hive高级数据类型

hive的高级数据类型主要包括：数组类型、map类型、结构体类型、集合类型，以下将分别详细介绍。

1）数组类型

array_type：array<data_type>

-- 建表语句

create table test.array_table(

name   string,

age    int,

addr   array<string>

)

row format delimited fields terminated by ','

collection items terminated by ':'

;

hive> desc test.array_table;

OK

name                    string                                     

age                     int                                        

addr                    array<string>

-- 测试数据

chavin,28,beijing:shanghai:shenyang

nope,28,beijing:shanghai:taiyuan

wzq,28,beijing:dalian:shenyang:fuxin

zhangshuai,28,beijing:shenyang:fuxin

-- 加载数据

hive> load data local inpath '/home/hadoop/array_table.txt' into table test.array_table;

Loading data to table test.array_table

OK

Time taken: 0.918 seconds

-- 查询

hive> select * from test.array_table;

OK

chavin    28    ["beijing","shanghai","shenyang"]

nope    28    ["beijing","shanghai","taiyuan"]

wzq    28    ["beijing","dalian","shenyang","fuxin"]

zhangshuai    28    ["beijing","shenyang","fuxin"]

Time taken: 0.199 seconds, Fetched: 4 row(s)

hive> select addr[1] from test.array_table;

OK

shanghai

shanghai

dalian

shenyang

Time taken: 0.186 seconds, Fetched: 4 row(s)

hive> select addr[0] from test.array_table;

OK

beijing

beijing

beijing

beijing

Time taken: 0.212 seconds, Fetched: 4 row(s)

2）map类型

map_type：map<primary_type,data_type>

-- 建表语句

create table test.map_table(

name      string,

detail    map<string,string>

)

row format delimited fields terminated by '\t'

collection items terminated by ','

map keys terminated by ':'

;

hive> desc test.map_table;

OK

name                    string                                     

detail                  map<string,string>                         

Time taken: 0.146 seconds, Fetched: 2 row(s)

-- 测试数据

chavin    age:28,addr:beijing,phone:15998888888,email:811651208@qq.com

nope    age:28,addr:beijing,phone:13899998888,email:nope@gmail.com

wzq    age:28,addr:shanghai,phone:13899998888,email:wzq@gmail.com

zhangshuai    age:28,addr:tianjing,phone:13899998888,email:zs@gmail.com

-- 加载数据

hive> load data local inpath '/home/hadoop/map.txt' into table test.map_table;

Loading data to table test.map_table

OK

Time taken: 0.641 seconds

-- 查询数据

hive> select * from test.map_table;

OK

chavin    {"age":"28","addr":"beijing","phone":"15998888888","email":"811651208@qq.com"}

nope    {"age":"28","addr":"beijing","phone":"13899998888","email":"nope@gmail.com"}

wzq    {"age":"28","addr":"shanghai","phone":"13899998888","email":"wzq@gmail.com"}

zhangshuai    {"age":"28","addr":"tianjing","phone":"13899998888","email":"zs@gmail.com"}

Time taken: 0.183 seconds, Fetched: 4 row(s)

hive> select detail from test.map_table;

OK

{"age":"28","addr":"beijing","phone":"15998888888","email":"811651208@qq.com"}

{"age":"28","addr":"beijing","phone":"13899998888","email":"nope@gmail.com"}

{"age":"28","addr":"shanghai","phone":"13899998888","email":"wzq@gmail.com"}

{"age":"28","addr":"tianjing","phone":"13899998888","email":"zs@gmail.com"}

Time taken: 0.177 seconds, Fetched: 4 row(s)

hive> select name,detail['email'] from test.map_table;

OK

chavin    811651208@qq.com

nope    nope@gmail.com

wzq    wzq@gmail.com

zhangshuai    zs@gmail.com

Time taken: 0.179 seconds, Fetched: 4 row(s)

3）结构体类型

struct_type：struct<col_name:data_type,... ...>

-- 建表语句

create table test.struct_table(

name      string,

detail    struct<age:int,phone:string,email:string>

)

row format delimited fields terminated by '\t'

collection items terminated by ','

;

hive> desc test.struct_table;

OK

name                    string                                     

detail                  struct<age:int,phone:string,email:string>                       

Time taken: 0.139 seconds, Fetched: 2 row(s)

-- 测试数据

chavin    28,15998888888,811651208@qq.com

nope    28,13899998888,nope@gmail.com

wzq    28,13899998888,wzq@gmail.com

zhangshuai    28,13899998888,zs@gmail.com

-- 加载数据

hive> load data local inpath '/home/hadoop/struct.txt' into table test.struct_table;

Loading data to table test.struct_table

OK

Time taken: 0.564 seconds

-- 查询数据

hive> select * from test.struct_table;

OK

chavin    {"age":28,"phone":"15998888888","email":"811651208@qq.com"}

nope    {"age":28,"phone":"13899998888","email":"nope@gmail.com"}

wzq    {"age":28,"phone":"13899998888","email":"wzq@gmail.com"}

zhangshuai    {"age":28,"phone":"13899998888","email":"zs@gmail.com"}

Time taken: 0.183 seconds, Fetched: 4 row(s)

hive> select name,detail.email from test.struct_table;

OK

chavin    811651208@qq.com

nope    nope@gmail.com

wzq    wzq@gmail.com

zhangshuai    zs@gmail.com

Time taken: 0.869 seconds, Fetched: 4 row(s)

4）集合类型

union_type：uniontype<data_type,data_type,... ...>

-- 建表语句

create table test.union_table(

detail    uniontype<int,string>

)

row format delimited fields terminated by '\t'

collection items terminated by ','

;

hive> desc test.union_table;

OK                                    

detail                  uniontype<int,string>                       

Time taken: 0.141 seconds, Fetched: 2 row(s)