External Java UDFs

This article provides a step-by-step guide for installing the RisingWave Java UDF SDK, defining functions using Java, starting a Java process as a UDF server, and declaring and using UDFs in RisingWave.

Prerequisites

• Ensure that you have Java Developer’s Kit (JDK) (11 or later) installed on your computer.
• Ensure that you have Apache Maven (3.0 or later) installed on your computer. Maven is a build tool that helps manage Java projects and dependencies.

1. Create a Maven project from template

The RisingWave Java UDF SDK is distributed as a Maven artifact. We have prepared a sample project so you don’t have to create it from scratch. Run the following command to clone the template repository.

git clone https://github.com/risingwavelabs/risingwave-java-udf-template.git

I'd like to start from scratch

To create a new project using the RisingWave Java UDF SDK, follow these steps:

Generate a new Maven project:

mvn archetype:generate -DgroupId=com.example -DartifactId=udf-example -DarchetypeArtifactId=maven-archetype-quickstart -DarchetypeVersion=1.4 -DinteractiveMode=false

Configure your pom.xml file as follows:

<?xml version="1.0" encoding="UTF-8"?>
<project xmlns="http://maven.apache.org/POM/4.0.0" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
    xsi:schemaLocation="http://maven.apache.org/POM/4.0.0 http://maven.apache.org/xsd/maven-4.0.0.xsd">
    <modelVersion>4.0.0</modelVersion>
    <groupId>com.example</groupId>
    <artifactId>udf-example</artifactId>
    <version>1.0-SNAPSHOT</version>

    <dependencies>
        <dependency>
            <groupId>com.risingwave.java</groupId>
            <artifactId>risingwave-udf</artifactId>
            <version>0.1.1</version>
        </dependency>
    </dependencies>
</project>

The --add-opens flag must be added when running unit tests through Maven:

<build>
    <plugins>
        <plugin>
            <groupId>org.apache.maven.plugins</groupId>
            <artifactId>maven-surefire-plugin</artifactId>
            <version>3.0.0-M7</version>
            <configuration>
                <argLine>--add-opens=java.base/java.nio=ALL-UNNAMED</argLine>
            </configuration>
        </plugin>
    </plugins>
</build>

2. Define your functions in Java

Scalar functions

A user-defined scalar function maps zero, one, or multiple scalar values to a new scalar value.

In order to define a scalar function, you have to create a new class that implements the ScalarFunctioninterface in com.risingwave.functions and implement exactly one evaluation method named eval(...). This method must be declared public and non-static.

Any data type listed in Data type mapping can be used as a parameter or return type of an evaluation method.

Here’s an example of a scalar function that calculates the greatest common divisor (GCD) of two integers:

import com.risingwave.functions.ScalarFunction;

public class Gcd implements ScalarFunction {
    public int eval(int a, int b) {
        while (b != 0) {
            int temp = b;
            b = a % b;
            a = temp;
        }
        return a;
    }
}

DIFFERENCES WITH FLINK

• The ScalarFunction is an interface instead of an abstract class.
• Multiple overloaded eval methods are not supported.
• Variable arguments such as eval(Integer...) are not supported.

Table functions

A user-defined table function maps zero, one, or multiple scalar values to one or multiple rows (structured types).

In order to define a table function, you have to create a new class that implements the TableFunctioninterface in com.risingwave.functions and implement exactly one evaluation method named eval(...). This method must be declared public and non-static.

The return type must be an Iterator of any data type listed in Data type mapping.

Similar to scalar functions, input and output data types are automatically extracted using reflection. This includes the generic argument T of the return value for determining an output data type.

Here’s an example of a table function that generates a series of integers:

import com.risingwave.functions.TableFunction;

public class Series implements TableFunction {
    public Iterator<Integer> eval(int n) {
        return java.util.stream.IntStream.range(0, n).iterator();
    }
}

DIFFERENCES WITH FLINK

• The TableFunction is an interface instead of an abstract class. It has no generic arguments.
• Instead of calling collect to emit a row, the eval method returns an Iterator of the output rows.
• Multiple overloaded eval methods are not supported.
• Variable arguments such as eval(Integer...) are not supported.
• In SQL, table functions can be used in the FROM clause directly. JOIN LATERAL TABLE is not supported.

3. Start a UDF server

Run the following command to create a UDF server and register for the functions you defined.

import com.risingwave.functions.UdfServer;

public class App {
    public static void main(String[] args) {
        try (var server = new UdfServer("0.0.0.0", 8815)) {
            // Register functions
            server.addFunction("gcd", new Gcd());
            server.addFunction("series", new Series());
            // Start the server
            server.start();
            server.awaitTermination();
        } catch (Exception e) {
            e.printStackTrace();
        }
    }
}

Run the following command start the UDF server.

_JAVA_OPTIONS="--add-opens=java.base/java.nio=ALL-UNNAMED" mvn exec:java -Dexec.mainClass="com.example.App"

The UDF server will start running, allowing you to call the defined UDFs from RisingWave.

4. Declare your functions in RisingWave

In RisingWave, use the CREATE FUNCTION command to declare the functions you defined.

Here are the SQL statements for declaring the two UDFs defined in step 3.

CREATE FUNCTION gcd(int, int) RETURNS int
AS gcd
USING LINK 'http://localhost:8815';

CREATE FUNCTION series(int) RETURNS TABLE (x int)
AS series
USING LINK 'http://localhost:8815';

5. Use your functions in RisingWave

Once the UDFs are created in RisingWave, you can use them in SQL queries just like any built-in functions. For example:

SELECT gcd(25, 15);
SELECT * FROM series(10);

Data type mapping

The RisingWave Java UDF SDK supports the following data types:

SQL Type	Java Type	Notes
BOOLEAN	boolean, Boolean
SMALLINT	short, Short
INT	int, Integer
BIGINT	long, Long
REAL	float, Float
DOUBLE PRECISION	double, Double
DECIMAL	BigDecimal
DATE	java.time.LocalDate
TIME	java.time.LocalTime
TIMESTAMP	java.time.LocalDateTime
INTERVAL	com.risingwave.functions.PeriodDuration
VARCHAR	String
BYTEA	byte[]
JSONB	String	Use @DataTypeHint(“JSONB”) String as the type. See example.
T[]	T’[]	T can be any of the above SQL types. T’ should be the corresponding Java type.
STRUCT<>	user-defined class	Define a data class as the type. See example.

Example - JSONB

Define the function in Java

import com.google.gson.Gson;

// Returns the i-th element of a JSON array.
public class JsonbAccess implements ScalarFunction {
    static Gson gson = new Gson();

    public @DataTypeHint("JSONB") String eval(@DataTypeHint("JSONB") String json, int index) {
        if (json == null)
            return null;
        var array = gson.fromJson(json, Object[].class);
        if (index >= array.length || index < 0)
            return null;
        var obj = array[index];
        return gson.toJson(obj);
    }
}

Create the function in RisingWave

CREATE FUNCTION jsonb_access(jsonb, int) RETURNS jsonb
AS jsonb_access USING link 'http://localhost:8815';

Example - Struct type

Define the function in Java

// Split a socket address into host and port.
public static class IpPort implements ScalarFunction {
    public static class SocketAddr {
        public String host;
        public short port;
    }

    public SocketAddr eval(String addr) {
        var socketAddr = new SocketAddr();
        var parts = addr.split(":");
        socketAddr.host = parts[0];
        socketAddr.port = Short.parseShort(parts[1]);
        return socketAddr;
    }
}

Create the function in RisingWave

CREATE FUNCTION ip_port(varchar) RETURNS struct<host varchar, port smallint>
AS ip_port USING link 'http://localhost:8815';

Assistant

Responses are generated using AI and may contain mistakes.