show tblproperties table_name

show tables

describe extended mydb.employees

–where查询自居没有加分区过滤的话就会禁止提交这个任务

set hive.mapred.mode=strict;

–查看表分区

show partitions Tablename;

show partitions table_name partition(country=’US’)

load data local inpath ‘${env:Home}/california-employees’

into Table employees

partition(country=’US’,sdate=’CA’);

–增加一个2012年1月2日的分区

alter table log_message

ADD PARTITION (year=2012,month=1,day=2)

location ‘hdfs://master_server/data/log_message/2012/01/02’

–删除表

drop table if exists employees;

–表重名

alter table log_message rename to logmsgs;

–增加、修改和删除表分区

alteer table log_message add if not exists

partion (year =2011,month=1,day=1) Location ‘/logs/2011/01/01’

–修改列信息

alter table log_message

change cloumn hms hours_minutes_seconds INT

cloumn ‘The hours ,minutes,and seconds part of the timestamp’

alter severity;

–增加列

alter table log_message add columns(

app_name string comment’Application name’,

session_id long comment ‘The current session_id’

);

–存储属性

alter table stocks

clustered by (exchange ,symbol)

stored by (symbol)

into 48 buckets；

–向管理表中转载数据

load data local inpath ‘${env:HOME}/california-employees’

overwrite into table employees

partion (country=’US’,state =’CA’);

–动态分区插入

insert overwrite table employees

partion(country,state)

select se.st,se.cnty

from staged_employees se;

–动态分区属性

set hive.exec.dynamic.partion=true;

set hive.exec.dynamic.partion.mode=nonstrict;

set hive.exec.max.dynamic.partions.pernode=1000;

–从宽表中拿数据，单个查询语句中

create table ca_employees

as select name ,salary,address

from employees

where se.state =’CA’

–导出数据

hadoop fs -cp source_path target_path

==========

insert overwrite local directory ‘/tmp/ca_employees’

select name, salary,address

from employees

where se.state=’CA’

–指定多个输出文件夹目录，向表里插入数据

from staged_employees se

insert overwrite directory’/tmp/or_employees’

select *

where se.cty=’US’ and se.st=’OR’

insert overwrite directory ‘/tmp/ca_employees’

select *

where se.cty=’US’ and se.st=’CA’

–建表

create table employees(

name string,

salary float,

subordinates array<string>,

deductions MAP<string ,float>,

address  struct<street:string ,city:string,state:string,zip:int>

)

partitioned by (country string,state string)

–查询

select name, subordinates from employees;

–查询 array 格式的数据

select name ,subordinates[0] from employees

–查询Map格式的数据

select name ,deductions[“State Taxes”] from employees

–查询Struct格式的数据

select name ,address.city from employees

–使用正则表达式来指定列,查询symbol和所有以price作为前缀的列

select symbol,’price.*’

from stocks

–列值计算,round()函数会返回一个Double 类型的最近整数

–floor,round,ceil（“向上取整”）输入的是Double类型的值，返回的是bigint

select upper(name),salary,deductions[“Federal Taxes”],

round(salary *(1-deductions[“Federal Taxes “]))

from employees;

–聚合函数，count，avg

select count(*),avg(salary)

from employees;

–count(*),count(distinct expr[,expr_.]),sum(col),sum(distinct col)

–avg(),min(col),max(col)

–设置属性 hive.map.aggr值为true来提高聚合性能

select hive.map.aggr=true;

select count(*),avg(salary)

from employees;

–字段为空的话将不会产生新的记录；不为空的话，每个数组的灭个元素将会产生新的纪录

select explode(subordinates) as sub

from employees

–limit 语句会限制返回的行数

select upper(name0,salary,deductions[“Federal Taxes”],

round(salary*(1-deductions[“Federal Taxes”]))

from employees

limit 2;

–case … when … then

select name , salary,

case

when salary<5000 then ‘low’

when salary>=5000 and salary<7000 then ‘middle’

when salary>=7000 and salary<10000 then ‘high’

else ‘very high’

end as bracket from employees

–hive中某些情况的查询可以不需要用到MapReduce就是所谓的本地模式

–属性设置为true的话，hive就会尝试本地模式

set  hive.exec.mode.local.auto =true;

–like 模糊匹配，RLike是hive中的扩展

–inner join的时候，表的大小从左到右依次增加

select s.ymd,s.symbol,s.price_close,d.divided

from stocks s join divided d

on s.ymd=d.ymd and s.symbol= d.symbol

where s|.symbol =’AAPL’;

–right outer join 返回右表中所有符合where语句的记录

–所有的表中只有一张小表，可以在最大的表通过mapper的时候将小表完全放内存中

set hive.auto.convert.join =true;

–order by 会对查询结果执行一个全局的扫描

–hive里添加了一个sort by，只会在每个reducer中对数据进行排序

–含有sort by 的distribute by

MapReduce job中传输的所有数据按照健-值对的方式进行组织的，默认情况下，

MapReduce计算框架会依据map输入的健计算相应的哈希值，

然后按照得到的哈希值将键-值对均匀发布到多个reducer中去

–distribute by 控制map的输出reducer中是如何划分的

–Hive要求Distribute by 写在Sort by 语句之前

select s.ymd,s.symbol,s.price_close

from stocks s

distribute by s.symbol

sort by s.symbol ASC,s.ymd ASC

–cluster by 等价于distribute by 和sort by 的综合使用

select s.ymd,s.symbol,s.price_close

from stocks s

cluster by s.symbol

–类型转换

select name,salary

from employees

where cast(salary as Float)<100000.0

select (2.0*cast(cast(b as string ) as double ))

from src;

–数据快抽样，hive提供了一种按照抽样百分比进行抽样的方式，这是基于行数的，

这样抽样的最小单元是HDFS数据块，如果小于128MB的话，会返回所有行

–hive目前暂不支持物化视图，嵌套子查询的查询

from（

select *

from people join cart

on (cart.people=people.id)

where firstname =’john’

）a  select a.lastname where a.id=3;

–创建索引

create table employees(

name string ,

salary float,

subordinates array<string>,

deductions map<string ,float>,

address  struct<street:string ,city:string,state:string,zip:int>

)

create index employees_index

on table employees(country)

as ‘org.apache.hadoop.hive.ql.index.compact.compactindexhandler’

with deferred rebuild

idxproperties (‘creator=’me’,’created_at’= ‘some_time’)

in table employees_index_table

partion by (country,name)

comment’Employees indexed by country and name.’

partioned by (country string, state string )

–修改索引

alter index employees_index

on table employees

partition (country=’US’)

rebuild;

–显示索引

show formatted index on employees;

–删除索引

drop index if exists employees_index on table employees;

–Bitmap 索引应用于排重后值比较少的列