tools/chapter_division.py

# import os
# import json
# import re
# from modules import outliner_dir, output_dir

# # ---------- Helper: sanitize file names ----------
# def safe_filename(name: str) -> str:
#     return re.sub(r'[\\/*?:"<>|]', "", name).strip()


# # ---------- Main function ----------
# def split_json(input_file: str, output_dir: str):
#     # Load JSON
#     with open(input_file, "r", encoding="utf-8") as f:
#         data = json.load(f)

#     main_headings = data.get("main_headings", [])
#     sub_headings = data.get("sub_headings", [])
#     sub_sub_headings = data.get("sub_sub_headings", None)

#     # Ensure output folder exists
#     os.makedirs(output_dir, exist_ok=True)

#     for idx, main in enumerate(main_headings):
#         headers = sub_headings[idx] if idx < len(sub_headings) else []

#         # لو فيه sub_sub_headings نجيبها، لو مش فيه نخليها null
#         if sub_sub_headings and idx < len(sub_sub_headings):
#             subs = sub_sub_headings[idx]
#         else:
#             subs = None

#         section = {"headers": headers, "sub_sub_headings": subs}

#         file_name = safe_filename(main) + ".json"
#         file_path = os.path.join(output_dir, file_name)

#         with open(file_path, "w", encoding="utf-8") as out_f:
#             json.dump(section, out_f, ensure_ascii=False, indent=2)

#         print(f"✅ Saved {file_name}")


# # ---------- Example usage ----------
# if __name__ == "__main__":
#     split_json("/content/ai_agent_output/step_1_outlines.json", "output_sections")