#!/usr/bin/env python3

import json
import os
import shutil

def main():
    # Load structure.json
    with open('docs/archive/structure.json', 'r') as f:
        data = json.load(f)

    # Create organized directory
    organized_dir = 'docs/archive/organized'
    os.makedirs(organized_dir, exist_ok=True)

    # Copy assets
    assets_src = 'docs/archive/assets'
    assets_dst = os.path.join(organized_dir, 'assets')
    if os.path.exists(assets_src):
        shutil.copytree(assets_src, assets_dst, dirs_exist_ok=True)

    pages = []

    for url, page_data in data.items():
        # Skip anchors and non-site URLs
        if '#' in url or not url.startswith('https://nixtamal.toast.al'):
            continue

        # Extract path
        if url == 'https://nixtamal.toast.al' or url == 'https://nixtamal.toast.al/':
            path = 'home'
        else:
            path = url[len('https://nixtamal.toast.al/'):].rstrip('/')
            if not path:
                path = 'home'

        # Create directory structure
        full_path = os.path.join(organized_dir, path)
        os.makedirs(full_path, exist_ok=True)

        # Get title and text
        title = page_data['title'].strip()
        text = page_data['text']

        # Clean text: remove excessive whitespace
        lines = text.split('\n')
        cleaned_lines = []
        for line in lines:
            stripped = line.strip()
            if stripped:
                cleaned_lines.append(stripped)
        text = '\n\n'.join(cleaned_lines)

        # Filename
        if '/' in path:
            filename = path.split('/')[-1] + '.md'
        else:
            filename = path + '.md'

        # Write markdown file
        md_path = os.path.join(full_path, filename)
        with open(md_path, 'w', encoding='utf-8') as f:
            f.write(f'# {title}\n\n{text}\n')

        # Collect for index
        pages.append((path, title, md_path.replace(organized_dir + '/', '')))

    # Create index.md
    index_path = os.path.join(organized_dir, 'index.md')
    with open(index_path, 'w', encoding='utf-8') as f:
        f.write('# Nixtamal Documentation Archive\n\n')
        f.write('This is an organized archive of the Nixtamal documentation.\n\n')
        f.write('## Contents\n\n')

        # Group by top-level section
        sections = {}
        for path, title, rel_path in pages:
            top = path.split('/')[0] if '/' in path else path
            if top not in sections:
                sections[top] = []
            sections[top].append((path, title, rel_path))

        for top in sorted(sections.keys()):
            f.write(f'### {top.capitalize()}\n\n')
            for path, title, rel_path in sorted(sections[top]):
                f.write(f'- [{title}]({rel_path})\n')
            f.write('\n')

if __name__ == '__main__':
    main()