十月十二日

本地excel/csv文件-->hive-->mysql

前提：Hadoop集群、hive配置完成

1、在node01开启hadoop集群： start-dfs.sh , start-yarn.sh

2、node01 : 开启 metastore

hive所在路径/hive/bin/hive --service metastore

3、克隆node01-->node01(1)，开启 hive

hive所在路径/hive/bin/hive

4、将本地文件上传至虚拟机

5、在hive中创建表

create table lingshou_csv (InvoiceNo String,
                         StockCode String,
                         Description String,
                         Quantity String,
                         InvoiceDate String,
                         UnitPrice String,
                         CustomerID String,
                         Country String) 
                         ROW format delimited fields terminated by ',' STORED AS TEXTFILE;

6、本地文件导入hive

load data  local inpath  '/kkb/install/hive/lingshou.csv' into table lingshou_csv;

7、查看是否导入成功

select * from lingshou_csv limit 10;

注意要用 limit 限制，否则数据过多等待时间太长

8、在mysql中的node01数据库（hive中对应mysql）中新建表

create table lingshou_csv (InvoiceNo VARCHAR(255),
                         StockCode VARCHAR(255),
                         Description  VARCHAR(255),
                         Quantity  VARCHAR(255),
                         InvoiceDate  VARCHAR(255),
                         UnitPrice  VARCHAR(255),
                         CustomerID  VARCHAR(255),
                         Country  VARCHAR(255)); 
                         ROW format delimited fields terminated by ',' STORED AS TEXTFILE;

9、hive数据表导入mysql

bin/sqoop export \
--connect jdbc:mysql://node01:3306/lingshu_csv \
--username root \
--password 20194023 \
--table testtable \
--num-mappers 1 \
--export-dir /user/hive/warehouse/lingshu_csv \
--input-fields-terminated-by ","