Develop a standalone Spark SQL application (using IntelliJ IDEA) that converts one or more string columns to upper case.
The standalone application should take at least two input parameters:
The output dataset should extend the current columns with new ones with their names including the “source”, e.g. if city
column were used, the output could be upper_city
.
Extra: Make sure the conversion happens on string columns only
Module: Spark SQL
Duration: 30 mins
+---+-----------------+-------+
| id| city|country|
+---+-----------------+-------+
| 0| Warsaw| Poland|
| 1|Villeneuve-Loubet| France|
| 2| Vranje| Serbia|
| 3| Pittsburgh| US|
+---+-----------------+-------+
id,city,country
0,Warsaw,Poland
1,Villeneuve-Loubet,France
2,Vranje,Serbia
3,Pittsburgh,US
+---+-----------------+-------+-----------------+
| id| city|country| upper_city|
+---+-----------------+-------+-----------------+
| 0| Warsaw| Poland| WARSAW|
| 1|Villeneuve-Loubet| France|VILLENEUVE-LOUBET|
| 2| Vranje| Serbia| VRANJE|
| 3| Pittsburgh| US| PITTSBURGH|
+---+-----------------+-------+-----------------+