How can I use an ExecuteScript processor to run a python script that reads unzipped data from an Unpack Content processor in Nifi?

Question

I am trying to read and use the data that has come from an Unpack content processor with my python script. I know that I am supposed to use an Execute script processor to run the python script but I am struggling to understand:

1. How to format my Python code to find the path of the unzipped file in Nifi?
2. How to format my Python code to set the output path of the processed file in Nifi?

This is what my current Nifi dataflow looks like: enter image description here

#!/usr/bin/env python
# coding: utf-8
from sysconfig import get_python_version
import pandas as pd
import os
from pathlib import Path
get_python_version().run_line_magic('pip', 'install pyarrow')

# Specifying the path of the unzipped insights file
path = Path(r'C:\Users\IT Admin\Desktop\Parquet_to_CSV_V2\insights_2023-06-29')
os.chdir(path)
cwd = Path.cwd()
cwd

# Adding the paths of every parquet file under the "complex_relations" folder to a list
target_dir = cwd / "complex_relation"
pq_files = []
for file in target_dir.rglob("*.parquet*"):
    pq_files.append(file)

# A loop for adding all the parquet files to one parquet file 
data_frames=[]

for parquet in pq_files:
    df = pd.read_parquet(parquet)
    data_frames.append(df)
    
concatenated_df = pd.concat(data_frames)


# Specifying the combined parquet file to a specific file_path
output_path = r'C:\Users\IT Admin\Desktop\Parquet_to_CSV_V2\output\complex_relations.parquet'
concatenated_df.to_parquet(output_path, engine = 'pyarrow')
compliled_pq = r'C:\Users\IT Admin\Desktop\Parquet_to_CSV_V2\output\complex_relations.parquet'
pd.read_parquet(compliled_pq, engine = "auto")