目录
正文
现有文件为csv格式,需要导入hive中,设csv内容如下
?
1 2 |
1001,zs,23 1002,lis,24 |
首先创建表
?
1 2 3 4 5 6 7 |
create table if not exists csv2( uid int , uname string, age int ) row format serde 'org.apache.hadoop.hive.serde2.OpenCSVSerde' stored as textfile ; |
导入数据及查询
?
1 2 |
load data local inpath '/data/csv2.csv' into table csv2; select * from csv2; |
其他注意事项
如果建表是parquet格式可否load导入csv文件?
?
1 2 3 4 5 6 7 8 9 10 11 |
drop table csv2; create table if not exists csv2( uid int , uname string, age int ) row format serde 'org.apache.hadoop.hive.serde2.OpenCSVSerde' stored as parquet ; load data local inpath '/data/csv2.csv' into table csv2; select * from csv2; |
使用时会报错
Failed with exception java.io.IOException:java.lang.RuntimeException: hdfs://192.168.10.101:8020/user/hive/warehouse/csv2/csv2.csv is not a Parquet file. expected magic number at tail [80, 65, 82, 49] but found [44, 50, 52, 10]
**不可以,需要先导入成textfile,之后再从临时表导入成parquet,**如下
?
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 |
drop table csv2; create table if not exists csv2 ( uid int , uname string, age int ) row format serde 'org.apache.hadoop.hive.serde2.OpenCSVSerde' stored as textfile; -- 先导入csv文件到表格csv2,保存格式是textfile load data local inpath '/data/csv2.csv' into table csv2; drop table csv3; -- 创建csv3,保存格式parquet create table if not exists csv3 ( uid int , uname string, age int ) row format delimited fields terminated by ',' stored as parquet; -- 提取csv2的数据插入到csv3 insert overwrite table csv3 select * from csv2; |
总结
- 关键是要引入org.apache.hadoop.hive.serde2.OpenCSVSerde
csv
要保存到hive
的parquet
,需要先保存成textfile