无法将变量传递给pyspark中的Spark SQL查询

Question

我有一个日期数据类型的python变量（我正在使用pyspark）：变量值为2016-10-31

print type(load_dt)

 >> <type 'datetime.date'>

我很难将其传递给sparksql查询：

    hive_context.sql("select * from  tbl t1 where cast (substring(t1.dt,1,10) as date) ={0}".format(load_dt));

    Error:

    u"cannot resolve '(cast(substring(dt,1,10) as date) = ((2016 - 10) - 31))' due to data type mismatch: differing types in '(cast(substring(period_dt,1,10) as date) = ((2016 - 10) - 31))'
 (date and int)

Answer 1

添加引号：

"select * from  tbl t1 where cast (substring(t1.dt,1,10) as date) = '{0}'"

否则，日期将转换为2016-10-31字符串并解释为算术表达式：

2016 - 10 - 31