ProwlerHTML2CSV.py

# AI WAS USED IN THE MAKING OF THIS PROGRAM. AI USED: ChatGPT-4o
# NAME:
#   ProwlerHTML2CSV
# CREATED BY:
#   Garrett Greathouse

import bs4
from bs4 import BeautifulSoup
import csv
import glob

# Open Prowler HTML output
html_file_list = glob.glob("prowler-output-*.html")
html_file_name = html_file_list[html_file_list.index(max(html_file_list))]  # Gets most recent file by name
with open(html_file_name, "r", encoding="utf-8") as file:
    html_file = file.read()

# Parse HTML
soup = BeautifulSoup(html_file, "html.parser")

# Grabs table with data
table = soup.find(id="findingsTable")

# TESTING PURPOSES ONLY
# with open("bs4-test.html", "w", encoding="utf-8") as output_file:
#     output_file.write(table.prettify())

# Extract headers
headers = [th.get_text(strip=True) for th in table.find_all("th")]

# Extract data
rows = []
for tr in table.find_all("tr", class_="table-danger")[1:]:
    cells = [td.get_text(strip=True) for td in tr.find_all("td")]
    if cells:
        rows.append(cells)

# Transfer data to new CSV
output_file = "prowler-csv-output.csv"
with open(output_file, "w", newline='', encoding="utf-8") as csv_file:
    writer = csv.writer(csv_file)
    writer.writerow(headers)  # Header row
    writer.writerows(rows)    # Date rows
    print(f"Data has been successfully written to {output_file}")