Update main.py

This commit is contained in:
Aira Catapang
2026-03-17 06:17:33 +00:00
committed by system
parent ed26a0172b
commit 99a9cdb91a

179
main.py
View File

@@ -5,10 +5,12 @@ import os
import httpx
from typing import Optional
from contextlib import asynccontextmanager
from fastapi import FastAPI
from fastapi import FastAPI, Query, Request
from fastapi.responses import StreamingResponse, Response, JSONResponse
from playwright.async_api import async_playwright, BrowserContext
BASE_URL = "https://animepahe.si"
ANIWATCHTV_BASE = "https://aniwatchtv.to"
ANILIST_API = "https://graphql.anilist.co"
JIKAN_API = "https://api.jikan.moe/v4"
IS_HEADLESS = os.environ.get("HEADLESS", "true").lower() == "true"
@@ -53,6 +55,10 @@ class AnimePahe:
async def _intercept_assets(self, route):
url = route.request.url.lower()
# Allow all requests from aniwatchtv so season posters can load
if "aniwatchtv.to" in url:
await route.continue_()
return
if any(ad in url for ad in self.ad_domains) or url.endswith(
(".png", ".jpg", ".jpeg", ".webp", ".woff")
):
@@ -619,6 +625,108 @@ class AnimePahe:
if page:
await page.close()
# ---------------- SEASONS ----------------
async def get_seasons(self, anime_id: str) -> dict:
"""
Scrape the 'More Seasons' section from aniwatchtv.to using the
existing Playwright browser context.
anime_id is the full slug, e.g. jujutsu-kaisen-the-culling-game-part-1-20401
"""
url = f"{ANIWATCHTV_BASE}/{anime_id}"
page = await self.context.new_page()
try:
await page.goto(url, wait_until="domcontentloaded", timeout=30000)
# Short wait for lazy-loaded images and JS rendering
await asyncio.sleep(1)
# Wait for season block — try common selectors gracefully
for selector in [".os-list", ".seasons-block", "[class*='season']", "main"]:
try:
await page.wait_for_selector(selector, timeout=5000)
break
except:
continue
seasons = await page.evaluate(f"""() => {{
const BASE = "{ANIWATCHTV_BASE}";
const currentId = "{anime_id}";
const results = [];
const seen = new Set();
// Strategy 1: dedicated season list block (.os-list or similar)
const block = (
document.querySelector('.os-list') ||
document.querySelector('.seasons-block') ||
document.querySelector('[class*="os-list"]') ||
document.querySelector('[class*="season-list"]')
);
// Strategy 2: find a heading that says "More Seasons" and walk up
const fallbackContainer = (() => {{
for (const el of document.querySelectorAll('*')) {{
if (/more\\s+seasons?/i.test(el.innerText?.trim() || '')) {{
let p = el.parentElement;
for (let i = 0; i < 5; i++) {{
if (!p) break;
if (p.querySelectorAll('a[href]').length > 0) return p;
p = p.parentElement;
}}
}}
}}
return null;
}})();
const container = block || fallbackContainer;
if (!container) return results;
for (const a of container.querySelectorAll('a[href]')) {{
const href = a.getAttribute('href') || '';
const fullUrl = href.startsWith('http') ? href
: href.startsWith('/') ? BASE + href
: null;
if (!fullUrl) continue;
const slug = fullUrl.replace(/\\/$/, '').split('/').pop();
// Include ALL slugs — current page included — dedupe only
if (!slug || seen.has(slug)) continue;
seen.add(slug);
const numericMatch = slug.match(/-(\\d+)$/);
const numericId = numericMatch ? numericMatch[1] : null;
const titleEl = a.querySelector('span, [class*="title"], [class*="name"]');
const title = (titleEl?.innerText?.trim() || a.innerText?.trim() || slug);
// Poster is in a sibling/child div.season-poster as a CSS background-image
const posterEl = a.querySelector('.season-poster') || a.closest('li, div')?.querySelector('.season-poster');
let poster = null;
if (posterEl) {{
const bg = posterEl.style.backgroundImage || window.getComputedStyle(posterEl).backgroundImage;
const bg2 = bg.split('url(').pop().split(')')[0].replace(/['"/]/g, '').trim();
if (bg2 && bg2.startsWith('http')) poster = bg2;
}}
results.push({{ title, id: slug, numericId, url: fullUrl, poster }});
}}
return results;
}}""")
return {
"id": anime_id,
"total": len(seasons),
"seasons": seasons,
}
except Exception as e:
print(f"[get_seasons] ERROR: {e}")
return {"id": anime_id, "total": 0, "seasons": [], "error": str(e)}
finally:
await page.close()
pahe = AnimePahe()
@@ -633,6 +741,23 @@ async def lifespan(app: FastAPI):
app = FastAPI(lifespan=lifespan)
@app.get("/", response_class=JSONResponse)
async def root():
return {
"status": "ok",
"routes": [
"/search?q=:title",
"/latest?p=:page",
"/info/:session",
"/ids/:session",
"/episodes/:session?p=:page&resolve=false|true",
"/resolve/:animeSession/:episodeSession",
"/seasons/:animeId - e.g. /seasons/jujutsu-kaisen-the-culling-game-part-1-20401",
"/poster?url=:cdnImageUrl - proxy hotlink-protected poster images",
],
}
@app.get("/search")
async def api_search(q: str):
return await pahe.search(q)
@@ -663,6 +788,58 @@ async def api_resolve(anime: str, episode: str):
return await pahe.resolve(anime, episode)
@app.get("/seasons/{anime_id:path}")
async def api_seasons(anime_id: str, request: Request):
"""
Scrape the More Seasons section from aniwatchtv.to.
Example:
GET /seasons/jujutsu-kaisen-the-culling-game-part-1-20401
Returns:
id - the slug passed in
total - number of other seasons found
seasons[] - list of { title, id, numericId, url, poster }
"""
data = await pahe.get_seasons(anime_id)
base_url = str(request.base_url).rstrip("/")
for season in data.get("seasons", []):
if season.get("poster"):
season["posterProxied"] = f"{base_url}/poster?url={season['poster']}"
else:
season["posterProxied"] = None
return data
@app.get("/poster")
async def api_poster(url: str = Query(..., description="CDN image URL to proxy")):
"""
Proxy a hotlink-protected poster image with the correct Referer header.
Use this to display season/anime posters in the browser.
Example:
GET /poster?url=https://cdn.noitatnemucod.net/thumbnail/100x200/100/abc123.jpg
"""
try:
async with httpx.AsyncClient(timeout=10, follow_redirects=True) as client:
resp = await client.get(
url,
headers={
"Referer": "https://aniwatchtv.to/",
"User-Agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/122 Safari/537.36",
},
)
resp.raise_for_status()
content_type = resp.headers.get("content-type", "image/jpeg")
return Response(content=resp.content, media_type=content_type)
except Exception as e:
return Response(
content=f"Failed to fetch image: {e}",
status_code=502,
media_type="text/plain",
)
if __name__ == "__main__":
import uvicorn