使用日期列Java-Spark写入分区

SparkSession spark = SparkSession.builder()
        .appName("Java-Spark Write to Partition")
        .master("local")
        .getOrCreate();

Dataset<Row> data = spark.read()
        .format("csv")
        .option("header", "true")
        .load("path/to/input.csv");

import org.apache.spark.sql.functions;

Dataset<Row> partitionedData = data.withColumn("partition", functions.date_format(functions.col("dateColumn"), "yyyy-MM-dd"));

partitionedData.write()
        .format("parquet")
        .mode(SaveMode.Overwrite)
        .partitionBy("partition")
        .save("path/to/output");