我有如下的json

{
          "customer_id": "CUST002",
          "address_code": "1",
          "addresses": [
            {
              "Address_Line1": "street-1",
              "Address_Line2": "st mathews",
              "Address_Line3": "liberal rk",
              "City": "NJ"
            },
            {
              "Address_Line1": "A1",
              "Address_Line2": "A2",
              "Address_Line3": "A3",
              "City": "C2"
            }
          ]
        }

我已经为json创建了流，它是：

CREATE STREAM TEST_SOURCE1_ARRAY(
          addresses ARRAY < STRUCT < Address_Line1 VARCHAR, 
          Address_Line2 VARCHAR, Address_Line3 VARCHAR, 
          City VARCHAR, state VARCHAR, zip VARCHAR >>, 
          customer_id VARCHAR, address_code VARCHAR
        ) WITH (
          KAFKA_TOPIC = 'source_1_data_new', 
          VALUE_FORMAT = 'JSON_SR', partitions = 1
        );
        SELECT
        SRC_JSON.customer_id,
        CASE 
        WHEN SRC_JSON.address_code='1' THEN
        ARRAY[STRUCT(
        StreetName := EXPLODE(SRC_JSON.addresses)->Addr_1,
        Province := EXPLODE(SRC_JSON.addresses)->Addr_2,
        City := EXPLODE(SRC_JSON.addresses)->City
        )
        from FROM source_data AS SRC_JSON EMIT CHANGES;

当我执行上述查询时，记录被复制，即不是为一个客户返回一行，而是为数组中的每个对象复制记录。

{
        "CUSTOMER_ID": "CUST002",
        "Addr_1": "street-1",
        "Addr_2": "st mathews",
        "City": "NJ"
        },
        {
        "CUSTOMER_ID": "CUST002",
        "Addr_1": "A1",
        "Addr_2": "A2",
        "City": "C2"
        }

我希望结果是

{
          "CUSTOMER_ID": "CUST002",
          "ADDRESSES": [
            {
              "Address_Line1": "street-1",
              "Address_Line2": "st mathews",
              "City": "NJ"
            },
            {
              "Address_Line1": "A1",
              "Address_Line2": "A2",
              "City": "C2"
            }
          ]
        }