters:
      colNames - (undocumented)
      Returns:
      (undocumented)
      Since:
      2.0.0

    • clusterBy

      public DataStreamWriter<T> clusterBy(String... colNames)
      Clusters the output by the given columns. If specified, the output is laid out such that records with similar values on the clustering column are grouped together in the same file.

      Clustering improves query efficiency by allowing queries with predicates on the clustering columns to skip unnecessary data. Unlike partitioning, clustering can be used on very high cardinality columns.

      Parameters:
      colNames - (undocumented)
      Returns:
      (undocumented)
      Since:
      4.0.0

    • outputMode

      public DataStreamWriter<T> outputMode(OutputMode outputMode)
      Specifies how data of a streaming DataFrame/Dataset is written to a streaming sink.
      • OutputMode.Append(): only the new rows in the streaming DataFrame/Dataset will be written to the sink.
      • OutputMode.Complete(): all the rows in the streaming DataFrame/Dataset will be written to the sink every time there are some updates.
      • OutputMode.Update(): only the rows that were updated in the streaming DataFrame/Dataset will be written to the sink every time there are some updates. If the query doesn't contain aggregations, it will be equivalent to OutputMode.Append() mode.

      Parameters:
      outputMode - (undocumented)
      Returns:
      (undocumented)
      Since:
      2.0.0

    • outputMode

      public DataStreamWriter<T> outputMode(String outputMode)
      Specifies how data of a streaming DataFrame/Dataset is written to a streaming sink.
      • append: only the new rows in the streaming DataFrame/Dataset will be written to the sink.
      • complete: all the rows in the streaming DataFrame/Dataset will be written to the sink every time there are some updates.
      • update: only the rows that were updated in the streaming DataFrame/Dataset will be written to the sink every time there are some updates. If the query doesn't contain aggregations, it will be equivalent to append mode.

      Parameters:
      outputMode - (undocumented)
      Returns:
      (undocumented)
      Since:
      2.0.0

    • trigger

      public DataStreamWriter<T> trigger(Trigger trigger)
      Set the trigger for the stream query. The default value is ProcessingTime(0) and it will run the query as fast as possible.

      Scala Example: 

      
   df.writeStream.trigger(ProcessingTime("10 seconds"))

   import scala.concurrent.duration._
   df.writeStream.trigger(ProcessingTime(10.seconds))

      Java Example: 

      
   df.writeStream().trigger(ProcessingTime.create("10 seconds"))

   import java.util.concurrent.TimeUnit
   df.writeStream().trigger(ProcessingTime.create(10, TimeUnit.SECONDS))

      Parameters:
      trigger - (undocumented)
      Returns:
      (undocumented)
      Since:
      2.0.0

    • queryName

      public DataStreamWriter<T> queryName(String queryName)
      Specifies the name of the StreamingQuery that can be started with start(). This name must be unique among all the currently active queries in the associated SQLContext.

      Parameters:
      queryName - (undocumented)
      Returns:
      (undocumented)
      Since:
      2.0.0

    • format

      public DataStreamWriter<T> format(String source)
      Specifies the underlying output data source.

      Parameters:
      source - (undocumented)
      Returns:
      (undocumented)
      Since:
      2.0.0

    • partitionBy

      public DataStreamWriter<T> partitionBy(scala.collection.immutable.Seq<String> colNames)
      Partitions the output by the given columns on the file system. If specified, the output is laid out on the file system similar to Hive's partitioning scheme. As an example, when we partition a dataset by year and then month, the directory layout would look like:

      • year=2016/month=01/
      • year=2016/month=02/

      Partitioning is one of the most widely used techniques to optimize physical data layout. It provides a coarse-grained index for skipping unnecessary data reads when queries have predicates on the partitioned columns. In order for partitioning to work well, the number of distinct values in each column should typically be less than tens of thousands.

      Parameters:
      colNames - (undocumented)
      Returns:
      (undocumented)
      Since:
      2.0.0

    • clusterBy

      public DataStreamWriter<T> clusterBy(scala.collection.immutable.Seq<String> colNames)
      Clusters the output by the given columns. If specified, the output is laid out such that records with similar values on the clustering column are grouped together in the same file.

      Clustering improves query efficiency by allowing queries with predicates on the clustering columns to skip unnecessary data. Unlike partitioning, clustering can be used on very high cardinality columns.

      Parameters:
      colNames - (undocumented)
      Returns:
      (undocumented)
      Since:
      4.0.0

    • option

      public DataStreamWriter<T> option(String key, String value)
      Adds an output option for the underlying data source.

      Parameters:
      key - (undocumented)
      value - (undocumented)
      Returns:
      (undocumented)
      Since:
      2.0.0

    • option

      public DataStreamWriter<T> option(String key, boolean value)
      Adds an output option for the underlying data source.

      Parameters:
      key - (undocumented)
      value - (undocumented)
      Returns:
      (undocumented)
      Since:
      2.0.0

    • option

      public DataStreamWriter<T> option(String key, long value)
      Adds an output option for the underlying data source.

      Parameters:
      key - (undocumented)
      value - (undocumented)
      Returns:
      (undocumented)
      Since:
      2.0.0

    • option

      public DataStreamWriter<T> option(String key, double value)
      Adds an output option for the underlying data source.

      Parameters:
      key - (undocumented)
      value - (undocumented)
      Returns:
      (undocumented)
      Since:
      2.0.0

    • options

      public DataStreamWriter<T> options(scala.collection.Map<String,String> options)
      (Scala-specific) Adds output options for the underlying data source.

      Parameters:
      options - (undocumented)
      Returns:
      (undocumented)
      Since:
      2.0.0

    • options

      public DataStreamWriter<T> options(Map<String,String> options)
      Adds output options for the underlying data source.

      Parameters:
      options - (undocumented)
      Returns:
      (undocumented)
      Since:
      2.0.0

    • start

      public StreamingQuery start(String path)
      Starts the execution of the streaming query, which will continually output results to the given path as new data arrives. The returned StreamingQuery object can be used to interact with the stream.

      Parameters:
      path - (undocumented)
      Returns:
      (undocumented)
      Since:
      2.0.0

    • start

      public StreamingQuery start() throws TimeoutException
      Starts the execution of the streaming query, which will continually output results to the given path as new data arrives. The returned StreamingQuery object can be used to interact with the stream. Throws a TimeoutException if the following conditions are met: - Another run of the same streaming query, that is a streaming query sharing the same checkpoint location, is already active on the same Spark Driver - The SQL configuration spark.sql.streaming.stopActiveRunOnRestart is enabled - The active run cannot be stopped within the timeout controlled by the SQL configuration spark.sql.streaming.stopTimeout

      Returns:
      (undocumented)
      Throws:
      TimeoutException
      Since:
      2.0.0

    • toTable

      public StreamingQuery toTable(String tableName) throws TimeoutException
      Starts the execution of the streaming query, which will continually output results to the given table as new data arrives. The returned StreamingQuery object can be used to interact with the stream.

      For v1 table, partitioning columns provided by partitionBy will be respected no matter the table exists or not. A new table will be created if the table not exists.

      For v2 table, partitionBy will be ignored if the table already exists. partitionBy will be respected only if the v2 table does not exist. Besides, the v2 table created by this API lacks some functionalities (e.g., customized properties, options, and serde info). If you need them, please create the v2 table manually before the execution to avoid creating a table with incomplete information.

      Parameters:
      tableName - (undocumented)
      Returns:
      (undocumented)
      Throws:
      TimeoutException
      Since:
      3.1.0

    • foreach

      public DataStreamWriter<T> foreach(ForeachWriter<T> writer)
      Sets the output of the streaming query to be processed using the provided writer object. object. See ForeachWriter for more details on the lifecycle and semantics.
      Parameters:
      writer - (undocumented)
      Returns:
      (undocumented)
      Since:
      2.0.0

    • foreachBatch

      public DataStreamWriter<T> foreachBatch(scala.Function2<Dataset<T>,Object,scala.runtime.BoxedUnit> function)
      :: Experimental ::

      (Scala-specific) Sets the output of the streaming query to be processed using the provided function. This is supported only in the micro-batch execution modes (that is, when the trigger is not continuous). In every micro-batch, the provided function will be called in every micro-batch with (i) the output rows as a Dataset and (ii) the batch identifier. The batchId can be used to deduplicate and transactionally write the output (that is, the provided Dataset) to external systems. The output Dataset is guaranteed to be exactly the same for the same batchId (assuming all operations are deterministic in the query).

      Parameters:
      function - (undocumented)
      Returns:
      (undocumented)
      Since:
      2.4.0

    • foreachBatch

      public DataStreamWriter<T> foreachBatch(VoidFunction2<Dataset<T>,Long> function)
      :: Experimental ::

      (Java-specific) Sets the output of the streaming query to be processed using the provided function. This is supported only in the micro-batch execution modes (that is, when the trigger is not continuous). In every micro-batch, the provided function will be called in every micro-batch with (i) the output rows as a Dataset and (ii) the batch identifier. The batchId can be used to deduplicate and transactionally write the output (that is, the provided Dataset) to external systems. The output Dataset is guaranteed to be exactly the same for the same batchId (assuming all operations are deterministic in the query).

      Parameters:
      function - (undocumented)
      Returns:
      (undocumented)
      Since:
      2.4.0