Import path in Java, Maven

Question

Following the tutorial about Kafka Streams located at: https://github.com/confluentinc/kafka-streams-examples/blob/4.0.0-post/src/main/java/io/confluent/examples/streams/WikipediaFeedAvroExample.java

There is a line:

import io.confluent.examples.streams.avro.WikiFeed

As I suppose it relates to this file: https://github.com/confluentinc/kafka-streams-examples/blob/4.0.0-post/src/main/resources/avro/io/confluent/examples/streams/wikifeed.avsc

• How does Maven knows it is in resource not java folder?
• Why io/confluent/examples/streams/avro/wikifeed.avsc instead of avro/io/confluent/examples/streams/wikifeed.avsc?

The other import is even more fantastic:

import io.confluent.kafka.serializers.AbstractKafkaAvroSerDeConfig;

• There is no kafka folder in the java/io/confluent folder.

https://github.com/confluentinc/kafka-streams-examples/tree/4.0.0-post/src/main/resources/avro/io/confluent.

How does all this magic suppose to work?

Answer 1

The magic is made by avro-maven-plugin which you can find in the pom.xml:

<plugin>
    <groupId>org.apache.avro</groupId>
    <artifactId>avro-maven-plugin</artifactId>
    <version>${avro.version}</version>
    <executions>
        <execution>
            <phase>generate-sources</phase>
            <goals>
                <goal>schema</goal>
            </goals>
            <configuration>
                <sourceDirectory>src/main/resources/avro/io/confluent/examples/streams</sourceDirectory>
                <outputDirectory>${project.build.directory}/generated-sources</outputDirectory>
                <stringType>String</stringType>
            </configuration>
        </execution>
    </executions>
</plugin>

Quoting from the documentation of the plugin:

Simple integration with dynamic languages. Code generation is not required to read or write data files nor to use or implement RPC protocols. Code generation as an optional optimization, only worth implementing for statically typed languages.

This is, at pre compile time, the plugin reads the content of avsc files and generate binary sources (for this case, Java classes) that then can be used in the code.

You can see the code generated by the plugin in target/generated-sources. There will be a folder structure and proper java (not class) files there.

Answer 2

The WikiFeed class is created dynamically at build time using the avro-maven-plugin from the .avsc file you linked to. You can check how it's configured in the <plugins> section of pom.xml.

The AbstractKafkaAvroSerDeConfig class comes from the kafka-avro-serializer dependency. Eclipse has a nice way of navigating from the individual class in the Editor view back to the Package Explorer which includes the Maven dependencies, like this: