在使用sqoop之前，需要提前启动hadoop, yarn和对应的数据库mysql

1. 导入数据

在sqoop中，导入的概念是从非大数据集群(关系型数据库向大数据集群(thdfs,hive]中传输数据,使用import关键字

2. 从mysql向hive导入数据

2.1 导入用户信息表

sqoop import  \
--connect jdbc:mysql://bigdata03:3306/mall \
--username root  \
--password 111111 \
--table t_user_info \
--num-mappers 1 \
--hive-import  \
--fields-terminated-by "," \
--hive-overwrite \
--hive-table mall_bigdata.ods_user_info

里面的 \ 是代表换行符，这里指令可以写在一行，也可以使用换行符将参数部分分来来写，显得更加直观，num-mappers是指定mapper任务个数，这里表只有一个，数据量也少，任务可以设为1，当表多，数据量大时，可以适当增大num-mappers参数，fields-terminated-by是指定分隔符

注：bigdata03虚拟机会开两个tab窗口，一个用于输入相关shell命令，一个用于开启hive命令行界面进行相关数据查询等。

hive 导入完成

查看导入的用户信息表数据

select * from mall_bigdata.ods_user_info;

2.导入订单表

2.2 导入订单表

sqoop import  \
--connect jdbc:mysql://bigdata03:3306/mall \
--username root  \
--password 111111 \
--table t_sale_order \
--num-mappers 1 \
--hive-import  \
--fields-terminated-by "," \
--hive-overwrite \
--hive-table mall_bigdata.ods_sale_info

查看导入的订单表数据

select * from mall_bigdata.ods_sale_info;

2.3 导入商品信息表

sqoop import  \
--connect jdbc:mysql://bigdata03:3306/mall \
--username root  \
--password 111111 \
--table dim_goods_info \
--num-mappers 1 \
--hive-import  \
--fields-terminated-by "," \
--hive-overwrite \
--hive-table mall_bigdata.dim_goods_info

select * from mall_bigdata.dim_goods_info;

2.4 导入国家信息表

sqoop import  \
--connect jdbc:mysql://bigdata03:3306/mall \
--username root  \
--password 111111 \
--table dim_country_info \
--num-mappers 1 \
--hive-import  \
--fields-terminated-by "," \
--hive-overwrite \
--hive-table mall_bigdata.dim_country_info

select * from mall_bigdata.dim_country_info;

2.5 导入省份信息表

sqoop import  \
--connect jdbc:mysql://bigdata03:3306/mall \
--username root  \
--password 111111 \
--table dim_province_info \
--num-mappers 1 \
--hive-import  \
--fields-terminated-by "," \
--hive-overwrite \
--hive-table mall_bigdata.dim_province_info

select * from mall_bigdata.dim_province_info;

2.6 导入城市信息表

sqoop import  \
--connect jdbc:mysql://bigdata03:3306/mall \
--username root  \
--password 111111 \
--table dim_city_info \
--num-mappers 1 \
--hive-import  \
--fields-terminated-by "," \
--hive-overwrite \
--hive-table mall_bigdata.dim_city_info

select * from mall_bigdata.dim_city_info;

创建tmp_dwd_user_info.sql 并上传到 /opt/file

-- 切换hive的数据库
use mall_bigdata;

-- 补全用户信息表中的国家名称,省份名称和城市名称
create table if not exists mall_bigdata.tmp_dwd_user_info
as
select
	 user_id
	,user_name
	,sex
	,age
	,country_name
	,province_name
	,city_name
from
(select
	 user_id
	,user_name
	,sex
	,age
	,country_code
	,province_code
	,city_code
from ods_user_info
) a
left join
(select
	country_code
	,country_name
	from dim_country_info
) b
on a.country_code=b.country_code
left join
(select
	province_code
	,province_name
	,country_code
	from dim province_info
) c
on a.province_code=c.province_code and a.country_code=c.country_code
left join
(select
	city_code
	,city_name
	,province_code
	from dim_city_info
) d
on a.city_code=d.city_code and a.province_code=d.province_code;

2.7 创建hive临时表文件

创建hive临时表文件tmp_dwd_user_info.txt

-- 切换hive的数据库
use mall_bigdata;

-- 补全用户信息表中的国家名称,省份名称和城市名称
create table if not exists mall_bigdata.tmp_dwd_user_info
as
select
	 user_id
	,user_name
	,sex
	,age
	,country_name
	,province_name
	,city_name
from
(select
	 user_id
	,user_name
	,sex
	,age
	,country_code
	,province_code
	,city_code
from ods_user_info
) a
left join
(select
	country_code
	,country_name
	from dim_country_info
) b
on a.country_code=b.country_code
left join
(select
	province_code
	,province_name
	,country_code
	from dim_province_info
) c
on a.province_code=c.province_code and a.country_code=c.country_code
left join
(select
	city_code
	,city_name
	,province_code
	from dim_city_info
) d
on a.city_code=d.city_code and a.province_code=d.province_code;

执行该hive文件

select * from tmp_dwd_user_info;

创建hive表文件dwd_sale_order_detail.sql到 /opt/file/目录

-- 切换hive的数据库
use mall_bigdata;

-- 补全用户信息表中的国家名称,省份名称和城市名称
create table if not exists mall_bigdata.tmp_dwd_user_info
as
select
	 user_id
	,user_name
	,sex
	,age
	,country_name
	,province_name
	,city_name
from
(select
	 user_id
	,user_name
	,sex
	,age
	,country_code
	,province_code
	,city_code
from ods_user_info
) a
left join
(select
	country_code
	,country_name
	from dim_country_info
) b
on a.country_code=b.country_code
left join
(select
	province_code
	,province_name
	,country_code
	from dim_province_info
) c
on a.province_code=c.province_code and a.country_code=c.country_code
left join
(select
	city_code
	,city_name
	,province_code
	from dim_city_info
) d
on a.city_code=d.city_code and a.province_code=d.province_code;


--补全订单表中的商品名称
--过滤国家名称为中国的订单记录
create table if not exists mall_bigdata.dwd_sale_order_detail
as
select
	sale_id,
	a.user_id,
	user_name,
	sex,
	age,
	country_name,
	province_name,
	city_name,
	a.goods_id,
	goods_name,
	price,
	sale_count,
	total_price,
	create_time
from
(select
	sale_id
	,user_id
	,goods_id
	,price
	,sale_count
	,total_price
	,create_time
from ods_sale_order) a
left join
(
	select
		goods_id
		,goods_name
	from dim_goods_info
) b
on a.goods_id=b.goods_id
left join
(
	select
		user_id
		,user_name
		,sex
		,age
		,country_name
		,province_name
		,city_name
	from tmp_dwd_user_info
)c
on a.user_id=c.user_id
where country_name='中国';

--删除临时表
--drop table if exists mall_bigdata.tmp_dwd_user_info;

执行该sql文件

hive -f dwd_sale_order_detail.sql