Hive和Impala建表和导入数据

1. Hive

建表

create table [库名].[表名](
	字段名1 STRING COMMENT '注释',
	字段名2 STRING COMMENT '注释',
	字段名3 DECIMAL(10,2) COMMENT '注释',
	字段名 DATE COMMENT '注释',
	...
)
[COMMENT '注释']
[partitioned by (分区字段名 分区字段类型)]
row format delimited fields terminated by '|'
[collection items terminated by ',']
[map keys terminated by ':']
stored as textfile;

导入数据

hadoop fs -ls /user/hive
hadoop fs -mkdir /user/hive/txt
hadoop fs -put 数据文件名称.txt /user/hive/txt
hadoop fs -chmod 775 /user/hive/txt

use 库名;
load data inpath 'hdfs文件路径' into table 表名 [partition(分区字段名=值)];

附Hive数据导入的方式https://blog.csdn.net/Gavin_chun/article/details/78012460

2. Impala

建表

create table [库名].[表名](
	字段名1 STRING COMMENT '注释',
	字段名2 STRING COMMENT '注释',
	字段名3 DECIMAL(10,2) COMMENT '注释',
	字段名4 DATE COMMENT '注释',
	...
)
[COMMENT '注释']
[partitioned by (分区字段名 分区字段类型)]
row format delimited fields terminated by '|'
stored as textfile;

导入数据

hadoop fs -ls /user/impala
hadoop fs -mkdir /user/impala/txt
hadoop fs -put 数据文件名称.txt /user/impala/txt
hadoop fs -chmod 775 /user/impala/txt

use 库名;
load data inpath 'hdfs文件路径' into table 表名 [partition(分区字段名=值)];

3. 分区表数据导入

https://blog.csdn.net/weixin_42547302/article/details/114135335