将数据从一个表复制到另一分区表 [英] copy data from one table to another partitioning table

问题描述

%hive
INSERT INTO NEWPARTITIONING partition(year(L_SHIPDATE)) select * from LINEITEM;

我想将订单项中的数据复制到分区表 NEWPARTITIONING 中，但是出现以下错误:

I want to copy the data from line item to the partitioning table NEWPARTITIONING but I got the following error:

1:54行无法识别语句中')''select''*'附近的输入.

line 1:54 cannot recognize input near ')' 'select' '*' in statement.

不明白为什么会发生此错误.谁能给我一些想法

Don't understand why this error occurs. Can anyone give me some ideas

推荐答案

Hive支持 DYNAMIC 或 STATIC 分区加载.

Hive supports DYNAMIC or STATIC partition loading.

分区规范仅允许使用列名或列列表(用于动态分区加载)，如果需要功能，请在选择中进行计算，请参见以下示例:

Partition specification allows only column name or column list (for dynamic partition load), if you need function, then calculate it in the select, see example below:

set hive.exec.dynamic.partition=true;
set hive.exec.dynamic.partition.mode=nonstrict;

insert into table NEWPARTITIONING partition (partition_column)
select i.col1,
       ...
       i.colN,
       year(L_SHIPDATE) as partition_column --Partition should be the last in column list
  from LINEITEM i 

或者您可以以 partition(partition_column ='value')的形式指定静态分区，在这种情况下，您无需选择分区表达式:

Or you can specify static partition in the form partition(partition_column='value'), in this case you do not need to select partition expression:

insert into table NEWPARTITIONING partition (partition_column='2020-01-01')
select i.col1,
       ...
       i.colN
  from LINEITEM i 
 where year(L_SHIPDATE)  = '2020-01-01' 

在STATIC和DYNAMIC两种情况下，Hive不支持分区规范中的功能.函数可以在查询中计算(动态负载)，也可以在包装器外壳中计算，然后作为参数传递给脚本(用于静态分区).

In both cases - STATIC and DYNAMIC, Hive does not support functions in partition specification. Functions can be calculated in the query (dynamic load) or calculated in a wrapper shell and passed as a parameter to the script (for static partition).

这篇关于将数据从一个表复制到另一分区表的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！