Aggregating the same column twice
There are cases where dot notation can be more cumbersome than SQL. This exercise calculates the first and last times for each train line. The following code does this using dot notation.
from pyspark.sql.functions import min, max, col
expr = [min(col("time")).alias('start'), max(col("time")).alias('end')]
dot_df = df.groupBy("train_id").agg(*expr)
dot_df.show()
+--------+-----+-----+
|train_id|start| end|
+--------+-----+-----+
| 217|6:06a|6:59a|
| 324|7:59a|9:05a|
+--------+-----+-----+
Your mission is to achieve this same result using a SQL query. The dataframe df has been registered as a table named schedule.
Deze oefening maakt deel uit van de cursus
Introduction to Spark SQL in Python
Oefeninstructies
- Write a SQL query that gives an identical result to the dot notation query.
Praktische interactieve oefening
Probeer deze oefening eens door deze voorbeeldcode in te vullen.
# Write a SQL query giving a result identical to dot_df
query = "SELECT ____ FROM schedule ____ ____ ____"
sql_df = spark.sql(query)
sql_df.show()