我试图将相关的子查询转换成pyspark代码，我得到以下错误。
具有相同名称的属性出现在操作中：
sql查询示例：

select(few columns)
from table0 spw1 
JOIN table1 spw2
       ON ( spw2.date_key=(SELECT MAX(date_key)
                            FROM table2 spw3
                            WHERE spw3.name =spw1.name
                            AND spw3.id = spw1.id
                            AND spw3.v_key=spw1.v_key
                            AND spw3.start_date < spw1.prev_start_date)