pyspark在spark sql中函数之间的使用范围

Question

当我跑步时

spark.sql(''' 
        select client,avg(amount) over (partition by client
         order by my_timestamp
         range between interval 30 days preceding and current row) as moving_avg_amount
         from my_table''')

它有效。

但如果我想排除最后 X 天 - 它会失败。

... range between interval 30 days preceding and 12 days preceding ..
   or :
 ... range between interval 30 days preceding and interval 12 days preceding .. 

正确的语法是什么？

Answer 1

在Spark SQL Window over interval of between two specified time boundaries - between 3 hours and 2 hours ago 中建议范围间隔在 SparkSQL 中不能正常工作。