from transformers import pipeline
import os
from typing import Dict

# Configuración de entorno para Hugging Face Spaces
os.environ["TRANSFORMERS_CACHE"] = "/tmp/transformers"
os.environ["HF_HOME"] = "/tmp/huggingface"

class SentimentAnalysisService:
    def __init__(self):
        try:
            print("[LOG] Cargando pipeline con modelo BETO...")
            self.pipeline = pipeline(
                "sentiment-analysis",
                model="finiteautomata/beto-sentiment-analysis",
                top_k=3  # ⬅️ Mostrar las 3 emociones más probables
            )
            print("[LOG] Pipeline cargado correctamente.")
        except Exception as e:
            print("[ERROR] Falló la carga del modelo:", e)
            raise

    def analyze(self, transcript: str) -> Dict:
        print("[LOG] Análisis de transcripción recibido.")
        try:
            results = self.pipeline(transcript)
            print("[LOG] Resultado del modelo:", results)

            # Emoción dominante = la primera (mayor score)
            dominant = results[0]

            emotion_mapping = {
                "POS": "entusiasta",
                "NEU": "neutro",
                "NEG": "frustrado"
            }

            dominant_emotion = emotion_mapping.get(dominant['label'], "desconocido")
            confidence = round(dominant['score'], 2)

            # Crear diccionario de probabilidades mapeadas
            emotion_probabilities = {
                emotion_mapping.get(r['label'], r['label']): round(r['score'], 2)
                for r in results
            }

        except Exception as e:
            print("[ERROR] Falló la predicción:", e)
            return {
                "dominant_emotion": "error",
                "emotion_probabilities": {},
                "confidence": 0.0
            }

        return {
            "dominant_emotion": dominant_emotion,
            "emotion_probabilities": emotion_probabilities,
            "confidence": confidence
        }