scraper.py

import requests
from bs4 import BeautifulSoup
import re
from playwright.sync_api import sync_playwright
import time

DOMENA = "https://mrkaj.me"

def search_movies(query):
    api_url = f"{DOMENA}/se/j/json?q={query}"
    headers = {"User-Agent": "Mozilla/5.0"}
    try:
        response = requests.get(api_url, headers=headers)
        if response.status_code == 200:
            return response.json()
    except Exception as e:
        print(f"Search error: {e}")
    return []

def get_details(slug, media_type):
    url = f"{DOMENA}/p/{slug}" if media_type in ["movie", "film"] else f"{DOMENA}/tv/{slug}/S01E01"
    headers = {"User-Agent": "Mozilla/5.0"}
    results = {"languages": [], "seasons": []}
    
    try:
        response = requests.get(url, headers=headers)
        if response.status_code != 200:
            return results
        
        soup = BeautifulSoup(response.text, "html.parser")
        
        # Languages
        keyword_div = soup.find("div", class_="movie-keyword")
        if keyword_div:
            for podtrieda in keyword_div.find_all(recursive=False):
                text = podtrieda.get_text(strip=True)
                if text:
                    results["languages"].append(text)
        
        # Seasons
        if media_type not in ["movie", "film"]:
            response_text = response.text
            x = 0
            while True:
                x = response_text.find("Séria", x+1)
                if x != -1:
                    seria_match = re.search(r'Séria\s+(\d+)', response_text[x:x+50])
                    if seria_match:
                        seria = seria_match.group(1)
                        y = response_text.find("</span>", x)
                        if y != -1:
                            pocet_epizod = ''.join(filter(str.isnumeric, response_text[y-10:y]))
                            results["seasons"].append({
                                "season": int(seria),
                                "episodes": int(pocet_epizod) if pocet_epizod else 0
                            })
                else:
                    break
    except Exception as e:
        print(f"Details error: {e}")
        
    return results

def get_stream_url(slug, media_type, season=None, episode=None, lang=None, source_idx=0):
    if media_type in ["movie", "film"]:
        url = f"{DOMENA}/p/{slug}"
    else:
        url = f"{DOMENA}/tv/{slug}/S{int(season):02d}E{int(episode):02d}"
    
    # URL parameters for source and language
    # Using the format expected by the site (?source=X&lang=Y)
    params = []
    if source_idx is not None:
        params.append(f"source={source_idx}")
    if lang:
        params.append(f"lng={requests.utils.quote(lang)}")
    
    if params:
        url += "?" + "&".join(params)
    
    print(f"Scanning URL: {url}")
    
    with sync_playwright() as p:
        # We need to install browsers in the Dockerfile
        browser = p.chromium.launch(headless=True)
        context = browser.new_context(user_agent="Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/110.0.0.0 Safari/537.36")
        page = context.new_page()
        
        found = {"stream": None, "vtt": None}
        
        def handle_request(request):
            if ".m3u8" in request.url and not found["stream"]:
                found["stream"] = request.url
            if request.url.endswith(".vtt") and not found["vtt"]:
                found["vtt"] = request.url
                
        page.on("request", handle_request)
        
        try:
            page.goto(url, wait_until="networkidle", timeout=30000)
            
            # Additional wait or click if needed
            for _ in range(10):
                if found["stream"]:
                    time.sleep(1) # Wait for VTT
                    break
                time.sleep(1)
            
            # If not found, try to click the iframe if it exists (like in app.py)
            if not found["stream"]:
                try:
                    iframe = page.locator("iframe#frm")
                    if iframe.count() > 0:
                        iframe.click(force=True)
                        for _ in range(10):
                            if found["stream"]:
                                time.sleep(1)
                                break
                            time.sleep(1)
                except:
                    pass
        except Exception as e:
            print(f"Playwright error: {e}")
            
        browser.close()
        return found