Flink SQL中的Map、Array、Row_flinksql中row类型的使用-优快云博客

本文链接：https://blog.youkuaiyun.com/zuixue_lei/article/details/127776927

本文通过一个实例展示了如何使用Flink SQL处理Kafka中的JSON数据，包括创建表、插入数据及执行查询操作。示例涉及了ROW、ARRAY、MAP等复杂数据类型，并给出了具体的SQL建表和查询语句。

摘要生成于 C知道，由 DeepSeek-R1 满血版支持，前往体验 >

1、kafka中样例数据

{
    "id":1238123899121,
    "name":"asdlkjasjkdla998y1122",
    "date":"1990-10-14",
    "obj":{
        "time1":"12:12:43Z",
        "str":"sfasfafs",
        "lg":2324342345
    },
    "arr":[
        {
            "f1":"f1str11",
            "f2":134
        },
        {
            "f1":"f1str22",
            "f2":555
        }
    ],
    "time":"12:12:43Z",
    "timestamp":"1990-10-14T12:12:43Z",
    "map":{
        "flink":123
    },
    "mapinmap":{
        "inner_map":{
            "key":234
        }
    }
}

2、建表语句

CREATE TABLE json_source (
    id            BIGINT,
    name          STRING,
    `date`        DATE,
    obj           ROW<time1 TIME,str STRING,lg BIGINT>,
    arr           ARRAY<ROW<f1 STRING,f2 INT>>,
    `time`        TIME,
    `timestamp`   TIMESTAMP(3),
    `map`         MAP<STRING,BIGINT>,
    mapinmap      MAP<STRING,MAP<STRING,INT>>,
    proctime as PROCTIME()
 ) WITH (
    'connector.type' = 'kafka',  
    'connector.topic' = 'test',  
    'connector.properties.zookeeper.connect' = 'localhost:2181',
    'connector.properties.bootstrap.servers' = 'localhost:9092',   
    'connector.properties.group.id' = 'testGroup',    
    'connector.version'='universal',  
    'format.type' = 'json',     
    'connector.startup-mode'='latest-offset' 
);

3、查询

select id, 
    name,
    `date`,
    obj.str,
    arr[1].f1,
    `map`['flink'],
    mapinmap['inner_map']['key'] from json_source;
    //注意数组index从1开始

4、插入数据

insert into json_source 
select 
    111 as id,
    'name' as name,
    Row(CURRENT_TIME,'ss',123) as obj,
    Array[Row('f',1),Row('s',2)] as arr,
    Map['k1','v1','k2','v2'] as `map`,
    Map['inner_map',Map['k','v']] as mapinmap
;

借鉴于：

Flink实战之Flink SQL中的Map、Array、Row_优优我心的博客-优快云博客_flinksql row