app.py

from fastapi import FastAPI, Query, Response
import undetected_chromedriver as uc
import requests
import time
import json
import os
from typing import Optional

app = FastAPI()

COOKIES_FILE = "cf_cookies.json"

def save_cookies(cookies: dict):
    with open(COOKIES_FILE, "w") as f:
        json.dump(cookies, f)

def load_cookies() -> Optional[dict]:
    if not os.path.exists(COOKIES_FILE):
        return None
    with open(COOKIES_FILE, "r") as f:
        return json.load(f)

def get_cf_cookie(url):
    print("🔍 Iniciando navegador para resolver challenge de Cloudflare...")

    try:
        options = uc.ChromeOptions()
        options.headless = True
        options.add_argument("--no-sandbox")
        options.add_argument("--disable-dev-shm-usage")
        options.add_argument("--disable-blink-features=AutomationControlled")

        driver = uc.Chrome(options=options)

        driver.get(url)

        # Esperar hasta 30 segundos para que Cloudflare entregue cf_clearance
        max_wait = 30
        cf_clearance = None
        domain = None

        for _ in range(max_wait):
            cookies = driver.get_cookies()
            for cookie in cookies:
                if cookie["name"] == "cf_clearance":
                    cf_clearance = cookie["value"]
                    domain = cookie["domain"]
                    break
            if cf_clearance:
                break
            time.sleep(1)

        driver.quit()

        if cf_clearance:
            print("✅ cf_clearance encontrada:", cf_clearance)
            result = {"cf_clearance": cf_clearance, "domain": domain}
            save_cookies(result)
            return result
        else:
            print("⚠️ No se encontró cf_clearance")
            return None

    except Exception as e:
        print("❌ Error con el navegador:", str(e))
        return None

@app.get("/")
def root():
    return {"message": "FastAPI + Cloudflare Bypass funcionando ✅"}

@app.get("/cookie")
def get_cookie(
    url: str = "https://www.visa.com.do/cmsapi/fx/rates?amount=1&fee=3&utcConvertedDate=10%2F19%2F2025&exchangedate=10%2F19%2F2025&fromCurr=DOP&toCurr=USD"
):
    result = get_cf_cookie(url)
    if result:
        return result
    return {"error": "No se pudo obtener cf_clearance"}

@app.get("/scrape")
def scrape(url: str = Query(..., description="URL a scrapear")):
    cookie_result = load_cookies()

    if not cookie_result:
        cookie_result = get_cf_cookie(url)
        if not cookie_result:
            return {
                "error": "No se pudo obtener cf_clearance. Puede que el sitio no esté protegido por Cloudflare o hubo un error con el navegador."
            }

    headers = {
        "User-Agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/120.0.0.0 Safari/537.36"
    }

    cookies = {
        "cf_clearance": cookie_result["cf_clearance"]
    }

    try:
        res = requests.get(url, headers=headers, cookies=cookies, timeout=10)

        if res.status_code in (403, 503):
            os.remove(COOKIES_FILE)
            return {
                "error": f"cf_clearance caducada o inválida (status {res.status_code}). Intenta regenerarla."
            }

        # Detectar tipo de contenido y devolver el body completo
        content_type = res.headers.get("Content-Type", "").lower()

        if "application/json" in content_type:
            return Response(content=res.text, media_type="application/json")
        else:
            return Response(content=res.text, media_type="text/html")

    except Exception as e:
        return {"error": str(e)}