AVRO表的创建

AVRO官方介绍

原始数据存放在 inner_person_t这张表中

select * from inner_person_t;

方式1

1. 创建schema

文件名为：schema.avsc
内容如下：

{
	"type": "record",
	"name": "avroPeople",
	"namespace": "com.china",
	"fields": [{
			"name": "id",
			"type": "int"
		}, {
			"name": "age",
			"type": "int"
		}, {
			"name": "sex",
			"type": "string"
		}
	]
}

2.将schema文件放到指定的hdfs目录上

hadoop fs -put schema.avsc /user/xiangyongqiao15/hivedata/avro_schema

3.创建表

CREATE TABLE IF NOT EXISTS avro_people
ROW FORMAT SERDE 'org.apache.hadoop.hive.serde2.avro.AvroSerDe'
WITH SERDEPROPERTIES ('avro.schema.url'='/user/xiangyongqiao15/hivedata/avro_schema/schema.avsc')
STORED AS INPUTFORMAT 'org.apache.hadoop.hive.ql.io.avro.AvroContainerInputFormat'
OUTPUTFORMAT 'org.apache.hadoop.hive.ql.io.avro.AvroContainerOutputFormat';

4.导入数据

 insert overwrite table avro_people select * from inner_person_t;

5.查看数据

select * from avro_people;

6.添加字段时的注意事项

AVRO表新添加字段时，一定要给这个新字段设置默认值

{
	"type": "record",
	"name": "avroPeople",
	"namespace": "com.china",
	"fields": [{
			"name": "id",
			"type": "int"
		}, {
			"name": "age",
			"type": "int"
		}, {
			"name": "sex",
			"type": "string"
		}
		, {
			"name": "country",
			"type": "string",
			"default":"China"
		}
	]
}