试图将表中的值更新到另一个表中，这两个表都有相同的字段名，但值不同，查询必须在任何普通数据库上正常工作，但在这里它返回
编译语句时出错：失败：parseexception行1:0无法识别“merge”附近的输入到“final”
MERGE
INTO FINAL
USING FIRST_STAGE
ON IMSI = FIRST_STAGE.IMSI and Site = FIRST_STAGE.Site
WHEN MATCHED THEN UPDATE SET
Min_Date = least(FIRST_STAGE.Min_Date, Min_Date),
Max_Date = greatest(FIRST_STAGE.Max_Date, Max_Date),
NoofDays = FIRST_STAGE.NoofDays + NoofDays,
Down_Link = FIRST_STAGE.Down_Link + Down_Link,
up_Link = FIRST_STAGE.up_Link + up_Link,
connection = FIRST_STAGE.connection + connection
WHEN NOT MATCHED THEN INSERT ( Min_Date,
Max_Date,
NoofDays,
IMSI,
Site,
Down_Link,
Up_Link,
Connection )
VALUES ( FIRST_STAGE.Min_Date,
FIRST_STAGE.Max_Date,
FIRST_STAGE.NoofDays,
FIRST_STAGE.IMSI,
FIRST_STAGE.Site,
FIRST_STAGE.Down_Link,
FIRST_STAGE.Up_Link,
FIRST_STAGE.Connection )
1条答案
按热度按时间0dxa2lsx1#
中引入了配置单元合并语句
Hortonworks distribution.这些的先决条件
merge statement要运行的是：最终表格需要用
transactional enabled ,ORC format ,and bucketed.afaik以防
Cloudera distribution我们需要使用Kudu从cloudera-5.10+.注：
Upsert statement only works for Impala tables that use the Kudu storage engine.我不认为我们可以运行中的帖子中提到的合并语句CDH到现在为止。