【4】HIVE中map，array和structs使用

1：怎样导入文本文件（格式是怎样的？），2：怎样查询数据，已经能否在join中使用？在子查询中使用？等等

知道怎么在hive中导入数组不？
例如：我想把 数组[1,2,3] 和 数组 ["a","b","c"]
导入到table1中
create table table2 ( a array<int> , b array<string>);

那么 我如何 导入呢？使得
select * from table1;
j结果为：
[1,2,3] ["a","b","c"]

同样 在 hive 中 对于 map
怎样 查询呢？
例如 
create table table2 ( a MAP<STRING,ARRAY<STRING>>);
select * from table2 结果为：
{"d01":["d011","d012"],"d02":["d021","d022"]}
{"d01":["d011","d012"],"d02":null}
{"d01":[null,"d012"],"d02":["d021","d022"]}
那么 我想获得 key 为 d01的value值 
该怎么操作呢

关于数组的操作说明：
drop table table2;

create table table2 (a array<string>, b array<string>)
ROW FORMAT DELIMITED
FIELDS TERMINATED BY '\t'
COLLECTION ITEMS TERMINATED BY ',';

load data local inpath "../hive/examples/files/arraytest.txt"  overwrite into table table2;

arraytest.txt中的数据形式为：（不同数组间用\t分割，同一数组内不同元素用逗号分割）
b00,b01        b00,b01
b00,b01        b00,b01
b00,b01        b00,b01
b00,b01        b00,b01


hive> select * from table2;

OK
["b00","b01"]   ["b00","b01"]
["b00","b01"]   ["b00","b01"]
["b00","b01"]   ["b00","b01"]
["b00","b01"]   ["b00","b01"]
Time taken: 0.056 seconds

hive> select a from table2;
OK
["b00","b01"]
["b00","b01"]
["b00","b01"]
["b00","b01"]
Time taken: 15.903 seconds

hive> select a[0] from table2;
OK
b00
b00
b00
b00
Time taken: 12.913 seconds

hive> select * from table2 where a[0] = b[0];
OK
["b00","b01"]   ["b00","b01"]
["b00","b01"]   ["b00","b01"]
["b00","b01"]   ["b00","b01"]
["b00","b01"]   ["b00","b01"]
Time taken: 11.803 seconds

 

关于map的操作说明：
drop table table2;

hive> CREATE TABLE table2 (foo STRING , bar MAP<STRING, STRING>)
    > ROW FORMAT DELIMITED
    > FIELDS TERMINATED BY '\t'
    > COLLECTION ITEMS TERMINATED BY ','
    > MAP KEYS TERMINATED BY ':'
    > STORED AS TEXTFILE;

hive> load data local inpath "../hive/examples/files/maptest.txt"  overwrite into table table2;
maptest.txt中的文件格式为：（不同列之间用一个tab分割，map中key和value用冒号分割，不同K/V间用逗号分割）
a00        b0:b01,b1:b11
a01        b1:b11,b2:b12
a02        b2:b12,b3:b13
a03        b3:b13,b4:b14

hive> select bar from table2;
OK
{"b0":"b01","b1":"b11"}
{"b1":"b11","b2":"b12"}
{"b2":"b12","b3":"b13"}
{"b3":"b13","b4":"b14"}
Time taken: 19.237 seconds
怎么根据 key来查询value呢？
hive> select bar['b1'] from table2;
OK
b11
b11
NULL
NULL
Time taken: 11.65 seconds

查看map中的键值对个数：
hive> select size(bar) from table2;
OK
2
2
2
2
Time taken: 12.137 seconds