Convert DSOMM Generated Activities to Excel

convert.py

import os
import sys

import yaml
from openpyxl import Workbook


# This script converts the given `generated.yaml` from DSOMM into an Excel file.
#
# Get the generated.yaml:
# curl https://raw.githubusercontent.com/devsecopsmaturitymodel/DevSecOps-MaturityModel-data/main/src/assets/YAML/generated/generated.yaml -o generated.yaml
#
# Requirements: pip install PyYAML openpyxl

class Activity:
  def __init__(self):
    self.dimension = ""
    self.sub_dimension = ""
    self.name = ""
    self.level = 0
    self.description = ""
    self.risk = ""
    self.measure = ""
    self.time = ""
    self.knowledge = ""
    self.resources = ""
    self.usefulness = ""
    self.iso27001_2017 = ""
    self.iso27001_2022 = ""

  def to_list(self):
    return [self.dimension,
            self.sub_dimension,
            self.name,
            self.level,
            self.description,
            self.risk,
            self.measure,
            self.time,
            self.knowledge,
            self.resources,
            self.usefulness,
            self.iso27001_2017,
            self.iso27001_2022]


def main():
  if len(sys.argv) < 2:
    print("Usage: python convert.py <filename>")
    sys.exit(1)

  filename = sys.argv[1]

  if not os.path.isfile(filename):
    print(f"Error: The file '{filename}' does not exist!")
    sys.exit(1)

  activities = read_activities_from_yaml(filename)
  write_excel(activities)


def read_activities_from_yaml(filename):
  with open(filename, 'r') as file:
    try:
      data = yaml.safe_load(file)
      return extract_dimensions(data)
    except yaml.YAMLError as e:
      print(f"Error parsing YAML file: {e}")


def extract_dimensions(data):
  activities = []

  for dimension_name, dimension in data.items():
    if isinstance(dimension, dict):
      activities += extract_sub_dimensions(dimension_name, dimension)
    else:
      print(f"Unexpected value for dimension: {dimension}")

  return activities


def extract_sub_dimensions(dimension_name, dimension):
  activities = []

  for sub_dimension_name, sub_dimension in dimension.items():
    if isinstance(sub_dimension, dict):
      activities += extract_activities(dimension_name, sub_dimension_name, sub_dimension)
    else:
      print(f"Unexpected value for sub dimension: {sub_dimension}")

  return activities


def extract_activities(dimension_name, sub_dimension_name, sub_dimension):
  activities = []

  for activity_name, activity in sub_dimension.items():
    if isinstance(activity, dict):
      activities.append(extract_activity(dimension_name, sub_dimension_name, activity_name, activity))
    else:
      print(f"Unexpected value for activity: {activity}")

  return activities


def extract_activity(dimension_name, sub_dimension_name, activity_name, activity):
  current_activity = Activity()
  current_activity.dimension = dimension_name
  current_activity.sub_dimension = sub_dimension_name
  current_activity.name = activity_name
  current_activity.level = map_level_label(activity['level'])

  if 'description' in activity:
    current_activity.description = activity['description']

  current_activity.risk = activity['risk']
  current_activity.measure = activity['measure']

  difficulty = activity['difficultyOfImplementation']
  current_activity.time = map_general_label(difficulty['time'])
  current_activity.knowledge = map_knowledge_label(difficulty['knowledge'])
  current_activity.resources = map_general_label(difficulty['resources'])
  current_activity.usefulness = map_general_label(activity['usefulness'])

  references = activity['references']
  iso_mapping = references.get('iso27001-2017', [])
  iso_mapping = [str(v) for v in iso_mapping]
  current_activity.iso27001_2017 = ', '.join(iso_mapping)

  iso_mapping = references.get('iso27001-2022', [])
  iso_mapping = [str(v) for v in iso_mapping]
  current_activity.iso27001_2022 = ', '.join(iso_mapping)

  return current_activity


def map_general_label(number):
  # Maps 'time', 'resources', 'usefulness'
  general_labels = ['Very Low', 'Low', 'Medium', 'High', 'Very High']
  # The numbers start from 1, so minus one bc array starts at 0.
  return general_labels[number - 1]


def map_knowledge_label(number):
  # Maps 'knowledge' to strings:
  knowledge_labels = [
    'Very Low (one discipline)',
    'Low (one discipline)',
    'Medium (two disciplines)',
    'High (two disciplines)',
    'Very High (three or more disciplines)',
  ]
  # The numbers start from 1, so minus one bc array starts at 0.
  return knowledge_labels[number - 1]


def map_level_label(number):
  # Maps 'level' to strings:
  level_labels = [
    'Level 1: Basic understanding of security practices',
    'Level 2: Adoption of basic security practices',
    'Level 3: High adoption of security practices',
    'Level 4: Very high adoption of security practices',
    'Level 5: Advanced deployment of security practices at scale',
  ]
  # The numbers start from 1, so minus one bc array starts at 0.
  return level_labels[number - 1]


def write_excel(activities):
  column_names = ["Dimension",
                  "Sub Dimension",
                  "Name",
                  "Level",
                  "Description",
                  "Risk",
                  "Measure",
                  "Time",
                  "Knowledge",
                  "Resources",
                  "Usefulness",
                  "ISO27001:2017",
                  "ISO27001:2022"]
  wb = Workbook()
  ws = wb.active
  ws.append(column_names)

  for activity in activities:
    ws.append(activity.to_list())

  filename = "activities.xlsx"
  wb.save(filename)
  print(f"Data written to {filename}")


if __name__ == "__main__":
  main()