scraper2.py

import asyncio
import urllib.parse
import json
from playwright.async_api import async_playwright
from bs4 import BeautifulSoup
import re
import os
import shutil
from pyvirtualdisplay import Display

DOMENA = "https://mrkaj.me"
EXT_PATH = "/app/dew_vpn"
EXT_ID = "pgpnoemmehpnealclfbkejocclofdbon"

class AsyncBrowserManager:
    _instance = None

    def __init__(self):
        self.playwright = None
        self.context = None
        self.display = None
        self._setup_done = False
        self._lock = asyncio.Lock()

    @classmethod
    def get_instance(cls):
        if cls._instance is None:
            cls._instance = cls()
        return cls._instance

    async def get_screenshot(self):
        if not self.context: return None
        pages = self.context.pages
        if pages:
            for p in pages:
                if "extension" in p.url: return await p.screenshot(type="png")
            return await pages[-1].screenshot(type="png")
        return None

    async def start(self):
        async with self._lock:
            if self._setup_done: return
            
            user_data_dir = "/tmp/browser_profile"
            shutil.rmtree(user_data_dir, ignore_errors=True)

            print("[BROWSER] Spúšťam virtuálnu obrazovku...")
            try:
                self.display = Display(visible=0, size=(1280, 1024))
                self.display.start()
            except Exception as e:
                print(f"[Xvfb ERROR] {e}")

            print("[BROWSER] Spúšťam Chrome...")
            self.playwright = await async_playwright().start()
            print("[BROWSER] Spúšťam Chrome krok 1...")
            launch_args = [
                "--no-sandbox",
                "--disable-setuid-sandbox",
                "--disable-dev-shm-usage",
                "--disable-blink-features=AutomationControlled",
                f"--disable-extensions-except={EXT_PATH}",
                f"--load-extension={EXT_PATH}"
            ]

            self.context = await self.playwright.chromium.launch_persistent_context(
                user_data_dir,
                headless=False,
                args=launch_args,
                ignore_default_args=["--enable-automation"],
                bypass_csp=True,
                user_agent="Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/122.0.0.0 Safari/537.36"
            )
            print("[BROWSER] Chrome spustený")
            await asyncio.sleep(12)
            await self.activate_dewvpn()
            self._setup_done = True

    async def activate_dewvpn(self):
        print("[VPN] Aktivujem DewVPN...")
        page = await self.context.new_page()
        try:
            await page.goto(f"chrome-extension://{EXT_ID}/popup.html", timeout=30000)
            await asyncio.sleep(5)
            
            # DewVPN má zvyčajne zoznam lokalít priamo na hlavnej obrazovke alebo pod tlačidlom
            # Skúsime nájsť Slovensko
            try:
                # Klik na výber krajiny
                await page.click(".current-location, .location-selector")
                await asyncio.sleep(2)
            except: pass
            await page.fill("input[placeholder*='Search']", "Slovakia")
            await asyncio.sleep(2)
            
            await page.click("text=Slovakia")
            print("[VPN] DewVPN: Kliknuté na Slovensko!")
            await asyncio.sleep(8)
            
        except Exception as e:
            print(f"[VPN ERROR] {e}")
            # Ak to zlyhá, urobíme screenshot, aby sme videli, ako DewVPN vyzerá
        finally:
            await page.close()

    async def get_page(self):
        if not self._setup_done: await self.start()
        return await self.context.new_page()

manager = AsyncBrowserManager.get_instance()

# --- HLAVNÉ FUNKCIE SCRAPERA ---

async def search_movies(query):
    print(f"[SEARCH] Hľadám: {query}")
    page = await manager.get_page()
    try:
        url = f"{DOMENA}/se/j/json?q={urllib.parse.quote(query)}"
        await page.goto(url, wait_until="networkidle", timeout=45000)
        content = await page.inner_text("body")
        return json.loads(content)
    except Exception as e:
        print(f"[SEARCH ERROR] {e}")
        return []
    finally: await page.close()

async def get_details(slug, media_type):
    print(f"[DETAILS] Info: {slug}")
    page = await manager.get_page()
    url = f"{DOMENA}/p/{slug}" if media_type in ["movie", "film"] else f"{DOMENA}/tv/{slug}/S01E01"
    res = {"languages": [], "seasons": []}
    try:
        await page.goto(url, wait_until="domcontentloaded", timeout=45000)
        soup = BeautifulSoup(await page.content(), "html.parser")
        kw = soup.find("div", class_="movie-keyword")
        if kw: res["languages"] = [p.get_text(strip=True) for p in kw.find_all(recursive=False)]
        
        matches = re.findall(r'Séria\s+(\d+)', soup.get_text())
        for seria in sorted(set(matches), key=int):
            res["seasons"].append({"season": int(seria), "episodes": 0})
    except Exception as e: print(f"[DETAILS ERROR] {e}")
    finally: await page.close()
    return res

async def get_stream_url(slug, media_type, season=None, episode=None, lang=None, source_idx=0):
    print(f"[STREAM] Skenujem: {slug}")
    page = await manager.get_page()
    url = f"{DOMENA}/p/{slug}" if media_type in ["movie", "film"] else f"{DOMENA}/tv/{slug}/S{int(season):02d}E{int(episode):02d}"
    
    params = [f"source={source_idx}"]
    if lang: params.append(f"lng={urllib.parse.quote(lang)}")
    url += "?" + "&".join(params)
    
    found = {"stream": None, "vtt": None}
    
    async def zachyt_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", zachyt_request)
    
    try:
        await page.goto(url, wait_until="domcontentloaded", timeout=60000)
        for _ in range(20):
            if found["stream"]: break
            await asyncio.sleep(1)
            
        if not found["stream"]:
            iframe = page.locator("iframe#frm, video").first
            if await iframe.count() > 0:
                await iframe.click(force=True)
                for _ in range(10):
                    if found["stream"]: break
                    await asyncio.sleep(1)
    except Exception as e: print(f"[STREAM ERROR] {e}")
    finally: await page.close()
    return found