| set hive.vectorized.execution.enabled=false; |
| |
| -- Sometimes, the user wants to create a table from just a portion of the file schema; |
| -- This test_n6 makes sure that this scenario works; |
| |
| DROP TABLE test_n6; |
| |
| -- Current file schema is: (id int, name string, address struct<number:int,street:string,zip:string>); |
| -- Creates a table from just a portion of the file schema, including struct elements (test_n6 lower/upper case as well) |
| CREATE TABLE test_n6 (Name string, address struct<Zip:string,Street:string>) STORED AS PARQUET; |
| |
| LOAD DATA LOCAL INPATH '../../data/files/HiveGroup.parquet' OVERWRITE INTO TABLE test_n6; |
| SELECT * FROM test_n6; |
| |
| DROP TABLE test_n6; |
