app/ht_rd.py

# app/ht_rd.py
import cloudscraper
import json
import random
from datetime import datetime, timedelta
from typing import Dict, List, Optional, Any
from fastapi import HTTPException

class HentaiRandomApp:
    def __init__(self):
        self.base_url = "https://hentaiz.dog/watch"
        self.api_url = (
            "https://hentaiz.dog/__data.json"
            "?x-sveltekit-trailing-slash=1"
            "&x-sveltekit-invalidated=001"
        )
        self.img_base = "https://storage.haiten.org"
        self.scraper = cloudscraper.create_scraper(
            browser={"browser": "chrome", "platform": "windows"}
        )
        self.cache_data: Optional[List[Dict]] = None
        self.cache_expire: Optional[datetime] = None
        self.cache_ttl = 600

    async def start(self):
        return

    async def stop(self):
        try:
            self.scraper.close()
        except:
            pass

    def _get_val(self, idx, pool):
        """Hàm lấy giá trị an toàn từ pool theo index"""
        if isinstance(idx, int) and 0 <= idx < len(pool):
            return pool[idx]
        return idx

    def _resolve_img(self, img_idx, pool):
        """Xử lý lấy URL ảnh từ cấu trúc lồng nhau"""
        img_obj = self._get_val(img_idx, pool)
        if isinstance(img_obj, dict) and "filePath" in img_obj:
            path = self._get_val(img_obj["filePath"], pool)
            if isinstance(path, str):
                return f"{self.img_base}{path}"
        return ""

    async def fetch_data(self) -> List[Dict]:
        if (self.cache_data and self.cache_expire and datetime.utcnow() < self.cache_expire):
            return self.cache_data

        response = self.scraper.get(self.api_url, timeout=20)
        if response.status_code != 200:
            return self.cache_data or []

        # Xử lý SvelteKit multi-line JSON
        lines = response.text.strip().split("\n")
        full_pool = []
        
        for line in lines:
            try:
                chunk = json.loads(line)
                # SvelteKit thường để data trong nodes -> data
                if "nodes" in chunk:
                    for node in chunk["nodes"]:
                        if node and "data" in node:
                            full_pool.extend(node["data"])
                elif "data" in chunk:
                    full_pool.extend(chunk["data"])
            except:
                continue

        if not full_pool:
            return []

        items = []
        # Duyệt qua pool để tìm các object là Anime (có title và slug)
        for entry in full_pool:
            if isinstance(entry, dict) and "title" in entry and "slug" in entry:
                try:
                    # Lấy Title & Slug
                    title = self._get_val(entry["title"], full_pool)
                    slug = self._get_val(entry["slug"], full_pool)
                    
                    if not isinstance(slug, str): continue

                    # Fix Encoding nếu bị lỗi latin1
                    if isinstance(title, str):
                        try: title = title.encode("latin1").decode("utf-8")
                        except: pass

                    # Episode
                    ep_val = self._get_val(entry.get("episodeNumber", 1), full_pool)
                    
                    # Ảnh
                    poster = self._resolve_img(entry.get("posterImage"), full_pool)
                    backdrop = self._resolve_img(entry.get("backdropImage"), full_pool)

                    # TRUY VẾT STUDIO (Logic từ ht_noti.py)
                    studios = []
                    st_ref = entry.get("studios")
                    st_list = self._get_val(st_ref, full_pool)
                    
                    if isinstance(st_list, list):
                        for st_idx in st_list:
                            st_wrapper = self._get_val(st_idx, full_pool)
                            if isinstance(st_wrapper, dict) and "studio" in st_wrapper:
                                st_info = self._get_val(st_wrapper["studio"], full_pool)
                                if isinstance(st_info, dict) and "name" in st_info:
                                    st_name = self._get_val(st_info["name"], full_pool)
                                    if isinstance(st_name, str):
                                        try: st_name = st_name.encode("latin1").decode("utf-8")
                                        except: pass
                                        studios.append(st_name)

                    items.append({
                        "title": title,
                        "slug": slug,
                        "episode": int(ep_val) if isinstance(ep_val, (int, float)) else 1,
                        "poster_url": poster,
                        "backdrop_url": backdrop,
                        "studios": studios,
                        "link": f"{self.base_url}/{slug}"
                    })
                except:
                    continue

        # Loại bỏ các item lỗi hoặc trùng lặp
        unique_items = list({it["slug"]: it for it in items}.values())
        self.cache_data = unique_items
        self.cache_expire = datetime.utcnow() + timedelta(seconds=self.cache_ttl)
        return unique_items

    async def get_random(self, apikey: Optional[str], system_key: Optional[str], limit: str = "1"):
        if system_key and apikey != system_key:
            raise HTTPException(status_code=403, detail="Invalid API key")

        try:
            if "-" in limit:
                low, high = map(int, limit.split("-"))
                count = random.randint(max(1, low), min(50, high))
            else:
                count = max(1, min(50, int(limit)))
        except:
            count = 1

        all_items = await self.fetch_data()
        if not all_items:
            return {"ok": False, "msg": "No data found"}

        selected = random.sample(all_items, min(count, len(all_items)))

        return {
            "ok": True,
            "total_pool": len(all_items),
            "item": selected if (count > 1 or "-" in limit) else selected[0],
            "source": "hentaiz.dog",
            "cached": True,
            "made_by": "devily"
        }