设计一个基于多个带标签SparkSQL模板作为配置文件和多组参数的PySPARK代码程序，实现根据不同的输入参数自动批量地将数据导出为Parquet、CSV和Excel文件到S3上，标签和多个参数（以“_”分割）为组成导出数据文件名，文件已经存在则覆盖原始文件。
代码如下：

import json
from pyspark.sql import SparkSessiondef load_config(config_path):with open(config_path, 'r') as f:return json.load(f)def main(config_path, base_s3_path):# 初始化SparkSession，配置S3和Excel支持spark = SparkSession.builder \.appName("DataExportJob") \.config("spark.jars.packages", "com.crealytics:spark-excel_2.12:0.13.7,org.apache.hadoop:hadoop-aws:3.3.1") \.getOrCreate()# 配置S3访问（根据实际环境配置）spark.sparkContext._jsc.hadoopConfiguration().set("fs.s3a.access.key", "YOUR_ACCESS_KEY")spark.sparkContext._jsc.hadoopConfiguration().set("fs.s3a.secret.key", "YOUR_SECRET_KEY")spark.sparkContext._jsc.hadoopConfiguration().set("fs.s3a.endpoint", "s3.amazonaws.com")config = load_config(config_path)for template in config['templates']:label = template['label']sql_template = template['sql_template']parameters_list = template['parameters']for params in parameters_list:# 验证参数数量是否匹配placeholders = sql_template.count('{')if len(params) != placeholders:raise ValueError(f"参数数量不匹配，模板需要{placeholders}个参数，但当前参数为{len(params)}个")# 替换SQL中的占位符formatted_sql = sql_template.format(*params)df = spark.sql(formatted_sql)# 生成文件名参数部分param_str = "_".join(params)base_filename = f"{label}_{param_str}"# 定义输出路径output_paths = {'parquet': f"{base_s3_path}/parquet/{base_filename}",'csv': f"{base_s3_path}/csv/{base_filename}",'excel': f"{base_s3_path}/excel/{base_filename}.xlsx"}# 写入Parquetdf.write.mode('overwrite').parquet(output_paths['parquet'])# 写入CSV（自动生成header）df.write.mode('overwrite') \.option("header", "true") \.csv(output_paths['csv'])# 写入Excel（使用spark-excel包）df.write.format("com.crealytics.spark.excel") \.option("header", "true") \.option("inferSchema", "true") \.mode("overwrite") \.save(output_paths['excel'])spark.stop()if __name__ == "__main__":import argparseparser = argparse.ArgumentParser()parser.add_argument('--config', type=str, required=True, help='Path to config JSON file')parser.add_argument('--s3-path', type=str, required=True, help='Base S3 path (e.g., s3a://your-bucket/data)')args = parser.parse_args()main(args.config, args.s3_path)

配置文件示例（config.json）

{"templates": [{"label": "sales_report","sql_template": "SELECT * FROM sales WHERE date = '{0}' AND region = '{1}'","parameters": [["202301", "north"],["202302", "south"]]},{"label": "user_activity","sql_template": "SELECT user_id, COUNT(*) AS cnt FROM activity WHERE day = '{0}' GROUP BY user_id","parameters": [["2023-01-01"],["2023-01-02"]]}]
}

使用说明

1. 依赖管理：

   • 确保Spark集群已安装Hadoop AWS和Spark Excel依赖：

     spark-submit --packages com.crealytics:spark-excel_2.12:0.13.7,org.apache.hadoop:hadoop-aws:3.3.1 your_script.py
     

2. S3配置：

   • 替换代码中的YOUR_ACCESS_KEY和YOUR_SECRET_KEY为实际AWS凭证
   • 根据S3兼容存储调整endpoint（如使用MinIO需特殊配置）

3. 执行命令：

   spark-submit --packages com.crealytics:spark-excel_2.12:0.13.7,org.apache.hadoop:hadoop-aws:3.3.1 \
   data_export.py --config config.json --s3-path s3a://your-bucket/exports
   

输出结构

s3a://your-bucket/exports
├── parquet
│   ├── sales_report_202301_north
│   ├── sales_report_202302_south
│   └── user_activity_2023-01-01
├── csv
│   ├── sales_report_202301_north
│   ├── sales_report_202302_south
│   └── user_activity_2023-01-01
└── excel├── sales_report_202301_north.xlsx├── sales_report_202302_south.xlsx└── user_activity_2023-01-01.xlsx