Was this helpful?
JSON Files
employee_test.json
{"name":"Michael", "salary":3000}
{"name":"Andy", "salary":4500}
{"name":"Justin", "salary":3500}
{"name":"Berta", "salary":4000}
{"name":"Raju", "salary":3000}
{"name":"Chandy", "salary":4500}
{"name":"Joey", "salary":3500}
{"name":"Mon", "salary":4000}
{"name":"Rachel", "salary":4000}
1. Create external table:
CREATE EXTERNAL TABLE employee_ex_json (
    name CHAR(25) NOT NULL,
    salary INTEGER NOT NULL
) USING SPARK WITH
REFERENCE='abfs://mycontainer@myaccount.dfs.core.windows.net/myblob/employee_test.json',
FORMAT = 'JSON';
IMPORTANT!  Ensure that your external table column names match the names used for columns in your source data header row.
2. Create native Actian Data Platform table
CREATE TABLE employee_json(
name CHAR(25) NOT NULL,
salary INTEGER NOT NULL
) WITH STRUCTURE=X100;
3. Load the Actian Data Platform table with the INSERT command:
INSERT INTO employee_json SELECT * FROM employee_ex_json
Last modified date: 10/30/2024