314.6. DataFrame ジョブ

spark:dataframe?dataFrame=#testDataFrame&dataFrameCallback=#transformation

 Where `dataFrame` option refers to the name of an DataFrame instance
(`instances of org.apache.spark.sql.Dataset and org.apache.spark.sql.Row`) from a Camel registry,
while `dataFrameCallback` refers to the implementation
of `org.apache.camel.component.spark.DataFrameCallback` interface (also
from a registry). DataFrame callback provides a single method used to
apply incoming messages against the given DataFrame. Results of callback
computations are saved as a body to an exchange.

public interface DataFrameCallback<T> {
    T onDataFrame(Dataset<Row> dataFrame, Object... payloads);
}

String model = "Micra";
long linesCount = producerTemplate.requestBody("spark:dataFrame?dataFrame=#cars&dataFrameCallback=#findCarWithModel", model, long.class);

@Bean
RddCallback<Long> findCarWithModel() {
    return new DataFrameCallback<Long>() {
        @Override
        public Long onDataFrame(Dataset<Row> dataFrame, Object... payloads) {
            String model = (String) payloads[0];
            return dataFrame.where(dataFrame.col("model").eqNullSafe(model)).count();
        }
    };
}

@Bean
Dataset<Row> cars(HiveContext hiveContext) {
    Dataset<Row> jsonCars = hiveContext.read().json("/var/data/cars.json");
    jsonCars.registerTempTable("cars");
    return jsonCars;
}