How can i add multiple columns in Spark Datframe in efficiently

Question

I have set of columns names and need to add those columns in existing dataframe which is also very huge in size, i need to add the all columns from set to dataframe with Str

Answer 1

So this is in PySpark.

df.select(
    '*', 
    *list(F.lit(None).alias(f'col{n}') for n in range(7,13))
).show()

+---+----------+----+----+----+-----+----+-----+----+-----+-----+-----+-----+----+-----+----+----+-----+-----+-----+-----+-----+----+-----+
|Id |Name      |col7|col8|col3|col17|col6|col20|col2|col14|col16|col21|col15|col9|col10|col5|col1|col13|col19|col11|col22|col18|col4|col12|
+---+----------+----+----+----+-----+----+-----+----+-----+-----+-----+-----+----+-----+----+----+-----+-----+-----+-----+-----+----+-----+
|1  |James     |null|null|null|null |null|null |null|null |null |null |null |null|null |null|null|null |null |null |null |null |null|null |
|2  |Michael   |null|null|null|null |null|null |null|null |null |null |null |null|null |null|null|null |null |null |null |null |null|null |
|3  |Robert    |null|null|null|null |null|null |null|null |null |null |null |null|null |null|null|null |null |null |null |null |null|null |
|4  |Washington|null|null|null|null |null|null |null|null |null |null |null |null|null |null|null|null |null |null |null |null |null|null |
|5  |Jefferson |null|null|null|null |null|null |null|null |null |null |null |null|null |null|null|null |null |null |null |null |null|null |
+---+----------+----+----+----+-----+----+-----+----+-----+-----+-----+-----+----+-----+----+----+-----+-----+-----+-----+-----+----+-----+

This logic translates to Scala spark if you understand how to replace the list comprehension in Scala with map.

This is faster as it creates the 22 columns to be executed at once, rather that adding them in iterations as foldleft does.