hive如何创建json表

背景：
在业务中，对于json数据一般通过get_json_object或者json_tuple来解析json数据。对于有json和常规数据混合这种，用这两个函数比较常见。但是对于全是文本的这种json数据，我们可以通过创建json表与json表进行映射去查询或者统计数据。

1.测试数据如下：

{"student":{"name":"xiaowang","age":11,"sex":"M"},"class":{"book":"语文","level":2,"score":81},"teacher":{"name":"t1","class":"语文"}}
{"student":{"name":"xiaoming","age":12,"sex":"M"},"class":{"book":"语文","level":2,"score":82},"teacher":{"name":"t2","class":"语文"}}
{"student":{"name":"xiaolan","age":13,"sex":"M"},"class":{"book":"语文","level":2,"score":83},"teacher":{"name":"t3","class":"语文"}}
{"student":{"name":"xiaohei","age":14,"sex":"M"},"class":{"book":"语文","level":2,"score":84},"teacher":{"name":"t1","class":"语文"}}
{"student":{"name":"xiaobai","age":15,"sex":"M"},"class":{"book":"语文","level":2,"score":86},"teacher":{"name":"t2","class":"语文"}}
{"student":{"name":"xiaohong","age":16,"sex":"M"},"class":{"book":"语文","level":2,"score":87},"teacher":{"name":"t3","class":"语文"}}

2.建表语句如下：

我们需要的映射类：org.apache.hive.hcatalog.data.JsonSerDe

create external table test.test_student_json(
  student map<string,string> comment "学生信息",
  class map<string,string> comment "课程信息",
  teacher map<string,string> comment "授课老师信息")
row format serde 
'org.apache.hive.hcatalog.data.JsonSerDe';

3.如何查询数据：

查询外层json数据：

spark-sql> select student from test.test_student_json;
{"age":"11","name":"xiaowang","sex":"M"}                                        
{"age":"12","name":"xiaoming","sex":"M"}
{"age":"13","name":"xiaolan","sex":"M"}
{"age":"14","name":"xiaohei","sex":"M"}
{"age":"15","name":"xiaobai","sex":"M"}
{"age":"16","name":"xiaohong","sex":"M"}
Time taken: 2.424 seconds, Fetched 6 row(s)

查询里层json数据：

spark-sql> select student['name'] from test.test_student_json;
xiaowang
xiaoming
xiaolan
xiaohei
xiaobai
xiaohong
Time taken: 0.173 seconds, Fetched 6 row(s)

对于全是文本的这种json数据，使用json表映射的方式还是比较方便的