#!/usr/bin/env python3

# Static Blog Generator for Neocities (or any other web host)
# Converts individual HTML files into a full blog with RSS support.

# Version 1.0 (11 May 2026) - Blue Oak Model License 1.0.0
# Project page: https://640kb.neocities.org/apps/apps.html

# Requirements: python 3.6 or above

import os
import re
import glob
from datetime import datetime
from xml.sax.saxutils import escape as xml_escape

# ===========================================================================
# CONFIGURATION: Change these values as needed. I recommend keeping the core
# defaults during your first testing (except SCRIPT_TITLE, of course).
# ===========================================================================

# Base URL for your blog (used in RSS feed links)
# Change this if your blog is not located at /blog/
BLOG_BASE_URL = "/blog/"

# Number of posts per page
POSTS_PER_PAGE = 20

# Number of recent posts to include in RSS feed
RSS_ITEMS = 10

# Directory containing your post fragments
POSTS_DIR = "posts"

# Your HTML template file
TEMPLATE_FILE = "template.html"

# Script title printed to terminal output after index*.html/rss creation
SCRIPT_TITLE = "640KB Static Blog Generator"

# ===========================================================================

def clean_old_index_files():
    for filepath in glob.glob("index*.html"):
        os.remove(filepath)
        print(f"Removed old file: {filepath}")

def get_all_posts():
    posts = []
    pattern = os.path.join(POSTS_DIR, "*", "*.html")
    
    for filepath in glob.glob(pattern):
        year = os.path.basename(os.path.dirname(filepath))
        
        filename = os.path.basename(filepath)
        
        match = re.match(r"(\d{2})\.(\d{2})--", filename)
        if not match:
            print(f"Warning: Skipping {filepath} - filename doesn't start with MM.DD--")
            continue
        
        month, day = match.group(1), match.group(2)
        
        try:
            post_date = datetime(int(year), int(month), int(day))
        except ValueError:
            print(f"Warning: Skipping {filepath} - invalid date {year}-{month}-{day}")
            continue
        
        with open(filepath, 'r', encoding='utf-8') as f:
            content = f.read()
        
        posts.append({
            'year': year,
            'month': month,
            'day': day,
            'date_obj': post_date,
            'filename': filename,
            'filepath': filepath,
            'content': content
        })
    
    posts.sort(key=lambda x: x['date_obj'], reverse=True)
    return posts

def render_posts(posts_slice, start_index=1, total_posts=0, current_page=1, total_pages=1):
    html_parts = []
    
    for post in posts_slice:
        post_slug = post['filename'].replace('.html', '')
        post_id = f"post-{post['date_obj'].strftime('%Y%m%d')}-{post_slug}"
        
        html_parts.append(f'<div id="{post_id}" class="post-content">')
        html_parts.append(post['content'])
        html_parts.append('</div>')
        
    posts_html = ''.join(html_parts)
    
    if total_pages > 1:
        nav_html = '<div class="nav">\n'
        
        if current_page > 1:
            if current_page == 2:
                nav_html += '  <a href="index.html">[Newer-posts]</a>\n'
            else:
                nav_html += f'  <a href="index-{current_page-1}.html">[Newer-posts]</a>\n'
        
        if current_page < total_pages:
            nav_html += f'  <a href="index-{current_page+1}.html">[Older-posts]</a>\n'
        
        nav_html += '</div>\n'
        posts_html += nav_html
    
    return posts_html

def strip_html(text):
    return re.sub(r'<[^>]+>', '', text)

def generate_rss(posts, base_url=""):
    rss_items = []
    
    for post in posts[:RSS_ITEMS]:
        title_match = re.search(r'<h2[^>]*>(.*?)</h2>', post['content'], re.IGNORECASE)
        raw_title = title_match.group(1) if title_match else "Untitled"
        title = xml_escape(strip_html(raw_title))
        
        post_slug = post['filename'].replace('.html', '')
        post_id = f"post-{post['date_obj'].strftime('%Y%m%d')}-{post_slug}"
        
        full_content = post['content']
        
        pub_date = post['date_obj'].strftime('%a, %d %b %Y 12:00:00 GMT')
        
        rss_items.append(f"""<item>
    <title>{title}</title>
    <link>{base_url}index.html#post-{post['date_obj'].strftime('%Y%m%d')}-{post_slug}</link>
    <guid>{base_url}index.html#post-{post['date_obj'].strftime('%Y%m%d')}-{post_slug}</guid>
    <pubDate>{pub_date}</pubDate>
    <description><![CDATA[{full_content}]]></description>
</item>""")
    
    rss_xml = f"""<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0">
<channel>
    <title>640kb&apos;s Blog</title>
    <link>{base_url}</link>
    <description>640KB blog - notes, updates, thoughts</description>
    <language>en-us</language>
    <lastBuildDate>{datetime.now().strftime('%a, %d %b %Y %H:%M:%S GMT')}</lastBuildDate>
    {''.join(rss_items)}
</channel>
</rss>"""
    
    return rss_xml

def main():
    print("=" * 50)
    print(SCRIPT_TITLE)
    print("=" * 50)

# ---------------------------------------------------------------------------
    # This safety check can be deleted after you're comfortable with the script
    response = input("This will delete existing index*.html and rss.xml. Continue? (y/N): ")
    if response.lower() != 'y':
        print("Exiting.")
        return
    print()
# ---------------------------------------------------------------------------

    if not os.path.exists(TEMPLATE_FILE):
        print(f"Error: {TEMPLATE_FILE} not found in current directory.")
        print("Please create template.html with <!-- POSTS GO HERE --> placeholder.")
        return
    
    with open(TEMPLATE_FILE, 'r', encoding='utf-8') as f:
        template = f.read()
    
    if "<!-- POSTS GO HERE -->" not in template:
        print("Error: template.html must contain <!-- POSTS GO HERE -->")
        return
    
    posts = get_all_posts()
    if not posts:
        print("No posts found. Nothing to generate.")
        return
    
    print(f"Found {len(posts)} total posts")

    clean_old_index_files()

    total_posts = len(posts)
    total_pages = (total_posts + POSTS_PER_PAGE - 1) // POSTS_PER_PAGE
    print(f"Generating {total_pages} page(s) with {POSTS_PER_PAGE} posts per page")
    
    for page_num in range(1, total_pages + 1):
        start_idx = (page_num - 1) * POSTS_PER_PAGE
        end_idx = min(start_idx + POSTS_PER_PAGE, total_posts)
        posts_slice = posts[start_idx:end_idx]
        
        if page_num == 1:
            page_title = "Thoughts & Notes"
        else:
            page_title = f"Thoughts & Notes - Page {page_num}"
        
        page_template = template.replace("<title>Thoughts & Notes</title>", f"<title>{page_title}</title>")
        
        posts_html = render_posts(posts_slice, start_idx+1, total_posts, page_num, total_pages)
        
        final_html = page_template.replace("<!-- POSTS GO HERE -->", posts_html)
        
        if page_num == 1:
            output_file = "index.html"
        else:
            output_file = f"index-{page_num}.html"
        
        with open(output_file, 'w', encoding='utf-8') as f:
            f.write(final_html)
        
        print(f"Generated: {output_file}")
    
    rss_content = generate_rss(posts, base_url=BLOG_BASE_URL)
    with open("rss.xml", 'w', encoding='utf-8') as f:
        f.write(rss_content)
    print("Generated: rss.xml")
    
    print("\n" + "=" * 50)
    print("Done! Upload to Neocities:")
    print("  - index.html, index-*.html")
    print("  - rss.xml")
    print("=" * 50)

if __name__ == "__main__":
    main()