hive行转多列LATERAL VIEW explode

该文参考了:http://blog.sina.com.cn/s/blog_7e04e0d00101csic.html

https://cwiki.apache.org/confluence/display/Hive/LanguageManual+LateralView

lateral view用于和split、explode等UDTF一起使用的,能将一行数据拆分成多行数据,在此基础上可以对拆分的数据进行聚合,lateral view首先为原始表的每行调用UDTF,UDTF会把一行拆分成一行或者多行,lateral view在把结果组合,产生一个支持别名表的虚拟表。

单个LATERAL VIEW:

源表(table1)数据{A:string B:array C:string}

A                         B                                C

190     [1030,1031,1032,1033,1190]      select id
191     [1030,1031,1032,1033,1190]      select id

希望的结果是:

190    1030  select id

190    1031  select id

190    1032  select id

190    1033  select id

190    1190  select id

191    1030  select id

191    1031  select id

191    1032  select id

191    1033  select id

191    1190  select id

故使用select A,B,C from table_1 LATERAL VIEW explode(B) table1 as B得到上述结果

多个LATERAL VIEW的介绍:

LATERAL VIEW clauses are applied in the order that they appear. For example with the following base table:

Array col1

Array col2

[1, 2]

[a", "b", "c"]

[3, 4]

[d", "e", "f"]

SELECT  myCol1, myCol2  FROM  baseTable
LATERAL  VIEW  explode(col1) myTable1  AS  myCol1
LATERAL  VIEW  explode(col2) myTable2  AS  myCol2;
结果如下:

int myCol1

string myCol2

1

"a"

1

"b"

1

"c"

2

"a"

2

"b"

2

"c"

3

"d"

3

"e"

3

"f"

4

"d"

4

"e"

4

"f"

 

 

 复杂方式:

select * from tb_split;

20141018  aa|bb 7|9|0|3

20141019  cc|dd  6|1|8|5

使用方式:select datenu,des,type from tb_split 

lateral view explode(split(des,"//|")) tb1 as des

lateral view explode(split(type,"//|")) tb2 as type

执行过程是先执行from到 as cloumn的列过程,在执行select 和where后边的语句;

转载于:https://www.cnblogs.com/judylucky/p/3713774.html

你可能感兴趣的:(大数据)