自定义函数

import org.apache.flink.table.api.*;
import org.apache.flink.table.functions.ScalarFunction;
import static org.apache.flink.table.api.Expressions.*;

// 定义函数逻辑
public static class SubstringFunction extends ScalarFunction {
  public String eval(String s, Integer begin, Integer end) {
    return s.substring(begin, end);
  }
}

TableEnvironment env = TableEnvironment.create(...);

// 在 Table API 里不经注册直接“内联”调用函数
env.from("MyTable").select(call(SubstringFunction.class, $("myField"), 5, 12));

// 注册函数
env.createTemporarySystemFunction("SubstringFunction", SubstringFunction.class);

// 在 Table API 里调用注册好的函数
env.from("MyTable").select(call("SubstringFunction", $("myField"), 5, 12));

// 在 SQL 里调用注册好的函数
env.sqlQuery("SELECT SubstringFunction(myField, 5, 12) FROM MyTable");

import org.apache.flink.table.api._
import org.apache.flink.table.functions.ScalarFunction

// define function logic
class SubstringFunction extends ScalarFunction {
  def eval(s: String, begin: Integer, end: Integer): String = {
    s.substring(begin, end)
  }
}

val env = TableEnvironment.create(...)

// 在 Table API 里不经注册直接“内联”调用函数
env.from("MyTable").select(call(classOf[SubstringFunction], $"myField", 5, 12))

// 注册函数
env.createTemporarySystemFunction("SubstringFunction", classOf[SubstringFunction])

// 在 Table API 里调用注册好的函数
env.from("MyTable").select(call("SubstringFunction", $"myField", 5, 12))

// 在 SQL 里调用注册好的函数
env.sqlQuery("SELECT SubstringFunction(myField, 5, 12) FROM MyTable")

import org.apache.flink.table.api.*;
import org.apache.flink.table.functions.ScalarFunction;
import static org.apache.flink.table.api.Expressions.*;

// 定义可参数化的函数逻辑
public static class SubstringFunction extends ScalarFunction {

  private boolean endInclusive;

  public SubstringFunction(boolean endInclusive) {
    this.endInclusive = endInclusive;
  }

  public String eval(String s, Integer begin, Integer end) {
    return s.substring(begin, endInclusive ? end + 1 : end);
  }
}

TableEnvironment env = TableEnvironment.create(...);

// 在 Table API 里不经注册直接“内联”调用函数
env.from("MyTable").select(call(new SubstringFunction(true), $("myField"), 5, 12));

// 注册函数
env.createTemporarySystemFunction("SubstringFunction", new SubstringFunction(true));

import org.apache.flink.table.api._
import org.apache.flink.table.functions.ScalarFunction

// 定义可参数化的函数逻辑
class SubstringFunction(val endInclusive) extends ScalarFunction {
  def eval(s: String, begin: Integer, end: Integer): String = {
    s.substring(endInclusive ? end + 1 : end)
  }
}

val env = TableEnvironment.create(...)

// 在 Table API 里不经注册直接“内联”调用函数
env.from("MyTable").select(call(new SubstringFunction(true), $"myField", 5, 12))

// 注册函数
env.createTemporarySystemFunction("SubstringFunction", new SubstringFunction(true))

import org.apache.flink.table.functions.ScalarFunction;

// 有多个重载求值方法的函数
public static class SumFunction extends ScalarFunction {

  public Integer eval(Integer a, Integer b) {
    return a + b;
  }

  public Integer eval(String a, String b) {
    return Integer.valueOf(a) + Integer.valueOf();
  }

  public Integer eval(Double... d) {
    double result = 0;
    for (double value : d)
      result += value;
    return (int) result;
  }
}

import org.apache.flink.table.functions.ScalarFunction
import scala.annotation.varargs

// 有多个重载求值方法的函数
class SumFunction extends ScalarFunction {

  def eval(a: Integer, b: Integer): Integer = {
    a + b
  }

  def eval(a: String, b: String): Integer = {
    Integer.valueOf(a) + Integer.valueOf(b)
  }

  @varargs // generate var-args like Java
  def eval(d: Double*): Integer = {
    d.sum.toInt
  }
}

import org.apache.flink.table.annotation.DataTypeHint;
import org.apache.flink.table.annotation.InputGroup;
import org.apache.flink.table.functions.ScalarFunction;
import org.apache.flink.types.Row;

// 有多个重载求值方法的函数
public static class OverloadedFunction extends ScalarFunction {

  // no hint required
  public Long eval(long a, long b) {
    return a + b;
  }

  // 定义 decimal 的精度和小数位
  public @DataTypeHint("DECIMAL(12, 3)") BigDecimal eval(double a, double b) {
    return BigDecimal.valueOf(a + b);
  }

  // 定义嵌套数据类型
  @DataTypeHint("ROW<s STRING, t TIMESTAMP(3) WITH LOCAL TIME ZONE>")
  public Row eval(int i) {
    return Row.of(String.valueOf(i), Instant.ofEpochSecond(i));
  }

  // 允许任意类型的符入，并输出序列化定制后的值
  @DataTypeHint(value = "RAW", bridgedTo = ByteBuffer.class)
  public ByteBuffer eval(@DataTypeHint(inputGroup = InputGroup.ANY) Object o) {
    return MyUtils.serializeToByteBuffer(o);
  }
}

import org.apache.flink.table.annotation.DataTypeHint
import org.apache.flink.table.annotation.InputGroup
import org.apache.flink.table.functions.ScalarFunction
import org.apache.flink.types.Row
import scala.annotation.varargs

// function with overloaded evaluation methods
class OverloadedFunction extends ScalarFunction {

  // no hint required
  def eval(a: Long, b: Long): Long = {
    a + b
  }

  // 定义 decimal 的精度和小数位
  @DataTypeHint("DECIMAL(12, 3)")
  def eval(double a, double b): BigDecimal = {
    java.lang.BigDecimal.valueOf(a + b)
  }

  // 定义嵌套数据类型
  @DataTypeHint("ROW<s STRING, t TIMESTAMP(3) WITH LOCAL TIME ZONE>")
  def eval(Int i): Row = {
    Row.of(java.lang.String.valueOf(i), java.time.Instant.ofEpochSecond(i))
  }

  // 允许任意类型的符入，并输出定制序列化后的值
  @DataTypeHint(value = "RAW", bridgedTo = classOf[java.nio.ByteBuffer])
  def eval(@DataTypeHint(inputGroup = InputGroup.ANY) Object o): java.nio.ByteBuffer = {
    MyUtils.serializeToByteBuffer(o)
  }
}

import org.apache.flink.table.annotation.DataTypeHint;
import org.apache.flink.table.annotation.FunctionHint;
import org.apache.flink.table.functions.TableFunction;
import org.apache.flink.types.Row;

// 为函数类的所有求值方法指定同一个输出类型
@FunctionHint(output = @DataTypeHint("ROW<s STRING, i INT>"))
public static class OverloadedFunction extends TableFunction<Row> {

  public void eval(int a, int b) {
    collect(Row.of("Sum", a + b));
  }

  // overloading of arguments is still possible
  public void eval() {
    collect(Row.of("Empty args", -1));
  }
}

// 解耦类型推导与求值方法，类型推导完全取决于 FunctionHint
@FunctionHint(
  input = {@DataTypeHint("INT"), @DataTypeHint("INT")},
  output = @DataTypeHint("INT")
)
@FunctionHint(
  input = {@DataTypeHint("BIGINT"), @DataTypeHint("BIGINT")},
  output = @DataTypeHint("BIGINT")
)
@FunctionHint(
  input = {},
  output = @DataTypeHint("BOOLEAN")
)
public static class OverloadedFunction extends TableFunction<Object> {

  // an implementer just needs to make sure that a method exists
  // that can be called by the JVM
  public void eval(Object... o) {
    if (o.length == 0) {
      collect(false);
    }
    collect(o[0]);
  }
}

import org.apache.flink.table.annotation.DataTypeHint
import org.apache.flink.table.annotation.FunctionHint
import org.apache.flink.table.functions.TableFunction
import org.apache.flink.types.Row

// 为函数类的所有求值方法指定同一个输出类型
@FunctionHint(output = new DataTypeHint("ROW<s STRING, i INT>"))
class OverloadedFunction extends TableFunction[Row] {

  def eval(a: Int, b: Int): Unit = {
    collect(Row.of("Sum", Int.box(a + b)))
  }

  // overloading of arguments is still possible
  def eval(): Unit = {
    collect(Row.of("Empty args", Int.box(-1)))
  }
}

// 解耦类型推导与求值方法，类型推导完全取决于 @FunctionHint
@FunctionHint(
  input = Array(new DataTypeHint("INT"), new DataTypeHint("INT")),
  output = new DataTypeHint("INT")
)
@FunctionHint(
  input = Array(new DataTypeHint("BIGINT"), new DataTypeHint("BIGINT")),
  output = new DataTypeHint("BIGINT")
)
@FunctionHint(
  input = Array(),
  output = new DataTypeHint("BOOLEAN")
)
class OverloadedFunction extends TableFunction[AnyRef] {

  // an implementer just needs to make sure that a method exists
  // that can be called by the JVM
  @varargs
  def eval(o: AnyRef*) = {
    if (o.length == 0) {
      collect(Boolean.box(false))
    }
    collect(o(0))
  }
}

import org.apache.flink.table.api.DataTypes;
import org.apache.flink.table.catalog.DataTypeFactory;
import org.apache.flink.table.functions.ScalarFunction;
import org.apache.flink.table.types.inference.TypeInference;
import org.apache.flink.types.Row;

public static class LiteralFunction extends ScalarFunction {
  public Object eval(String s, String type) {
    switch (type) {
      case "INT":
        return Integer.valueOf(s);
      case "DOUBLE":
        return Double.valueOf(s);
      case "STRING":
      default:
        return s;
    }
  }

  // 禁用自动的反射式类型推导，使用如下逻辑进行类型推导
  @Override
  public TypeInference getTypeInference(DataTypeFactory typeFactory) {
    return TypeInference.newBuilder()
      // 指定输入参数的类型，必要时参数会被隐式转换
      .typedArguments(DataTypes.STRING(), DataTypes.STRING())
      // specify a strategy for the result data type of the function
      .outputTypeStrategy(callContext -> {
        if (!callContext.isArgumentLiteral(1) || callContext.isArgumentNull(1)) {
          throw callContext.newValidationError("Literal expected for second argument.");
        }
        // 基于字符串值返回数据类型
        final String literal = callContext.getArgumentValue(1, String.class).orElse("STRING");
        switch (literal) {
          case "INT":
            return Optional.of(DataTypes.INT().notNull());
          case "DOUBLE":
            return Optional.of(DataTypes.DOUBLE().notNull());
          case "STRING":
          default:
            return Optional.of(DataTypes.STRING());
        }
      })
      .build();
  }
}

import org.apache.flink.table.api.*;
import org.apache.flink.table.functions.FunctionContext;
import org.apache.flink.table.functions.ScalarFunction;

public static class HashCodeFunction extends ScalarFunction {

    private int factor = 0;

    @Override
    public void open(FunctionContext context) throws Exception {
        // 获取参数 "hashcode_factor"
        // 如果不存在，则使用默认值 "12"
        factor = Integer.parseInt(context.getJobParameter("hashcode_factor", "12"));
    }

    public int eval(String s) {
        return s.hashCode() * factor;
    }
}

TableEnvironment env = TableEnvironment.create(...);

// 设置任务参数
env.getConfig().addJobParameter("hashcode_factor", "31");

// 注册函数
env.createTemporarySystemFunction("hashCode", HashCodeFunction.class);

// 调用函数
env.sqlQuery("SELECT myField, hashCode(myField) FROM MyTable");

import org.apache.flink.table.api._
import org.apache.flink.table.functions.FunctionContext
import org.apache.flink.table.functions.ScalarFunction

class HashCodeFunction extends ScalarFunction {

  private var factor: Int = 0

  override def open(context: FunctionContext): Unit = {
    // 获取参数 "hashcode_factor"
    // 如果不存在，则使用默认值 "12"
    factor = context.getJobParameter("hashcode_factor", "12").toInt
  }

  def eval(s: String): Int = {
    s.hashCode * factor
  }
}

val env = TableEnvironment.create(...)

// 设置任务参数
env.getConfig.addJobParameter("hashcode_factor", "31")

// 注册函数
env.createTemporarySystemFunction("hashCode", classOf[HashCodeFunction])

// 调用函数
env.sqlQuery("SELECT myField, hashCode(myField) FROM MyTable")

import org.apache.flink.table.annotation.InputGroup;
import org.apache.flink.table.api.*;
import org.apache.flink.table.functions.ScalarFunction;
import static org.apache.flink.table.api.Expressions.*;

public static class HashFunction extends ScalarFunction {

  // 接受任意类型输入，返回 INT 型输出
  public int eval(@DataTypeHint(inputGroup = InputGroup.ANY) Object o) {
    return o.hashCode();
  }
}

TableEnvironment env = TableEnvironment.create(...);

// 在 Table API 里不经注册直接“内联”调用函数
env.from("MyTable").select(call(HashFunction.class, $("myField")));

// 注册函数
env.createTemporarySystemFunction("HashFunction", HashFunction.class);

// 在 Table API 里调用注册好的函数
env.from("MyTable").select(call("HashFunction", $("myField")));

// 在 SQL 里调用注册好的函数
env.sqlQuery("SELECT HashFunction(myField) FROM MyTable");

import org.apache.flink.table.annotation.InputGroup
import org.apache.flink.table.api._
import org.apache.flink.table.functions.ScalarFunction

class HashFunction extends ScalarFunction {

  // 接受任意类型输入，返回 INT 型输出
  def eval(@DataTypeHint(inputGroup = InputGroup.ANY) o: AnyRef): Int {
    return o.hashCode();
  }
}

val env = TableEnvironment.create(...)

// 在 Table API 里不经注册直接“内联”调用函数
env.from("MyTable").select(call(classOf[HashFunction], $"myField"))

// 注册函数
env.createTemporarySystemFunction("HashFunction", classOf[HashFunction])

// 在 Table API 里调用注册好的函数
env.from("MyTable").select(call("HashFunction", $"myField"))

// 在 SQL 里调用注册好的函数
env.sqlQuery("SELECT HashFunction(myField) FROM MyTable")

import org.apache.flink.table.annotation.DataTypeHint;
import org.apache.flink.table.annotation.FunctionHint;
import org.apache.flink.table.api.*;
import org.apache.flink.table.functions.TableFunction;
import org.apache.flink.types.Row;
import static org.apache.flink.table.api.Expressions.*;

@FunctionHint(output = @DataTypeHint("ROW<word STRING, length INT>"))
public static class SplitFunction extends TableFunction<Row> {

  public void eval(String str) {
    for (String s : str.split(" ")) {
      // use collect(...) to emit a row
      collect(Row.of(s, s.length()));
    }
  }
}

TableEnvironment env = TableEnvironment.create(...);

// 在 Table API 里不经注册直接“内联”调用函数
env
  .from("MyTable")
  .joinLateral(call(SplitFunction.class, $("myField")))
  .select($("myField"), $("word"), $("length"));
env
  .from("MyTable")
  .leftOuterJoinLateral(call(SplitFunction.class, $("myField")))
  .select($("myField"), $("word"), $("length"));

// 在 Table API 里重命名函数字段
env
  .from("MyTable")
  .leftOuterJoinLateral(call(SplitFunction.class, $("myField")).as("newWord", "newLength"))
  .select($("myField"), $("newWord"), $("newLength"));

// 注册函数
env.createTemporarySystemFunction("SplitFunction", SplitFunction.class);

// 在 Table API 里调用注册好的函数
env
  .from("MyTable")
  .joinLateral(call("SplitFunction", $("myField")))
  .select($("myField"), $("word"), $("length"));
env
  .from("MyTable")
  .leftOuterJoinLateral(call("SplitFunction", $("myField")))
  .select($("myField"), $("word"), $("length"));

// 在 SQL 里调用注册好的函数
env.sqlQuery(
  "SELECT myField, word, length " +
  "FROM MyTable, LATERAL TABLE(SplitFunction(myField))");
env.sqlQuery(
  "SELECT myField, word, length " +
  "FROM MyTable " +
  "LEFT JOIN LATERAL TABLE(SplitFunction(myField)) ON TRUE");

// 在 SQL 里重命名函数字段
env.sqlQuery(
  "SELECT myField, newWord, newLength " +
  "FROM MyTable " +
  "LEFT JOIN LATERAL TABLE(SplitFunction(myField)) AS T(newWord, newLength) ON TRUE");

import org.apache.flink.table.annotation.DataTypeHint
import org.apache.flink.table.annotation.FunctionHint
import org.apache.flink.table.api._
import org.apache.flink.table.functions.TableFunction
import org.apache.flink.types.Row

@FunctionHint(output = new DataTypeHint("ROW<word STRING, length INT>"))
class SplitFunction extends TableFunction[Row] {

  def eval(str: String): Unit = {
    // use collect(...) to emit a row
    str.split(" ").foreach(s => collect(Row.of(s, Int.box(s.length))))
  }
}

val env = TableEnvironment.create(...)

// 在 Table API 里不经注册直接“内联”调用函数
env
  .from("MyTable")
  .joinLateral(call(classOf[SplitFunction], $"myField")
  .select($"myField", $"word", $"length")
env
  .from("MyTable")
  .leftOuterJoinLateral(call(classOf[SplitFunction], $"myField"))
  .select($"myField", $"word", $"length")

// 在 Table API 里重命名函数字段
env
  .from("MyTable")
  .leftOuterJoinLateral(call(classOf[SplitFunction], $"myField").as("newWord", "newLength"))
  .select($"myField", $"newWord", $"newLength")

// 注册函数
env.createTemporarySystemFunction("SplitFunction", classOf[SplitFunction])

// 在 Table API 里调用注册好的函数
env
  .from("MyTable")
  .joinLateral(call("SplitFunction", $"myField"))
  .select($"myField", $"word", $"length")
env
  .from("MyTable")
  .leftOuterJoinLateral(call("SplitFunction", $"myField"))
  .select($"myField", $"word", $"length")

// 在 SQL 里调用注册好的函数
env.sqlQuery(
  "SELECT myField, word, length " +
  "FROM MyTable, LATERAL TABLE(SplitFunction(myField))");
env.sqlQuery(
  "SELECT myField, word, length " +
  "FROM MyTable " +
  "LEFT JOIN LATERAL TABLE(SplitFunction(myField)) ON TRUE")

// 在 SQL 里重命名函数字段
env.sqlQuery(
  "SELECT myField, newWord, newLength " +
  "FROM MyTable " +
  "LEFT JOIN LATERAL TABLE(SplitFunction(myField)) AS T(newWord, newLength) ON TRUE")

/**
  * Base class for user-defined aggregates and table aggregates.
  *
  * @param <T>   the type of the aggregation result.
  * @param <ACC> the type of the aggregation accumulator. The accumulator is used to keep the
  *             aggregated values which are needed to compute an aggregation result.
  */
public abstract class UserDefinedAggregateFunction<T, ACC> extends UserDefinedFunction {

  /**
    * Creates and init the Accumulator for this (table)aggregate function.
    *
    * @return the accumulator with the initial value
    */
  public ACC createAccumulator(); // MANDATORY

  /**
    * Returns the TypeInformation of the (table)aggregate function's result.
    *
    * @return The TypeInformation of the (table)aggregate function's result or null if the result
    *         type should be automatically inferred.
    */
  public TypeInformation<T> getResultType = null; // PRE-DEFINED

  /**
    * Returns the TypeInformation of the (table)aggregate function's accumulator.
    *
    * @return The TypeInformation of the (table)aggregate function's accumulator or null if the
    *         accumulator type should be automatically inferred.
    */
  public TypeInformation<ACC> getAccumulatorType = null; // PRE-DEFINED
}

/**
  * Base class for aggregation functions.
  *
  * @param <T>   the type of the aggregation result
  * @param <ACC> the type of the aggregation accumulator. The accumulator is used to keep the
  *             aggregated values which are needed to compute an aggregation result.
  *             AggregateFunction represents its state using accumulator, thereby the state of the
  *             AggregateFunction must be put into the accumulator.
  */
public abstract class AggregateFunction<T, ACC> extends UserDefinedAggregateFunction<T, ACC> {

  /** Processes the input values and update the provided accumulator instance. The method
    * accumulate can be overloaded with different custom types and arguments. An AggregateFunction
    * requires at least one accumulate() method.
    *
    * @param accumulator           the accumulator which contains the current aggregated results
    * @param [user defined inputs] the input value (usually obtained from a new arrived data).
    */
  public void accumulate(ACC accumulator, [user defined inputs]); // MANDATORY

  /**
    * Retracts the input values from the accumulator instance. The current design assumes the
    * inputs are the values that have been previously accumulated. The method retract can be
    * overloaded with different custom types and arguments. This function must be implemented for
    * datastream bounded over aggregate.
    *
    * @param accumulator           the accumulator which contains the current aggregated results
    * @param [user defined inputs] the input value (usually obtained from a new arrived data).
    */
  public void retract(ACC accumulator, [user defined inputs]); // OPTIONAL

  /**
    * Merges a group of accumulator instances into one accumulator instance. This function must be
    * implemented for datastream session window grouping aggregate and dataset grouping aggregate.
    *
    * @param accumulator  the accumulator which will keep the merged aggregate results. It should
    *                     be noted that the accumulator may contain the previous aggregated
    *                     results. Therefore user should not replace or clean this instance in the
    *                     custom merge method.
    * @param its          an {@link java.lang.Iterable} pointed to a group of accumulators that will be
    *                     merged.
    */
  public void merge(ACC accumulator, java.lang.Iterable<ACC> its); // OPTIONAL

  /**
    * Called every time when an aggregation result should be materialized.
    * The returned value could be either an early and incomplete result
    * (periodically emitted as data arrive) or the final result of the
    * aggregation.
    *
    * @param accumulator the accumulator which contains the current
    *                    aggregated results
    * @return the aggregation result
    */
  public T getValue(ACC accumulator); // MANDATORY

  /**
    * Resets the accumulator for this [[AggregateFunction]]. This function must be implemented for
    * dataset grouping aggregate.
    *
    * @param accumulator  the accumulator which needs to be reset
    */
  public void resetAccumulator(ACC accumulator); // OPTIONAL

  /**
    * Returns true if this AggregateFunction can only be applied in an OVER window.
    *
    * @return true if the AggregateFunction requires an OVER window, false otherwise.
    */
  public Boolean requiresOver = false; // PRE-DEFINED
}

/**
  * Base class for user-defined aggregates and table aggregates.
  *
  * @tparam T   the type of the aggregation result.
  * @tparam ACC the type of the aggregation accumulator. The accumulator is used to keep the
  *             aggregated values which are needed to compute an aggregation result.
  */
abstract class UserDefinedAggregateFunction[T, ACC] extends UserDefinedFunction {

  /**
    * Creates and init the Accumulator for this (table)aggregate function.
    *
    * @return the accumulator with the initial value
    */
  def createAccumulator(): ACC // MANDATORY

  /**
    * Returns the TypeInformation of the (table)aggregate function's result.
    *
    * @return The TypeInformation of the (table)aggregate function's result or null if the result
    *         type should be automatically inferred.
    */
  def getResultType: TypeInformation[T] = null // PRE-DEFINED

  /**
    * Returns the TypeInformation of the (table)aggregate function's accumulator.
    *
    * @return The TypeInformation of the (table)aggregate function's accumulator or null if the
    *         accumulator type should be automatically inferred.
    */
  def getAccumulatorType: TypeInformation[ACC] = null // PRE-DEFINED
}

/**
  * Base class for aggregation functions.
  *
  * @tparam T   the type of the aggregation result
  * @tparam ACC the type of the aggregation accumulator. The accumulator is used to keep the
  *             aggregated values which are needed to compute an aggregation result.
  *             AggregateFunction represents its state using accumulator, thereby the state of the
  *             AggregateFunction must be put into the accumulator.
  */
abstract class AggregateFunction[T, ACC] extends UserDefinedAggregateFunction[T, ACC] {

  /**
    * Processes the input values and update the provided accumulator instance. The method
    * accumulate can be overloaded with different custom types and arguments. An AggregateFunction
    * requires at least one accumulate() method.
    *
    * @param accumulator           the accumulator which contains the current aggregated results
    * @param [user defined inputs] the input value (usually obtained from a new arrived data).
    */
  def accumulate(accumulator: ACC, [user defined inputs]): Unit // MANDATORY

  /**
    * Retracts the input values from the accumulator instance. The current design assumes the
    * inputs are the values that have been previously accumulated. The method retract can be
    * overloaded with different custom types and arguments. This function must be implemented for
    * datastream bounded over aggregate.
    *
    * @param accumulator           the accumulator which contains the current aggregated results
    * @param [user defined inputs] the input value (usually obtained from a new arrived data).
    */
  def retract(accumulator: ACC, [user defined inputs]): Unit // OPTIONAL

  /**
    * Merges a group of accumulator instances into one accumulator instance. This function must be
    * implemented for datastream session window grouping aggregate and dataset grouping aggregate.
    *
    * @param accumulator  the accumulator which will keep the merged aggregate results. It should
    *                     be noted that the accumulator may contain the previous aggregated
    *                     results. Therefore user should not replace or clean this instance in the
    *                     custom merge method.
    * @param its          an [[java.lang.Iterable]] pointed to a group of accumulators that will be
    *                     merged.
    */
  def merge(accumulator: ACC, its: java.lang.Iterable[ACC]): Unit // OPTIONAL

  /**
    * Called every time when an aggregation result should be materialized.
    * The returned value could be either an early and incomplete result
    * (periodically emitted as data arrive) or the final result of the
    * aggregation.
    *
    * @param accumulator the accumulator which contains the current
    *                    aggregated results
    * @return the aggregation result
    */
  def getValue(accumulator: ACC): T // MANDATORY

  /**
    * Resets the accumulator for this [[AggregateFunction]]. This function must be implemented for
    * dataset grouping aggregate.
    *
    * @param accumulator  the accumulator which needs to be reset
    */
  def resetAccumulator(accumulator: ACC): Unit // OPTIONAL

  /**
    * Returns true if this AggregateFunction can only be applied in an OVER window.
    *
    * @return true if the AggregateFunction requires an OVER window, false otherwise.
    */
  def requiresOver: Boolean = false // PRE-DEFINED
}

/**
 * Accumulator for WeightedAvg.
 */
public static class WeightedAvgAccum {
    public long sum = 0;
    public int count = 0;
}

/**
 * Weighted Average user-defined aggregate function.
 */
public static class WeightedAvg extends AggregateFunction<Long, WeightedAvgAccum> {

    @Override
    public WeightedAvgAccum createAccumulator() {
        return new WeightedAvgAccum();
    }

    @Override
    public Long getValue(WeightedAvgAccum acc) {
        if (acc.count == 0) {
            return null;
        } else {
            return acc.sum / acc.count;
        }
    }

    public void accumulate(WeightedAvgAccum acc, long iValue, int iWeight) {
        acc.sum += iValue * iWeight;
        acc.count += iWeight;
    }

    public void retract(WeightedAvgAccum acc, long iValue, int iWeight) {
        acc.sum -= iValue * iWeight;
        acc.count -= iWeight;
    }

    public void merge(WeightedAvgAccum acc, Iterable<WeightedAvgAccum> it) {
        Iterator<WeightedAvgAccum> iter = it.iterator();
        while (iter.hasNext()) {
            WeightedAvgAccum a = iter.next();
            acc.count += a.count;
            acc.sum += a.sum;
        }
    }

    public void resetAccumulator(WeightedAvgAccum acc) {
        acc.count = 0;
        acc.sum = 0L;
    }
}

// 注册函数
StreamTableEnvironment tEnv = ...
tEnv.registerFunction("wAvg", new WeightedAvg());

// 使用函数
tEnv.sqlQuery("SELECT user, wAvg(points, level) AS avgPoints FROM userScores GROUP BY user");

import java.lang.{Long => JLong, Integer => JInteger}
import org.apache.flink.api.java.tuple.{Tuple1 => JTuple1}
import org.apache.flink.api.java.typeutils.TupleTypeInfo
import org.apache.flink.table.api.Types
import org.apache.flink.table.functions.AggregateFunction

/**
 * Accumulator for WeightedAvg.
 */
class WeightedAvgAccum extends JTuple1[JLong, JInteger] {
  sum = 0L
  count = 0
}

/**
 * Weighted Average user-defined aggregate function.
 */
class WeightedAvg extends AggregateFunction[JLong, CountAccumulator] {

  override def createAccumulator(): WeightedAvgAccum = {
    new WeightedAvgAccum
  }

  override def getValue(acc: WeightedAvgAccum): JLong = {
    if (acc.count == 0) {
        null
    } else {
        acc.sum / acc.count
    }
  }

  def accumulate(acc: WeightedAvgAccum, iValue: JLong, iWeight: JInteger): Unit = {
    acc.sum += iValue * iWeight
    acc.count += iWeight
  }

  def retract(acc: WeightedAvgAccum, iValue: JLong, iWeight: JInteger): Unit = {
    acc.sum -= iValue * iWeight
    acc.count -= iWeight
  }

  def merge(acc: WeightedAvgAccum, it: java.lang.Iterable[WeightedAvgAccum]): Unit = {
    val iter = it.iterator()
    while (iter.hasNext) {
      val a = iter.next()
      acc.count += a.count
      acc.sum += a.sum
    }
  }

  def resetAccumulator(acc: WeightedAvgAccum): Unit = {
    acc.count = 0
    acc.sum = 0L
  }

  override def getAccumulatorType: TypeInformation[WeightedAvgAccum] = {
    new TupleTypeInfo(classOf[WeightedAvgAccum], Types.LONG, Types.INT)
  }

  override def getResultType: TypeInformation[JLong] = Types.LONG
}

// 注册函数
val tEnv: StreamTableEnvironment = ???
tEnv.registerFunction("wAvg", new WeightedAvg())

// 使用函数
tEnv.sqlQuery("SELECT user, wAvg(points, level) AS avgPoints FROM userScores GROUP BY user")

'''
Java code:

/**
 * Accumulator for WeightedAvg.
 */
public static class WeightedAvgAccum {
    public long sum = 0;
    public int count = 0;
}

// The java class must have a public no-argument constructor and can be founded in current java classloader.
// Java 类必须有一个 public 的无参构造函数，并且可以在当前类加载器中加载到。

/**
 * Weighted Average user-defined aggregate function.
 */
public static class WeightedAvg extends AggregateFunction<Long, WeightedAvgAccum> {

    @Override
    public WeightedAvgAccum createAccumulator() {
        return new WeightedAvgAccum();
    }

    @Override
    public Long getValue(WeightedAvgAccum acc) {
        if (acc.count == 0) {
            return null;
        } else {
            return acc.sum / acc.count;
        }
    }

    public void accumulate(WeightedAvgAccum acc, long iValue, int iWeight) {
        acc.sum += iValue * iWeight;
        acc.count += iWeight;
    }

    public void retract(WeightedAvgAccum acc, long iValue, int iWeight) {
        acc.sum -= iValue * iWeight;
        acc.count -= iWeight;
    }

    public void merge(WeightedAvgAccum acc, Iterable<WeightedAvgAccum> it) {
        Iterator<WeightedAvgAccum> iter = it.iterator();
        while (iter.hasNext()) {
            WeightedAvgAccum a = iter.next();
            acc.count += a.count;
            acc.sum += a.sum;
        }
    }

    public void resetAccumulator(WeightedAvgAccum acc) {
        acc.count = 0;
        acc.sum = 0L;
    }
}
'''

# 注册函数
t_env = ...  # type: StreamTableEnvironment
t_env.register_java_function("wAvg", "my.java.function.WeightedAvg")

# 使用函数
t_env.sql_query("SELECT user, wAvg(points, level) AS avgPoints FROM userScores GROUP BY user")

/**
  * Base class for user-defined aggregates and table aggregates.
  *
  * @param <T>   the type of the aggregation result.
  * @param <ACC> the type of the aggregation accumulator. The accumulator is used to keep the
  *             aggregated values which are needed to compute an aggregation result.
  */
public abstract class UserDefinedAggregateFunction<T, ACC> extends UserDefinedFunction {

  /**
    * Creates and init the Accumulator for this (table)aggregate function.
    *
    * @return the accumulator with the initial value
    */
  public ACC createAccumulator(); // MANDATORY

  /**
    * Returns the TypeInformation of the (table)aggregate function's result.
    *
    * @return The TypeInformation of the (table)aggregate function's result or null if the result
    *         type should be automatically inferred.
    */
  public TypeInformation<T> getResultType = null; // PRE-DEFINED

  /**
    * Returns the TypeInformation of the (table)aggregate function's accumulator.
    *
    * @return The TypeInformation of the (table)aggregate function's accumulator or null if the
    *         accumulator type should be automatically inferred.
    */
  public TypeInformation<ACC> getAccumulatorType = null; // PRE-DEFINED
}

/**
  * Base class for table aggregation functions.
  *
  * @param <T>   the type of the aggregation result
  * @param <ACC> the type of the aggregation accumulator. The accumulator is used to keep the
  *             aggregated values which are needed to compute a table aggregation result.
  *             TableAggregateFunction represents its state using accumulator, thereby the state of
  *             the TableAggregateFunction must be put into the accumulator.
  */
public abstract class TableAggregateFunction<T, ACC> extends UserDefinedAggregateFunction<T, ACC> {

  /** Processes the input values and update the provided accumulator instance. The method
    * accumulate can be overloaded with different custom types and arguments. A TableAggregateFunction
    * requires at least one accumulate() method.
    *
    * @param accumulator           the accumulator which contains the current aggregated results
    * @param [user defined inputs] the input value (usually obtained from a new arrived data).
    */
  public void accumulate(ACC accumulator, [user defined inputs]); // MANDATORY

  /**
    * Retracts the input values from the accumulator instance. The current design assumes the
    * inputs are the values that have been previously accumulated. The method retract can be
    * overloaded with different custom types and arguments. This function must be implemented for
    * datastream bounded over aggregate.
    *
    * @param accumulator           the accumulator which contains the current aggregated results
    * @param [user defined inputs] the input value (usually obtained from a new arrived data).
    */
  public void retract(ACC accumulator, [user defined inputs]); // OPTIONAL

  /**
    * Merges a group of accumulator instances into one accumulator instance. This function must be
    * implemented for datastream session window grouping aggregate and dataset grouping aggregate.
    *
    * @param accumulator  the accumulator which will keep the merged aggregate results. It should
    *                     be noted that the accumulator may contain the previous aggregated
    *                     results. Therefore user should not replace or clean this instance in the
    *                     custom merge method.
    * @param its          an {@link java.lang.Iterable} pointed to a group of accumulators that will be
    *                     merged.
    */
  public void merge(ACC accumulator, java.lang.Iterable<ACC> its); // OPTIONAL

  /**
    * Called every time when an aggregation result should be materialized. The returned value
    * could be either an early and incomplete result  (periodically emitted as data arrive) or
    * the final result of the  aggregation.
    *
    * @param accumulator the accumulator which contains the current
    *                    aggregated results
    * @param out         the collector used to output data
    */
  public void emitValue(ACC accumulator, Collector<T> out); // OPTIONAL

  /**
    * Called every time when an aggregation result should be materialized. The returned value
    * could be either an early and incomplete result (periodically emitted as data arrive) or
    * the final result of the aggregation.
    *
    * Different from emitValue, emitUpdateWithRetract is used to emit values that have been updated.
    * This method outputs data incrementally in retract mode, i.e., once there is an update, we
    * have to retract old records before sending new updated ones. The emitUpdateWithRetract
    * method will be used in preference to the emitValue method if both methods are defined in the
    * table aggregate function, because the method is treated to be more efficient than emitValue
    * as it can outputvalues incrementally.
    *
    * @param accumulator the accumulator which contains the current
    *                    aggregated results
    * @param out         the retractable collector used to output data. Use collect method
    *                    to output(add) records and use retract method to retract(delete)
    *                    records.
    */
  public void emitUpdateWithRetract(ACC accumulator, RetractableCollector<T> out); // OPTIONAL

  /**
    * Collects a record and forwards it. The collector can output retract messages with the retract
    * method. Note: only use it in {@code emitRetractValueIncrementally}.
    */
  public interface RetractableCollector<T> extends Collector<T> {

      /**
        * Retract a record.
        *
        * @param record The record to retract.
        */
      void retract(T record);
  }
}

/**
  * Base class for user-defined aggregates and table aggregates.
  *
  * @tparam T   the type of the aggregation result.
  * @tparam ACC the type of the aggregation accumulator. The accumulator is used to keep the
  *             aggregated values which are needed to compute an aggregation result.
  */
abstract class UserDefinedAggregateFunction[T, ACC] extends UserDefinedFunction {

  /**
    * Creates and init the Accumulator for this (table)aggregate function.
    *
    * @return the accumulator with the initial value
    */
  def createAccumulator(): ACC // MANDATORY

  /**
    * Returns the TypeInformation of the (table)aggregate function's result.
    *
    * @return The TypeInformation of the (table)aggregate function's result or null if the result
    *         type should be automatically inferred.
    */
  def getResultType: TypeInformation[T] = null // PRE-DEFINED

  /**
    * Returns the TypeInformation of the (table)aggregate function's accumulator.
    *
    * @return The TypeInformation of the (table)aggregate function's accumulator or null if the
    *         accumulator type should be automatically inferred.
    */
  def getAccumulatorType: TypeInformation[ACC] = null // PRE-DEFINED
}

/**
  * Base class for table aggregation functions.
  *
  * @tparam T   the type of the aggregation result
  * @tparam ACC the type of the aggregation accumulator. The accumulator is used to keep the
  *             aggregated values which are needed to compute an aggregation result.
  *             TableAggregateFunction represents its state using accumulator, thereby the state of
  *             the TableAggregateFunction must be put into the accumulator.
  */
abstract class TableAggregateFunction[T, ACC] extends UserDefinedAggregateFunction[T, ACC] {

  /**
    * Processes the input values and update the provided accumulator instance. The method
    * accumulate can be overloaded with different custom types and arguments. A TableAggregateFunction
    * requires at least one accumulate() method.
    *
    * @param accumulator           the accumulator which contains the current aggregated results
    * @param [user defined inputs] the input value (usually obtained from a new arrived data).
    */
  def accumulate(accumulator: ACC, [user defined inputs]): Unit // MANDATORY

  /**
    * Retracts the input values from the accumulator instance. The current design assumes the
    * inputs are the values that have been previously accumulated. The method retract can be
    * overloaded with different custom types and arguments. This function must be implemented for
    * datastream bounded over aggregate.
    *
    * @param accumulator           the accumulator which contains the current aggregated results
    * @param [user defined inputs] the input value (usually obtained from a new arrived data).
    */
  def retract(accumulator: ACC, [user defined inputs]): Unit // OPTIONAL

  /**
    * Merges a group of accumulator instances into one accumulator instance. This function must be
    * implemented for datastream session window grouping aggregate and dataset grouping aggregate.
    *
    * @param accumulator  the accumulator which will keep the merged aggregate results. It should
    *                     be noted that the accumulator may contain the previous aggregated
    *                     results. Therefore user should not replace or clean this instance in the
    *                     custom merge method.
    * @param its          an [[java.lang.Iterable]] pointed to a group of accumulators that will be
    *                     merged.
    */
  def merge(accumulator: ACC, its: java.lang.Iterable[ACC]): Unit // OPTIONAL

  /**
    * Called every time when an aggregation result should be materialized. The returned value
    * could be either an early and incomplete result  (periodically emitted as data arrive) or
    * the final result of the  aggregation.
    *
    * @param accumulator the accumulator which contains the current
    *                    aggregated results
    * @param out         the collector used to output data
    */
  def emitValue(accumulator: ACC, out: Collector[T]): Unit // OPTIONAL

  /**
    * Called every time when an aggregation result should be materialized. The returned value
    * could be either an early and incomplete result (periodically emitted as data arrive) or
    * the final result of the aggregation.
    *
    * Different from emitValue, emitUpdateWithRetract is used to emit values that have been updated.
    * This method outputs data incrementally in retract mode, i.e., once there is an update, we
    * have to retract old records before sending new updated ones. The emitUpdateWithRetract
    * method will be used in preference to the emitValue method if both methods are defined in the
    * table aggregate function, because the method is treated to be more efficient than emitValue
    * as it can outputvalues incrementally.
    *
    * @param accumulator the accumulator which contains the current
    *                    aggregated results
    * @param out         the retractable collector used to output data. Use collect method
    *                    to output(add) records and use retract method to retract(delete)
    *                    records.
    */
  def emitUpdateWithRetract(accumulator: ACC, out: RetractableCollector[T]): Unit // OPTIONAL

  /**
    * Collects a record and forwards it. The collector can output retract messages with the retract
    * method. Note: only use it in `emitRetractValueIncrementally`.
    */
  trait RetractableCollector[T] extends Collector[T] {

    /**
      * Retract a record.
      *
      * @param record The record to retract.
      */
    def retract(record: T): Unit
  }
}

/**
 * Accumulator for Top2.
 */
public class Top2Accum {
    public Integer first;
    public Integer second;
}

/**
 * The top2 user-defined table aggregate function.
 */
public static class Top2 extends TableAggregateFunction<Tuple2<Integer, Integer>, Top2Accum> {

    @Override
    public Top2Accum createAccumulator() {
        Top2Accum acc = new Top2Accum();
        acc.first = Integer.MIN_VALUE;
        acc.second = Integer.MIN_VALUE;
        return acc;
    }


    public void accumulate(Top2Accum acc, Integer v) {
        if (v > acc.first) {
            acc.second = acc.first;
            acc.first = v;
        } else if (v > acc.second) {
            acc.second = v;
        }
    }

    public void merge(Top2Accum acc, java.lang.Iterable<Top2Accum> iterable) {
        for (Top2Accum otherAcc : iterable) {
            accumulate(acc, otherAcc.first);
            accumulate(acc, otherAcc.second);
        }
    }

    public void emitValue(Top2Accum acc, Collector<Tuple2<Integer, Integer>> out) {
        // emit the value and rank
        if (acc.first != Integer.MIN_VALUE) {
            out.collect(Tuple2.of(acc.first, 1));
        }
        if (acc.second != Integer.MIN_VALUE) {
            out.collect(Tuple2.of(acc.second, 2));
        }
    }
}

// 注册函数
StreamTableEnvironment tEnv = ...
tEnv.registerFunction("top2", new Top2());

// 初始化表
Table tab = ...;

// 使用函数
tab.groupBy("key")
    .flatAggregate("top2(a) as (v, rank)")
    .select("key, v, rank");

import java.lang.{Integer => JInteger}
import org.apache.flink.table.api.Types
import org.apache.flink.table.functions.TableAggregateFunction

/**
 * Accumulator for top2.
 */
class Top2Accum {
  var first: JInteger = _
  var second: JInteger = _
}

/**
 * The top2 user-defined table aggregate function.
 */
class Top2 extends TableAggregateFunction[JTuple2[JInteger, JInteger], Top2Accum] {

  override def createAccumulator(): Top2Accum = {
    val acc = new Top2Accum
    acc.first = Int.MinValue
    acc.second = Int.MinValue
    acc
  }

  def accumulate(acc: Top2Accum, v: Int) {
    if (v > acc.first) {
      acc.second = acc.first
      acc.first = v
    } else if (v > acc.second) {
      acc.second = v
    }
  }

  def merge(acc: Top2Accum, its: JIterable[Top2Accum]): Unit = {
    val iter = its.iterator()
    while (iter.hasNext) {
      val top2 = iter.next()
      accumulate(acc, top2.first)
      accumulate(acc, top2.second)
    }
  }

  def emitValue(acc: Top2Accum, out: Collector[JTuple2[JInteger, JInteger]]): Unit = {
    // emit the value and rank
    if (acc.first != Int.MinValue) {
      out.collect(JTuple2.of(acc.first, 1))
    }
    if (acc.second != Int.MinValue) {
      out.collect(JTuple2.of(acc.second, 2))
    }
  }
}

// 初始化表
val tab = ...

// 使用函数
tab
  .groupBy('key)
  .flatAggregate(top2('a) as ('v, 'rank))
  .select('key, 'v, 'rank)

/**
 * Accumulator for Top2.
 */
public class Top2Accum {
    public Integer first;
    public Integer second;
    public Integer oldFirst;
    public Integer oldSecond;
}

/**
 * The top2 user-defined table aggregate function.
 */
public static class Top2 extends TableAggregateFunction<Tuple2<Integer, Integer>, Top2Accum> {

    @Override
    public Top2Accum createAccumulator() {
        Top2Accum acc = new Top2Accum();
        acc.first = Integer.MIN_VALUE;
        acc.second = Integer.MIN_VALUE;
        acc.oldFirst = Integer.MIN_VALUE;
        acc.oldSecond = Integer.MIN_VALUE;
        return acc;
    }

    public void accumulate(Top2Accum acc, Integer v) {
        if (v > acc.first) {
            acc.second = acc.first;
            acc.first = v;
        } else if (v > acc.second) {
            acc.second = v;
        }
    }

    public void emitUpdateWithRetract(Top2Accum acc, RetractableCollector<Tuple2<Integer, Integer>> out) {
        if (!acc.first.equals(acc.oldFirst)) {
            // if there is an update, retract old value then emit new value.
            if (acc.oldFirst != Integer.MIN_VALUE) {
                out.retract(Tuple2.of(acc.oldFirst, 1));
            }
            out.collect(Tuple2.of(acc.first, 1));
            acc.oldFirst = acc.first;
        }

        if (!acc.second.equals(acc.oldSecond)) {
            // if there is an update, retract old value then emit new value.
            if (acc.oldSecond != Integer.MIN_VALUE) {
                out.retract(Tuple2.of(acc.oldSecond, 2));
            }
            out.collect(Tuple2.of(acc.second, 2));
            acc.oldSecond = acc.second;
        }
    }
}

// 注册函数
StreamTableEnvironment tEnv = ...
tEnv.registerFunction("top2", new Top2());

// 初始化表
Table tab = ...;

// 使用函数
tab.groupBy("key")
    .flatAggregate("top2(a) as (v, rank)")
    .select("key, v, rank");

import java.lang.{Integer => JInteger}
import org.apache.flink.table.api.Types
import org.apache.flink.table.functions.TableAggregateFunction

/**
 * Accumulator for top2.
 */
class Top2Accum {
  var first: JInteger = _
  var second: JInteger = _
  var oldFirst: JInteger = _
  var oldSecond: JInteger = _
}

/**
 * The top2 user-defined table aggregate function.
 */
class Top2 extends TableAggregateFunction[JTuple2[JInteger, JInteger], Top2Accum] {

  override def createAccumulator(): Top2Accum = {
    val acc = new Top2Accum
    acc.first = Int.MinValue
    acc.second = Int.MinValue
    acc.oldFirst = Int.MinValue
    acc.oldSecond = Int.MinValue
    acc
  }

  def accumulate(acc: Top2Accum, v: Int) {
    if (v > acc.first) {
      acc.second = acc.first
      acc.first = v
    } else if (v > acc.second) {
      acc.second = v
    }
  }

  def emitUpdateWithRetract(
    acc: Top2Accum,
    out: RetractableCollector[JTuple2[JInteger, JInteger]])
  : Unit = {
    if (acc.first != acc.oldFirst) {
      // if there is an update, retract old value then emit new value.
      if (acc.oldFirst != Int.MinValue) {
        out.retract(JTuple2.of(acc.oldFirst, 1))
      }
      out.collect(JTuple2.of(acc.first, 1))
      acc.oldFirst = acc.first
    }
    if (acc.second != acc.oldSecond) {
      // if there is an update, retract old value then emit new value.
      if (acc.oldSecond != Int.MinValue) {
        out.retract(JTuple2.of(acc.oldSecond, 2))
      }
      out.collect(JTuple2.of(acc.second, 2))
      acc.oldSecond = acc.second
    }
  }
}

// 初始化表
val tab = ...

// 使用函数
tab
  .groupBy('key)
  .flatAggregate(top2('a) as ('v, 'rank))
  .select('key, 'v, 'rank)