Impala的简单使用

• 创建表

create table if not exists touch.table_test(
  name string comment '姓名',
  age int not null comment '年龄',
  sex tinyint not null comment '性别',
  create_time timestamp not null comment '创建时间',
  primary key(name) 
  ) partition  by hash(name) partitions  10 
stored as kudu

• 查看表定义

describe table_test

• 修改表的指定列

ALTER TABLE table_test CHANGE name user_name String

• 表的添加某些列

ALTER TABLE table_test ADD COLUMNS( update_time timestamp)
ALTER TABLE table_test ADD COLUMNS( email string,phone string)

• 表的删除指定列

ALTER TABLE table_test drop sex

• 表的插入数据

insert into table_test values('ggr',23,now(),now(),'[email protected]','17895642365');

• 表的更新数据

insert overwrite table_test values('ggr',24,now(),now(),'[email protected]','17895642365');
//2.8版本之后
UPSERT INTO kudu_table (pk, c1, c2, c3) VALUES (0, 'hello', 50, true), (1, 'world', -1, false);

UPSERT INTO production_table SELECT * FROM staging_table;
 | Impala SQL Language Reference | 418
UPSERT INTO production_table SELECT * FROM staging_table WHERE c1 IS NOT
 NULL AND c2 > 0;

• 表的查询数据

SELECT * FROM table_test;
SELECT user_name,age FROM table_test;   

• 视图

Create View IF NOT EXISTS view_name as Select statement
DROP VIEW database_name.view_name;
ALTER VIEW database_name.view_name为Select语句

• Order by 排序

select * from table_name ORDER BY col_name [ASC|DESC] [NULLS FIRST|NULLS LAST]

• Group by分组

select data from table_name Group BY col_name;

• Having 子句
  Impala中的Having子句允许您指定过滤哪些组结果显示在最终结果中的条件。
  一般来说，Having子句与group by子句一起使用; 它将条件放置在由GROUP BY子句创建的组上。

select * from table_name ORDER BY col_name [ASC|DESC] [NULLS FIRST|NULLS LAST]

• DISTINCT运算符(去重)

select distinct columns… from table_name;

• Union 使用

select * from customers order by id limit 3
 union select * from employee order by id limit 3;

• 偏移量（offset） & 限制（limit）
  OFFSET 需要使用order by 条件语句配合。

select * from table_test order by create_time limit 2 offset 1