Hive常用操作

1. 绑定数据

1.1 创建表&创建文本文件

create table  fantj.t3(id int,name string,age int) row format delimited fields terminated by ','stored as textfile;

hive> create table  fantj.t3(id int,name string,age int) row format delimited fields terminated by ','stored as textfile;
OK
Time taken: 4.467 seconds
hive> select * from fantj.t3;
OK
Time taken: 2.82 seconds

表示行格式用逗号来分割字段。

创建文本文件test.txt

我创建在/home/fantj目录下。

1,jiao,18
2,fantj,20
3,laowang,30
4,laotie,40

1.2 从本地导入到hive

LOAD DATA LOCAL INPATH '/home/fantj/test.txt' OVERWRITE INTO TABLE t3;

hive> LOAD DATA LOCAL INPATH '/home/fantj/test.txt' OVERWRITE INTO TABLE fantj.t3;
Loading data to table fantj.t3
[Warning] could not update stats.
OK
Time taken: 26.334 seconds

select * from fantj.t3;

hive> select * from fantj.t3;
OK
1   jiao    18
2   fantj   20
3   laowang 30
4   laotie  40
Time taken: 2.303 seconds, Fetched: 4 row(s)

导入成功！

1.3 从hdfs导入到hive

先将test文件上传到hdfs中

[root@s166 fantj]# hadoop fs -put test.txt /hdfs2hive

-rw-r--r--   3 root supergroup         46    /hdfs2hive/test.txt

进入hive，创建表t5

create table  fantj.t5(id int,name string,age int) row format delimited fields terminated by ','stored as textfile;

hive> create table  fantj.t5(id int,name string,age int) row format delimited fields terminated by ','stored as textfile;
OK
Time taken: 3.214 seconds

执行导入

LOAD DATA INPATH '/hdfs2hive/test.txt' OVERWRITE INTO TABLE fantj.t5;

hive> LOAD DATA INPATH '/hdfs2hive/test.txt' OVERWRITE INTO TABLE fantj.t5;
Loading data to table fantj.t5
[Warning] could not update stats.
OK
Time taken: 25.498 seconds

检查是否成功：

hive> select * from fantj.t5;
OK
1   jiao    18
2   fantj   20
3   laowang 30
4   laotie  40
Time taken: 5.046 seconds, Fetched: 4 row(s)