HIVE 第八章 UDF

Posted 2021-04-29 麦田里的守望者

公告：更新任在持续，更多精彩内容敬请关注！

当前浏览器不支持播放音乐或语音，请在微信或其他浏览器中播放

Hive UDF

 FUNCTION

dual表的概念来自oracle，数据库建立时即与数据字典一起初始化，该表只有一个varchar2类型名为dummy的字段，表数据只有一行“X”，用来查询一些系统信息，如select sysdate from dual; select user from dual;select seq.nextval from dual等。为了能在hive中测试一些时间、数学、聚合函数，可以仿照oracle创建dual表。

[root@hive ~]# echo X > dual.txt

hive>create dual (dummy String);

hive>load data local inpath '/root/dual.txt' overwrite into table dual;

Hive内部自定义函数UDF

 

 等值比较: =，> ,<,>=,<=,<>,!=,is [not] null 

hive> select 1 from dual where 1=1;
1

模糊搜索 like

hive> select 1  from dual where 'football' like '%foot%';

OK

1

hive> select 1  from dual where 'football' like '%foot____';

OK

1

注意：符”_”表示任意单个字符，而字符”%”表示任意数量的字符。

正则搜索RLIKE 或者 REGEXP

hive> select 1 from dual where 'football' rlike '^f.*l$';

OK

1

hive> select 1 from dual where 'footbar' regexp '^f.*r$';

OK

1

数学运算符 +， -，*，/ ,% 

hive> select 10/3 from dual;

OK

3.3333333333333335

位操作 &，|，~，^

hive> select 1|2 from dual;

OK

3

Time taken: 0.086 seconds, Fetched: 1 row(s)

hive> select 1&2 from dual;

OK

0

Time taken: 0.1 seconds, Fetched: 1 row(s)

hive> select 1^2 from dual;

OK

3

Time taken: 0.081 seconds, Fetched: 1 row(s)

hive> select ~1 from dual;

OK

-2

布尔表达式 and ，or， not

hive> select 1 from dual where 1 =2 or 3=3;

OK

1

hive> select 1 from dual where 1 =2 and 3=3;

OK

hive> select 1 from dual where not 3=3;

OK

hive> select 1 from dual where not 3=2;

OK

1

复合类型构建操作 

① map类型

hive>  create table udftest as select map('100','tom','200','mary') as t from dual;

hive> desc udftest;

OK

t                       map<string,string>  

hive>  select t from udftest;

OK

{"100":"tom","200":"mary"}

hive>  select t[100] from udftest;

OK

tom

hive>  select t[200] from udftest;

OK

mary

② struct 类型

hive> drop table udftest;

hive>create table udftest as select struct('tom','mary','tim') as t from dual;

hive> desc  udftest;

OK

t    struct<col1:string,col2:string,col3:string>                         

hive> select  t.col1,t.col2 ,t.col3 from  udftest;

OK

tom     mary    tim

hive> select t from udftest;

OK

{"col1":"tom","col2":"mary","col3":"tim"}

③ array 类型

hive> drop table udftest;

hive> create table udftest as select array('jiangzz','wangwu','lisi') as t from dual;

hive> describe udftest;

OK

t      array<string>                    

hive> select t[0],t[1],t[2] from udftest;

OK

jiangzz wangwu  lisi

 数值计算

round 

hive> select round(3.1415926) from dual;

OK

3.0

hive> select round(3.1415926,3) from dual;

OK

3.142

floor、ceil|ceiling

hive> select floor(3.1415926) from dual;

OK

3

hive> select ceil(3.1415926) from dual;

OK

4

hive> select ceiling(3.1415926) from dual;

OK

4

rand 

hive> select rand() from dual;

OK

0.8883078127190064

hive> select rand() from dual;

OK

0.7074664833607184

hive> select rand(100) from dual;

OK

0.7220096548596434

pow|power、sqrt

hive> select pow(10,2) from dual;

OK

100.0

hive> select power(10,2) from dual;

OK

100.0

hive> select sqrt(3) from dual;

OK

1.7320508075688772

bin

hive> select bin(2) from dual;

OK

10

Time taken: 0.116 seconds, Fetched: 1 row(s)

hive> select bin(8) from dual;

OK

1000

hex、unhex

hive> select hex(15) from dual;

OK

F

hive> select hex(16) from dual;

OK

10

hive> select hex('jiangzz') from dual;

OK

6A69616E677A7A

hive> select unhex('6A69616E677A7A') from dual;

OK

jiangzz

conv(BIGINT num, int from_base, int to_base)

hive> select conv(10,10,8) from dual;

OK

12

hive> select conv(10,2,10) from dual;

OK

2

abs 

hive> select abs(-10) from dual;

OK

10

pmod

hive> select pmod(10,3) from dual;

OK

1

Time taken: 0.073 seconds, Fetched: 1 row(s)

hive> select pmod(10.5,3) from dual;

OK

1.5

Time taken: 0.064 seconds, Fetched: 1 row(s)

本期内容暂时更新到此，更多精彩内容请持续关注jiangzz_wy公众账号，如有任何疑问也可以添加小编jiangzz_wx微信。

添加关注