# modules/discourse/discourse/discourse_live_interface.py

import streamlit as st
from streamlit_float import *
from streamlit_antd_components import *
import pandas as pd
import logging
import io
import matplotlib.pyplot as plt

# Configuración del logger
logger = logging.getLogger(__name__)

# Importaciones locales
from .discourse_process import perform_discourse_analysis
from .discourse_interface import display_discourse_results  # Añadida esta importación
from ..utils.widget_utils import generate_unique_key
from ..database.discourse_mongo_db import store_student_discourse_result
from ..database.chat_mongo_db import store_chat_history, get_chat_history


#####################################################################################################
def fig_to_bytes(fig):
    """Convierte una figura de matplotlib a bytes."""
    try:
        buf = io.BytesIO()
        fig.savefig(buf, format='png', dpi=300, bbox_inches='tight')
        buf.seek(0)
        return buf.getvalue()
    except Exception as e:
        logger.error(f"Error en fig_to_bytes: {str(e)}")
        return None

#################################################################################################

def display_discourse_live_interface(lang_code, nlp_models, discourse_t):
    """
    Interfaz para el análisis del discurso en vivo
    """
    try:
        if 'discourse_live_state' not in st.session_state:
            st.session_state.discourse_live_state = {
                'analysis_count': 0,
                'current_text1': '',
                'current_text2': '',
                'last_result': None,
                'text_changed': False
            }

        # Crear columnas con proporción 1:3
        input_col, result_col = st.columns([1, 3])

        # Columna izquierda: Entrada de textos
        with input_col:
            st.subheader(discourse_t.get('enter_text', 'Ingrese sus textos'))
            
            # Primer área de texto
            st.markdown("**Texto 1 (Patrón)**")
            text_input1 = st.text_area(
                "Texto 1",
                height=250,
                key="discourse_live_text1",
                value=st.session_state.discourse_live_state.get('current_text1', ''),
                label_visibility="collapsed"
            )
            st.session_state.discourse_live_state['current_text1'] = text_input1

            # Segundo área de texto
            st.markdown("**Texto 2 (Comparación)**")
            text_input2 = st.text_area(
                "Texto 2",
                height=250,
                key="discourse_live_text2",
                value=st.session_state.discourse_live_state.get('current_text2', ''),
                label_visibility="collapsed"
            )
            st.session_state.discourse_live_state['current_text2'] = text_input2

            # Botón de análisis
            analyze_button = st.button(
                discourse_t.get('analyze_button', 'Analizar'),
                key="discourse_live_analyze",
                type="primary",
                icon="🔍",
                disabled=not (text_input1 and text_input2),
                use_container_width=True
            )

            if analyze_button and text_input1 and text_input2:
                try:
                    with st.spinner(discourse_t.get('processing', 'Procesando...')):
                        result = perform_discourse_analysis(
                            text_input1,
                            text_input2,
                            nlp_models[lang_code],
                            lang_code
                        )

                        if result['success']:
                            # Procesar ambos gráficos
                            for graph_key in ['graph1', 'graph2']:
                                if graph_key in result and result[graph_key] is not None:
                                    bytes_key = f'{graph_key}_bytes'
                                    graph_bytes = fig_to_bytes(result[graph_key])
                                    if graph_bytes:
                                        result[bytes_key] = graph_bytes
                                    plt.close(result[graph_key])

                            st.session_state.discourse_live_state['last_result'] = result
                            st.session_state.discourse_live_state['analysis_count'] += 1
                            
                            store_student_discourse_result(
                                st.session_state.username,
                                text_input1,
                                text_input2,
                                result
                            )
                        else:
                            st.error(result.get('message', 'Error en el análisis'))

                except Exception as e:
                    logger.error(f"Error en análisis: {str(e)}")
                    st.error(discourse_t.get('error_processing', f'Error al procesar el texto: {str(e)}'))

        # Columna derecha: Visualización de resultados
        with result_col:
            st.subheader(discourse_t.get('live_results', 'Resultados en vivo'))

            if 'last_result' in st.session_state.discourse_live_state and \
               st.session_state.discourse_live_state['last_result'] is not None:
                
                result = st.session_state.discourse_live_state['last_result']
                if all(key in result for key in ['graph1', 'graph2']):
                    display_discourse_results(result, lang_code, discourse_t)
                else:
                    logger.error(f"Faltan gráficos en el resultado: {list(result.keys())}")
                    st.error(discourse_t.get('missing_graphs', 'Error: No se pudieron generar todos los gráficos'))
            else:
                st.info(discourse_t.get('initial_message', 
                    'Ingrese los textos y presione Analizar para ver los resultados.'))

    except Exception as e:
        logger.error(f"Error general en interfaz del discurso en vivo: {str(e)}")
        st.error(discourse_t.get('general_error', "Se produjo un error. Por favor, intente de nuevo."))