amazon athena 创建带分区的请求

Posted 2023-03-31

【中文标题】amazon athena 创建带分区的请求

【英文标题】：amazon athena create request with partitions

【发布时间】：2018-08-06 19:00:00

【问题描述】：

我创建了一个表，其分区如下：首先按年、月和日。

问题：我希望得到 12/2017 和 03/2018 的数据，我该怎么做？ 我的想法：

where (year='2017' and month='12') and ( year ='2018' and month='03')

正确吗？我不会感到困惑，因此 Amazon Athena 获取以下数据：

12/2017 and 03/2018 and 03/2017 and 12/2018 

因为 and 运算符？

PS：我无法测试，我只有免费帐户。 谢谢。

【参考方案1】：

无论如何，我尝试了一组迷你数据，发现 Amazon Athena 考虑了括号。

我的测试如下： 生成的表的 DDl：

CREATE EXTERNAL TABLE `manyands`(
  `years` int COMMENT 'from deserializer', 
  `months` int COMMENT 'from deserializer', 
  `days` int COMMENT 'from deserializer')
PARTITIONED BY ( 
  `year` string, 
  `month` string)
ROW FORMAT SERDE 
  'org.openx.data.jsonserde.JsonSerDe' 
STORED AS INPUTFORMAT 
  'org.apache.hadoop.mapred.TextInputFormat' 
OUTPUTFORMAT 
  'org.apache.hadoop.hive.ql.io.IgnoreKeyTextOutputFormat'
LOCATION
  's3://mybucket/'

我的一组数据测试：

我的测试：

1-SELECT * FROM "atlasdatabase"."manyands" where month='1'; 我得到了 CSV 格式：

"years","months","days","year","month"
"2017","1","21","2017","1"
"2018","1","81","2018","1"

2-SELECT * FROM "atlasdatabase"."manyands" where month='1' and year='2017';

"years","months","days","year","month"
"2017","1","21","2017","1"

3-SELECT * FROM "atlasdatabase"."manyands" where (month='1' and year='2018') and (month='3' and year='2017') ;

empty (Zéro enregistrements renvoyés)

4-SELECT * FROM "atlasdatabase"."manyands" where (month='1' and year='2018') or (month='3' ) ;

"years","months","days","year","month"
"2018","1","81","2018","1"
"2017","3","73","2017","3"
"2018","3","73","2018","3"

结论：在多个分区实例之间添加OR运算符。