在Google Dataflow上安装apt-get dependencies with Beam Java SDK

import org.apache.beam.sdk.extensions.gcp.util.gcsfs.GcsPath;
import org.apache.beam.sdk.options.PipelineOptionsFactory;

public class MyDataflowJob {
  public static void main(String[] args) {
    PipelineOptionsFactory.register(MyOptions.class);
    MyOptions options = PipelineOptionsFactory.fromArgs(args).withValidation().as(MyOptions.class);
    options.setFilesToStage(getDependencies());
    // ...
  }

  private static List<String> getDependencies() {
    List<String> dependencies = new ArrayList<>();
    dependencies.add("apt-get");
    // Add other dependencies here
    return dependencies;
  }
}

<dependencies>
  <dependency>
    <groupId>org.apache.beam</groupId>
    <artifactId>beam-sdks-java-core</artifactId>
    <version>2.33.0</version>
  </dependency>
  <!-- Add other dependencies here -->
</dependencies>

import java.io.IOException;

public class MyDataflowJob {
  public static void main(String[] args) {
    // ...
    try {
      Process process = Runtime.getRuntime().exec("apt-get install <dependency>");
      process.waitFor();
    } catch (IOException | InterruptedException e) {
      e.printStackTrace();
    }
    // ...
  }
}

$ mvn compile -e exec:java \
  -Dexec.mainClass=com.example.MyDataflowJob \
  -Dexec.args="--runner=DataflowRunner \
  --project=<your-project-id> \
  --stagingLocation=gs://<your-bucket>/staging \
  --tempLocation=gs://<your-bucket>/temp \
  --output=gs://<your-bucket>/output"