HBase基础之Hbase shell常用操作

一般操作

查看服务器状态

status

查看hbase版本

version

DDL操作

创建表

create 'member','member_id','address','info'

创建了3个列族，分别是member_id, address, info

知识点回顾：cf是schema的一部分，而column不是。

查看表信息

describe 'member'


DESCRIPTION                                                              ENABLED                               
 'member', {NAME => 'address', DATA_BLOCK_ENCODING => 'NONE', BLOOMFILTE true                                  
 R => 'ROW', REPLICATION_SCOPE => '0', VERSIONS => '1', COMPRESSION => '                                       
 NONE', MIN_VERSIONS => '0', TTL => '2147483647', KEEP_DELETED_CELLS =>                                        
 'false', BLOCKSIZE => '65536', IN_MEMORY => 'false', BLOCKCACHE => 'tru                                       
 e'}, {NAME => 'info', DATA_BLOCK_ENCODING => 'NONE', BLOOMFILTER => 'RO                                       
 W', REPLICATION_SCOPE => '0', VERSIONS => '1', COMPRESSION => 'NONE', M                                       
 IN_VERSIONS => '0', TTL => '2147483647', KEEP_DELETED_CELLS => 'false',                                       
  BLOCKSIZE => '65536', IN_MEMORY => 'false', BLOCKCACHE => 'true'}, {NA                                       
 ME => 'member_id', DATA_BLOCK_ENCODING => 'NONE', BLOOMFILTER => 'ROW',                                       
  REPLICATION_SCOPE => '0', VERSIONS => '1', COMPRESSION => 'NONE', MIN_                                       
 VERSIONS => '0', TTL => '2147483647', KEEP_DELETED_CELLS => 'false', BL                                       
 OCKSIZE => '65536', IN_MEMORY => 'false', BLOCKCACHE => 'true'}                                               
1 row(s) in 0.1800 seconds

查询所有的表

list

删除一个列族

member表建了3个列族，但是发现member_id这个列族是多余的，因为他就是主键，所以我们要将其删除。

alter 'member','delete'=>'member_id' 

drop表

为了测试drop，先创建一个表

create 'tmp_table','info'

删除表之前需要先将表disable再drop(新版本不用disable)

disable 'tmp_table'
drop 'tmp_table'

判断表是否enable

is_enabled 'member'

判断表是否disabled

is_disabled 'member'

DML操作

插入记录

格式：put 表名 row_key cf:column value

put 'member','luogankun','info:age','27'
put 'member','luogankun','info:birthday','1986-09-05'
put 'member','luogankun','info:company','asinainfo-linkage'
put 'member','luogankun','address:country','china'
put 'member','luogankun','address:province','beijing'
put 'member','luogankun','address:city','beijing'

put 'member','spring','info:age','27'
put 'member','spring','info:birthday','1986-05-14'
put 'member','spring','info:company','asinainfo-linkage'
put 'member','spring','address:country','china'
put 'member','spring','address:province','hubei'
put 'member','spring','address:city','wuhan'
put 'member','spring','info:favorite','shopping'

知识点回顾： column完全动态扩展，每行可以有不同的columns。

获取一个rowkey的所有数据

格式：get 表名 row_key

get 'member','luogankun'

COLUMN                       CELL                                                                              
 address:city                timestamp=1409122962541, value=beijing                                            
 address:country             timestamp=1409122962468, value=china                                              
 address:province            timestamp=1409122962510, value=beijing                                            
 info:age                    timestamp=1409122962328, value=27                                                 
 info:birthday               timestamp=1409122962399, value=1986-09-05                                         
 info:company                timestamp=1409122962434, value=asinainfo-linkage 

知识点回顾：htable按rowkey字典序(1,10,100,2)自动排序，每行包含任意数量的columns，columns按照columnkey(address:city,address:country,address:province,info:age,info:birthday,info:company)自动排序。

获取一个id，一个列族的所有数据
格式： get 表名 row_key column

get 'member','luogankun','info'

COLUMN                       CELL                                                                              
 info:age                    timestamp=1409122962328, value=27                                                 
 info:birthday               timestamp=1409122962399, value=1986-09-05                                         
 info:company                timestamp=1409122962434, value=asinainfo-linkage 

获取一个id，一个列族中一个列的所有数据
格式：get 表名 row_key cf:column

get 'member','luogankun','info:age'

COLUMN                       CELL                                                                              
 info:age                    timestamp=1409122962328, value=27

更新一条记录
格式： put 表名 row_key cf:column value
将luogankun的年龄改成18

put 'member','luogankun','info:age','18'

get 'member','luogankun','info:age'

COLUMN                       CELL                                                                              
 info:age                    timestamp=1409123175384, value=18 

知识点回顾：查询默认返回最新的值。

通过timestamp来获取指定版本的数据
格式： get 表名 row_key {COLUMN=>'cf:column',TIMESTAMP=>xxxxxx}

get 'member','luogankun',{COLUMN=>'info:age',TIMESTAMP=>1409122962328}

COLUMN                       CELL                                                                              
 info:age                    timestamp=1409122962328, value=27                                                 


get 'member','luogankun',{COLUMN=>'info:age',TIMESTAMP=>1409123175384}

COLUMN                       CELL                                                                              
 info:age                    timestamp=1409123175384, value=18  

知识点回顾：每个column可以有任意数量的values，按timestamp倒序自动排序；tableName+rowkey+column+timestamp==>value

全表扫描
格式：scan 表名

scan 'member'

ROW                          COLUMN+CELL                                                                       
 luogankun                   column=address:city, timestamp=1409122962541, value=beijing                       
 luogankun                   column=address:country, timestamp=1409122962468, value=china                      
 luogankun                   column=address:province, timestamp=1409122962510, value=beijing                   
 luogankun                   column=info:age, timestamp=1409123175384, value=18                                
 luogankun                   column=info:birthday, timestamp=1409122962399, value=1986-09-05                   
 luogankun                   column=info:company, timestamp=1409122962434, value=asinainfo-linkage             
 spring                      column=address:city, timestamp=1409122962828, value=wuhan                         
 spring                      column=address:country, timestamp=1409122962754, value=china                      
 spring                      column=address:province, timestamp=1409122962787, value=hubei                     
 spring                      column=info:age, timestamp=1409122962592, value=27                                
 spring                      column=info:birthday, timestamp=1409122962623, value=1986-05-14                   
 spring                      column=info:company, timestamp=1409122962670, value=asinainfo-linkage             
 spring                      column=info:favorite, timestamp=1409122963494, value=shopping      

删除id为spring的值的'info:age'字段
格式：delete 表名 row_key cf:column
先查看

get 'member','spring','info:age'

COLUMN                       CELL                                                                              
 info:age                    timestamp=1409122962592, value=27

再删除

delete 'member','spring','info:age'

再查看对比是否已经删除

get 'member','spring','info:age'

COLUMN                       CELL                                                                              
0 row(s) 

查询表中有多少行
格式：count 表名

count 'member'

删除整行
格式: deleteall 表名 row_key

deleteall 'member','spring'

将整张表清空
格式： truncate 表名

truncate 'member'

Truncating 'member' table (it may take a while):
 - Disabling table...
 - Dropping table...
 - Creating table...

可以看出，hbase是先将掉disable掉，然后drop掉后重建表来实现truncate的功能的。