Chaining decorated functions

Calls of the decorated functions can be chained by passing function names as decorator arguments:

@transformation(read_table("bronze.tbl_customers"))
def tbl_customers(df: DataFrame):
    return df

@transformation(tbl_customers)
def active_customers_only(df: DataFrame):
    return df.filter(f.col("active") == 1)

@transformation(active_customers_only, display=True)
def save(df: DataFrame):
    return df

More compact way:

@transformation(read_table("bronze.tbl_customers"), display=True)
def tbl_active_customers(df: DataFrame):
    return df.filter(f.col("active") == 1)

Once you run the active_customers_only function's cell, it gets is automatically called with the dataframe loaded by the customers_table function.

Similarly, once you run the save_output function's cell, it gets automatically called with the filtered dataframe returned from the active_customers_only function.