聊聊flink Table的Time Attributes

序

本文主要研究一下flink Table的Time Attributes

Processing time

經過fromDataStream定義

DataStream<Tuple2<String, String>> stream = ...;

// declare an additional logical field as a processing time attribute
Table table = tEnv.fromDataStream(stream, "Username, Data, UserActionTime.proctime");

WindowedTable windowedTable = table.window(Tumble.over("10.minutes").on("UserActionTime").as("userActionWindow"));

• 從DataStream建立Table的話，能夠在fromDataStream裏頭進行定義Processing time

經過TableSource定義

// define a table source with a processing attribute
public class UserActionSource implements StreamTableSource<Row>, DefinedProctimeAttribute {

    @Override
    public TypeInformation<Row> getReturnType() {
        String[] names = new String[] {"Username" , "Data"};
        TypeInformation[] types = new TypeInformation[] {Types.STRING(), Types.STRING()};
        return Types.ROW(names, types);
    }

    @Override
    public DataStream<Row> getDataStream(StreamExecutionEnvironment execEnv) {
        // create stream
        DataStream<Row> stream = ...;
        return stream;
    }

    @Override
    public String getProctimeAttribute() {
        // field with this name will be appended as a third field
        return "UserActionTime";
    }
}

// register table source
tEnv.registerTableSource("UserActions", new UserActionSource());

WindowedTable windowedTable = tEnv
    .scan("UserActions")
    .window(Tumble.over("10.minutes").on("UserActionTime").as("userActionWindow"));

• 經過TableSource建立Table的話，能夠經過實現DefinedProctimeAttribute接口來定義Processing time

Event time

經過fromDataStream定義

// Option 1:

// extract timestamp and assign watermarks based on knowledge of the stream
DataStream<Tuple2<String, String>> stream = inputStream.assignTimestampsAndWatermarks(...);

// declare an additional logical field as an event time attribute
Table table = tEnv.fromDataStream(stream, "Username, Data, UserActionTime.rowtime");


// Option 2:

// extract timestamp from first field, and assign watermarks based on knowledge of the stream
DataStream<Tuple3<Long, String, String>> stream = inputStream.assignTimestampsAndWatermarks(...);

// the first field has been used for timestamp extraction, and is no longer necessary
// replace first field with a logical event time attribute
Table table = tEnv.fromDataStream(stream, "UserActionTime.rowtime, Username, Data");

// Usage:

WindowedTable windowedTable = table.window(Tumble.over("10.minutes").on("UserActionTime").as("userActionWindow"));

• 從DataStream建立Table的話，能夠在fromDataStream裏頭進行定義Event time；具體有兩種方式，一種是額外定義一個字段，一種是覆蓋原有的字段

經過TableSource定義

// define a table source with a rowtime attribute
public class UserActionSource implements StreamTableSource<Row>, DefinedRowtimeAttributes {

    @Override
    public TypeInformation<Row> getReturnType() {
        String[] names = new String[] {"Username", "Data", "UserActionTime"};
        TypeInformation[] types =
            new TypeInformation[] {Types.STRING(), Types.STRING(), Types.LONG()};
        return Types.ROW(names, types);
    }

    @Override
    public DataStream<Row> getDataStream(StreamExecutionEnvironment execEnv) {
        // create stream
        // ...
        // assign watermarks based on the "UserActionTime" attribute
        DataStream<Row> stream = inputStream.assignTimestampsAndWatermarks(...);
        return stream;
    }

    @Override
    public List<RowtimeAttributeDescriptor> getRowtimeAttributeDescriptors() {
        // Mark the "UserActionTime" attribute as event-time attribute.
        // We create one attribute descriptor of "UserActionTime".
        RowtimeAttributeDescriptor rowtimeAttrDescr = new RowtimeAttributeDescriptor(
            "UserActionTime",
            new ExistingField("UserActionTime"),
            new AscendingTimestamps());
        List<RowtimeAttributeDescriptor> listRowtimeAttrDescr = Collections.singletonList(rowtimeAttrDescr);
        return listRowtimeAttrDescr;
    }
}

// register the table source
tEnv.registerTableSource("UserActions", new UserActionSource());

WindowedTable windowedTable = tEnv
    .scan("UserActions")
    .window(Tumble.over("10.minutes").on("UserActionTime").as("userActionWindow"));

• 經過TableSource建立Table的話，能夠經過實現DefinedRowtimeAttributes接口來定義Event time

definedTimeAttributes

flink-table_2.11-1.7.0-sources.jar!/org/apache/flink/table/sources/definedTimeAttributes.scala

/**
  * Extends a [[TableSource]] to specify a processing time attribute.
  */
trait DefinedProctimeAttribute {

  /**
    * Returns the name of a processing time attribute or null if no processing time attribute is
    * present.
    *
    * The referenced attribute must be present in the [[TableSchema]] of the [[TableSource]] and of
    * type [[Types.SQL_TIMESTAMP]].
    */
  @Nullable
  def getProctimeAttribute: String
}

/**
  * Extends a [[TableSource]] to specify rowtime attributes via a
  * [[RowtimeAttributeDescriptor]].
  */
trait DefinedRowtimeAttributes {

  /**
    * Returns a list of [[RowtimeAttributeDescriptor]] for all rowtime attributes of the table.
    *
    * All referenced attributes must be present in the [[TableSchema]] of the [[TableSource]] and of
    * type [[Types.SQL_TIMESTAMP]].
    *
    * @return A list of [[RowtimeAttributeDescriptor]].
    */
  def getRowtimeAttributeDescriptors: util.List[RowtimeAttributeDescriptor]
}

/**
  * Describes a rowtime attribute of a [[TableSource]].
  *
  * @param attributeName The name of the rowtime attribute.
  * @param timestampExtractor The timestamp extractor to derive the values of the attribute.
  * @param watermarkStrategy The watermark strategy associated with the attribute.
  */
class RowtimeAttributeDescriptor(
  val attributeName: String,
  val timestampExtractor: TimestampExtractor,
  val watermarkStrategy: WatermarkStrategy) {

  /** Returns the name of the rowtime attribute. */
  def getAttributeName: String = attributeName

  /** Returns the [[TimestampExtractor]] for the attribute. */
  def getTimestampExtractor: TimestampExtractor = timestampExtractor

  /** Returns the [[WatermarkStrategy]] for the attribute. */
  def getWatermarkStrategy: WatermarkStrategy = watermarkStrategy

  override def equals(other: Any): Boolean = other match {
    case that: RowtimeAttributeDescriptor =>
        Objects.equals(attributeName, that.attributeName) &&
        Objects.equals(timestampExtractor, that.timestampExtractor) &&
        Objects.equals(watermarkStrategy, that.watermarkStrategy)
    case _ => false
  }

  override def hashCode(): Int = {
    Objects.hash(attributeName, timestampExtractor, watermarkStrategy)
  }
}

• DefinedProctimeAttribute定義了getProctimeAttribute方法，返回String，用於定義Process time的字段名；DefinedRowtimeAttributes定義了getRowtimeAttributeDescriptors方法，返回的是RowtimeAttributeDescriptor的List，RowtimeAttributeDescriptor有3個屬性，分別是attributeName、timestampExtractor及watermarkStrategy

小結

• 在從DataStream或者TableSource建立Table時能夠指定Time Attributes，指定了以後就能夠做爲field來使用或者參與time-based的操做
• 針對Processing time，若是從DataStream建立Table的話，能夠在fromDataStream裏頭進行定義；經過TableSource建立Table的話，能夠經過實現DefinedProctimeAttribute接口來定義Processing time；DefinedProctimeAttribute定義了getProctimeAttribute方法，返回String，用於定義Process time的字段名
• 針對Event time，若是從DataStream建立Table的話，能夠在fromDataStream裏頭進行定義；具體有兩種方式，一種是額外定義一個字段，一種是覆蓋原有的字段；經過TableSource建立Table的話，能夠經過實現DefinedRowtimeAttributes接口來定義Event time；DefinedRowtimeAttributes定義了getRowtimeAttributeDescriptors方法，返回的是RowtimeAttributeDescriptor的List，RowtimeAttributeDescriptor有3個屬性，分別是attributeName、timestampExtractor及watermarkStrategy

doc

• Time Attributes