Creating a new column in PySpark, and indexing a row in the dataframe that returns a value 1 before

Question

I have a pyspark dataframe that I am cleaning and it is based off of logic previously built into Alteryx.

Inside the dataframe, you picture there being the following columns: col1 = 'Part', col2 = 'Tool'

I am creating a new column called 'Fill' and this column is created with this alteryx logic assigned to it:

'''if !IsNull(Part) then Part ELSE [row-1:Fill]'''

How would in Pyspark would I go about taking the previous row and putting its 'Fill' value in the new row if the condition is met?