#!/usr/bin/env python3
"""Build combined Joomla article from 4 Prime Day deal files."""

import re
import json

def extract_deal_blocks(html_content):
    """Extract all deal div blocks from HTML content."""
    # Find all deal div blocks (flex containers with images)
    blocks = re.findall(
        r'<div style="display:flex;gap:1rem;padding:1rem 0;border-bottom:1px solid #eee;">(.*?)</div>\s*</div>',
        html_content, re.DOTALL
    )
    # Also try the card class variant from devices file
    if not blocks:
        blocks = re.findall(
            r'<div class="card">(.*?)</div>\s*</div>',
            html_content, re.DOTALL
        )
    return blocks

def extract_deal_blocks_full(html_content):
    """Extract full deal div blocks including the outer div."""
    blocks = re.findall(
        r'(<div style="display:flex;gap:1rem;padding:1rem 0;border-bottom:1px solid #eee;">.*?</div>\s*</div>)',
        html_content, re.DOTALL
    )
    if not blocks:
        blocks = re.findall(
            r'(<div class="card">.*?</div>\s*</div>)',
            html_content, re.DOTALL
        )
    return blocks

def extract_deal_blocks_devices(html_content):
    """Extract deal blocks from the devices file (uses class='card')."""
    blocks = re.findall(
        r'<div class="card">(.*?)</div>(?=\s*(?:<h2>|<p style|<div class="card">|</body>))',
        html_content, re.DOTALL
    )
    return blocks

def convert_device_block_to_standard(block_html):
    """Convert a device card block to standard deal format."""
    # Extract image
    img_match = re.search(r'<img src="([^"]+)"', block_html)
    img_src = img_match.group(1) if img_match else ''
    
    # Extract title
    title_match = re.search(r'<strong>(.*?)</strong>', block_html)
    title = title_match.group(1) if title_match else ''
    
    # Extract price
    price_match = re.search(r'<span class="price">(.*?)</span>', block_html)
    price = price_match.group(1) if price_match else ''
    
    # Extract UVP
    uvp_match = re.search(r'<span class="uvp">(.*?)</span>', block_html)
    uvp = uvp_match.group(1) if uvp_match else ''
    
    # Extract badge (discount %)
    badge_match = re.search(r'<span class="badge">(.*?)</span>', block_html)
    badge = badge_match.group(1) if badge_match else ''
    
    # Extract link
    link_match = re.search(r'<a href="([^"]+)"[^>]*>Zu Amazon →</a>', block_html)
    link = link_match.group(1) if link_match else ''
    
    # Calculate savings
    savings = ''
    if price and uvp:
        try:
            price_val = float(price.replace('€', '').replace(',', '.').strip())
            uvp_val = float(uvp.replace('UVP ', '').replace('€', '').replace(',', '.').strip())
            saved = round(uvp_val - price_val, 2)
            savings = f'({saved:.2f} € gespart)'
        except:
            pass
    
    # Build standard block
    block = f'''<div style="display:flex;gap:1rem;padding:1rem 0;border-bottom:1px solid #eee;">
<img src="{img_src}" alt="" style="width:100px;height:100px;object-fit:contain;">
<div style="flex:1;"><strong>{title}</strong><br>
<span style="color:#b12704;font-size:1.2rem;font-weight:bold;">{price}</span>
<span style="text-decoration:line-through;color:#999;margin-left:0.5rem;">{uvp}</span>
<span style="background:#b12704;color:#fff;padding:2px 6px;border-radius:3px;margin-left:0.5rem;font-size:0.85rem;">{badge}</span>'''
    
    if savings:
        block += f'\n<span style="color:#b12704;margin-left:0.3rem;">{savings}</span>'
    
    block += f'<br>\n<a href="{link}" target="_blank" style="color:#007185;">Zu Amazon →</a></div></div>'
    
    return block

def main():
    # Read all 4 files
    files = {
        'mass_market': '/home/hermes/workspace/amazon_deals_2026-06-23.html',
        'kennerspiele': '/home/hermes/workspace/kennerspiele_deals_2026-06-23.html',
        'top_deals': '/home/hermes/workspace/top_deals_2026-06-23.html',
        'devices': '/home/hermes/workspace/amazon_devices_prime_2026-06-23.html',
    }
    
    contents = {}
    for key, path in files.items():
        with open(path, 'r') as f:
            contents[key] = f.read()
    
    # Extract deal blocks from each file
    mass_market_blocks = extract_deal_blocks_full(contents['mass_market'])
    kennerspiele_blocks = extract_deal_blocks_full(contents['kennerspiele'])
    
    # For top_deals, extract blocks by category
    top_html = contents['top_deals']
    
    # Extract sections from top_deals
    # Heißluftfritteusen
    hf_section = re.search(r'<h2[^>]*>Heißluftfritteusen.*?</h2>(.*?)(?=<h2)', top_html, re.DOTALL)
    hf_blocks = extract_deal_blocks_full(hf_section.group(1)) if hf_section else []
    
    # Staubsauger
    st_section = re.search(r'<h2[^>]*>Staubsauger.*?</h2>(.*?)(?=<h2)', top_html, re.DOTALL)
    st_blocks = extract_deal_blocks_full(st_section.group(1)) if st_section else []
    
    # Laptops
    lp_section = re.search(r'<h2[^>]*>Laptops.*?</h2>(.*?)(?=<h2)', top_html, re.DOTALL)
    lp_blocks = extract_deal_blocks_full(lp_section.group(1)) if lp_section else []
    
    # Tablets
    tb_section = re.search(r'<h2[^>]*>Tablets.*?</h2>(.*?)(?=<h2)', top_html, re.DOTALL)
    tb_blocks = extract_deal_blocks_full(tb_section.group(1)) if tb_section else []
    
    # Smartphones
    sp_section = re.search(r'<h2[^>]*>Smartphones.*?</h2>(.*?)(?=<h2)', top_html, re.DOTALL)
    sp_blocks = extract_deal_blocks_full(sp_section.group(1)) if sp_section else []
    
    # Apple
    ap_section = re.search(r'<h2[^>]*>Apple.*?</h2>(.*?)(?=<p class="meta")', top_html, re.DOTALL)
    ap_blocks = extract_deal_blocks_full(ap_section.group(1)) if ap_section else []
    
    # Devices - extract all card blocks
    device_blocks_raw = extract_deal_blocks_devices(contents['devices'])
    device_blocks = [convert_device_block_to_standard(b) for b in device_blocks_raw]
    
    # Count totals
    total = (len(kennerspiele_blocks) + len(mass_market_blocks) + 
             len(hf_blocks) + len(st_blocks) + len(lp_blocks) + 
             len(tb_blocks) + len(sp_blocks) + len(ap_blocks) + 
             len(device_blocks))
    
    print(f"Kennerspiele: {len(kennerspiele_blocks)}")
    print(f"Mass-Market Brettspiele: {len(mass_market_blocks)}")
    print(f"Heißluftfritteusen: {len(hf_blocks)}")
    print(f"Staubsauger: {len(st_blocks)}")
    print(f"Laptops: {len(lp_blocks)}")
    print(f"Tablets: {len(tb_blocks)}")
    print(f"Smartphones: {len(sp_blocks)}")
    print(f"Apple: {len(ap_blocks)}")
    print(f"Devices: {len(device_blocks)}")
    print(f"TOTAL: {total}")
    
    # Build the article
    article = []
    
    # Title and intro
    article.append('<h1>Prime Days Sommer 2026: Die große Brettspiel- & Technik-Deal-Übersicht</h1>')
    article.append('')
    article.append('<p>Die Amazon Prime Days Sommer 2026 sind da! Wir haben für euch die besten Deals aus den Bereichen Brettspiele, Kennerspiele, Haushaltselektronik und Technik zusammengestellt. Insgesamt erwarten euch <strong>über 100 Angebote</strong> — von familienfreundlichen Brettspiel-Klassikern über anspruchsvolle Kennerspiele mit BGG-Ranking bis hin zu Heißluftfritteusen, Staubsaugern, Laptops, Tablets, Smartphones und Apple-Produkten. Alle Preise sind Momentaufnahmen und können sich schnell ändern — also schnell zugreifen!</p>')
    article.append('')
    article.append('<hr id="system-readmore" />')
    article.append('')
    
    # Section 1: Kennerspiele
    article.append('<h2>🔥 Kennerspiele mit BGG-Ranking</h2>')
    article.append('<p>Für alle Vielspieler und Strategen: Hier kommen die anspruchsvolleren Brettspiele mit BGG-Weight-Rating. Viele davon sind exklusive Prime-Day-Angebote (⚡) — also nur für Prime-Mitglieder!</p>')
    article.append('')
    for block in kennerspiele_blocks:
        article.append(block)
        article.append('')
    
    # Section 2: Mass-Market Brettspiele
    article.append('<h2>🎲 Alle Brettspiel-Angebote</h2>')
    article.append('<p>Familienspiele, Klassiker und Partyspiele — hier ist für jeden Spieleabend etwas dabei. Von Monopoly über Catan bis hin zu UNO und Trivial Pursuit.</p>')
    article.append('')
    for block in mass_market_blocks:
        article.append(block)
        article.append('')
    
    # Section 3: Heißluftfritteusen
    article.append('<h2>🍟 Heißluftfritteusen</h2>')
    article.append('<p>Knusprig, schnell und energiesparend: Die besten Airfryer-Deals der Prime Days. Von Tefal über Ninja bis Philips — hier sind die Top-Angebote.</p>')
    article.append('')
    for block in hf_blocks:
        article.append(block)
        article.append('')
    
    # Section 4: Staubsauger
    article.append('<h2>🧹 Staubsauger & Wischroboter</h2>')
    article.append('<p>Saugroboter, Akku-Staubsauger und klassische Bodenstaubsauger — die besten Deals für ein sauberes Zuhause.</p>')
    article.append('')
    for block in st_blocks:
        article.append(block)
        article.append('')
    
    # Section 5: Laptops
    article.append('<h2>💻 Laptops</h2>')
    article.append('<p>Notebooks und Laptops für Arbeit, Studium und Freizeit — mit kräftigen Rabatten zum Prime Day.</p>')
    article.append('')
    for block in lp_blocks:
        article.append(block)
        article.append('')
    
    # Section 6: Tablets
    article.append('<h2>📱 Tablets</h2>')
    article.append('<p>Android-Tablets und Samsung Galaxy Tabs — perfekt für Unterhaltung unterwegs.</p>')
    article.append('')
    for block in tb_blocks:
        article.append(block)
        article.append('')
    
    # Section 7: Smartphones
    article.append('<h2>📱 Smartphones</h2>')
    article.append('<p>Samsung, Xiaomi und Motorola — die besten Smartphone-Deals zum Prime Day.</p>')
    article.append('')
    for block in sp_blocks:
        article.append(block)
        article.append('')
    
    # Section 8: Apple
    article.append('<h2>🍎 Apple-Produkte</h2>')
    article.append('<p>iPhones und iPads zu reduzierten Preisen — die besten Apple-Deals der Prime Days.</p>')
    article.append('')
    for block in ap_blocks:
        article.append(block)
        article.append('')
    
    # Section 9: Amazon Devices & Gaming
    article.append('<h2>📦 Gaming & Zubehör</h2>')
    article.append('<p>Gaming-Headsets, Controller, Nintendo Switch 2 und weiteres Zubehör — die besten Deals für Gamer.</p>')
    article.append('')
    for block in device_blocks:
        article.append(block)
        article.append('')
    
    # Footer
    article.append('<p style="margin-top:2rem;color:#999;font-size:0.85rem;">Quellen: Amazon.de über CreatorAPI. Preise Stand 23.06.2026 07:00 Uhr. Preise und Verfügbarkeit können sich schnell ändern. Über die Partner-Links (Tag 60pro05-21) kauft ihr zum gleichen Preis — Amazon zahlt eine kleine Provision.</p>')
    
    full_article = '\n'.join(article)
    
    # Split at readmore
    parts = full_article.split('<hr id="system-readmore" />')
    introtext = parts[0].strip() + '\n<hr id="system-readmore" />'
    fulltext = parts[1].strip() if len(parts) > 1 else ''
    
    # Save the article
    with open('/home/hermes/workspace/combined_article_introtext.html', 'w') as f:
        f.write(introtext)
    with open('/home/hermes/workspace/combined_article_fulltext.html', 'w') as f:
        f.write(fulltext)
    with open('/home/hermes/workspace/combined_article_full.html', 'w') as f:
        f.write(full_article)
    
    print(f"\nIntrotext length: {len(introtext)} chars")
    print(f"Fulltext length: {len(fulltext)} chars")
    print(f"Total article length: {len(full_article)} chars")
    
    # Save JSON payload for Joomla API
    payload = {
        "title": "Prime Days Sommer 2026: Die große Brettspiel- & Technik-Deal-Übersicht",
        "introtext": introtext,
        "fulltext": fulltext,
        "catid": 61,
        "state": 0,
        "access": 6,
        "created_by": 560,
        "language": "de",
        "featured": 0
    }
    
    with open('/home/hermes/workspace/joomla_payload.json', 'w') as f:
        json.dump(payload, f, ensure_ascii=False, indent=2)
    
    print("\nPayload saved to joomla_payload.json")

if __name__ == '__main__':
    main()