routers/getnews.py

from fastapi import APIRouter, HTTPException
import httpx
from typing import List, Dict
from bs4 import BeautifulSoup
import re

router = APIRouter()

GRAPHQL_URL = "https://api.graphql.imdb.com"
HEADERS = {"Content-Type": "application/json"}

QUERY = """
query GetNews($first: Int!) {
  movieNews: news(first: $first, category: MOVIE) {
    edges {
      node {
        id
        articleTitle { plainText }
        externalUrl
        date
        text { plaidHtml }
        image { url }
      }
    }
  }
  tvNews: news(first: $first, category: TV) {
    edges {
      node {
        id
        articleTitle { plainText }
        externalUrl
        date
        text { plaidHtml }
        image { url }
      }
    }
  }
}
"""

def clean_html(raw_html: str) -> str:
    # Remove tags HTML
    text = BeautifulSoup(raw_html or "", "html.parser").get_text(separator=" ", strip=True)
    
    # Remove múltiplos espaços, tabs, quebras de linha, etc.
    text = re.sub(r"\s+", " ", text)

    # Remove espaços antes de pontuações
    text = re.sub(r"\s+([.,;:!?])", r"\1", text)

    # Remove espaços após parênteses de abertura e antes de fechamento
    text = re.sub(r"\(\s+", "(", text)
    text = re.sub(r"\s+\)", ")", text)

    # Remove espaços desnecessários entre colchetes ou chaves se quiser estender
    text = re.sub(r"\[\s+", "[", text)
    text = re.sub(r"\s+\]", "]", text)
    text = re.sub(r"\{\s+", "{", text)
    text = re.sub(r"\s+\}", "}", text)

    return text.strip()

@router.get("/news")
async def get_news(first: int = 15) -> List[Dict]:
    payload = {
        "query": QUERY,
        "variables": {"first": first}
    }

    async with httpx.AsyncClient(timeout=10.0) as client:
        response = await client.post(GRAPHQL_URL, headers=HEADERS, json=payload)

    if response.status_code != 200:
        raise HTTPException(status_code=502, detail="Erro ao acessar a API do IMDb")

    data = response.json().get("data")
    if not data:
        raise HTTPException(status_code=500, detail="Resposta inválida da API")

    combined = []

    for category_key in ["movieNews", "tvNews"]:
        for edge in data.get(category_key, {}).get("edges", []):
            node = edge.get("node", {})
            combined.append({
                "id": node.get("id"),
                "title": node.get("articleTitle", {}).get("plainText"),
                "url": node.get("externalUrl"),
                "date": node.get("date"),
                "text": clean_html(node.get("text", {}).get("plaidHtml")),
                "image": node.get("image", {}).get("url"),
                "category": category_key.replace("News", "").upper()
            })

    combined.sort(key=lambda x: x.get("date"), reverse=True)
    return combined