Spaces:
Sleeping
Sleeping
#modules/semantic/semantic_interface.py | |
import streamlit as st | |
from streamlit_float import * | |
from streamlit_antd_components import * | |
from streamlit.components.v1 import html | |
import spacy_streamlit | |
import io | |
from io import BytesIO | |
import base64 | |
import matplotlib.pyplot as plt | |
import pandas as pd | |
import re | |
import logging | |
# Configuración del logger | |
logger = logging.getLogger(__name__) | |
# Importaciones locales | |
from .semantic_process import ( | |
process_semantic_input, | |
format_semantic_results | |
) | |
from ..utils.widget_utils import generate_unique_key | |
from ..database.semantic_mongo_db import store_student_semantic_result | |
from ..database.chat_mongo_db import store_chat_history, get_chat_history | |
# from ..database.semantic_export import export_user_interactions | |
############################### | |
# En semantic_interface.py | |
def display_semantic_interface(lang_code, nlp_models, semantic_t): | |
try: | |
# 1. Inicializar el estado de la sesión | |
if 'semantic_state' not in st.session_state: | |
st.session_state.semantic_state = { | |
'analysis_count': 0, | |
'last_analysis': None, | |
'current_file': None, | |
'pending_analysis': False # Nuevo flag para controlar el análisis pendiente | |
} | |
# 2. Área de carga de archivo con mensaje informativo | |
st.info(semantic_t.get('initial_instruction', | |
'Para comenzar un nuevo análisis semántico, cargue un archivo de texto (.txt)')) | |
uploaded_file = st.file_uploader( | |
semantic_t.get('semantic_file_uploader', 'Upload a text file for semantic analysis'), | |
type=['txt'], | |
key=f"semantic_file_uploader_{st.session_state.semantic_state['analysis_count']}" | |
) | |
# 2.1 Verificar si hay un archivo cargado y un análisis pendiente | |
if uploaded_file is not None and st.session_state.semantic_state.get('pending_analysis', False): | |
try: | |
with st.spinner(semantic_t.get('processing', 'Processing...')): | |
# Realizar análisis | |
text_content = uploaded_file.getvalue().decode('utf-8') | |
analysis_result = process_semantic_input( | |
text_content, | |
lang_code, | |
nlp_models, | |
semantic_t | |
) | |
if analysis_result['success']: | |
# Guardar resultado | |
st.session_state.semantic_result = analysis_result | |
st.session_state.semantic_state['analysis_count'] += 1 | |
st.session_state.semantic_state['current_file'] = uploaded_file.name | |
# Guardar en base de datos | |
storage_success = store_student_semantic_result( | |
st.session_state.username, | |
text_content, | |
analysis_result['analysis'] | |
) | |
if storage_success: | |
st.success( | |
semantic_t.get('analysis_complete', | |
'Análisis completado y guardado. Para realizar un nuevo análisis, cargue otro archivo.') | |
) | |
else: | |
st.error(semantic_t.get('error_message', 'Error saving analysis')) | |
else: | |
st.error(analysis_result['message']) | |
# Restablecer el flag de análisis pendiente | |
st.session_state.semantic_state['pending_analysis'] = False | |
except Exception as e: | |
logger.error(f"Error en análisis semántico: {str(e)}") | |
st.error(semantic_t.get('error_processing', f'Error processing text: {str(e)}')) | |
# Restablecer el flag de análisis pendiente en caso de error | |
st.session_state.semantic_state['pending_analysis'] = False | |
# 3. Columnas para los botones y mensajes | |
col1, col2 = st.columns([1,4]) | |
# 4. Botón de análisis | |
with col1: | |
analyze_button = st.button( | |
semantic_t.get('semantic_analyze_button', 'Analyze'), | |
key=f"semantic_analyze_button_{st.session_state.semantic_state['analysis_count']}", | |
type="primary", | |
icon="🔍", | |
disabled=uploaded_file is None, | |
use_container_width=True | |
) | |
# 5. Procesar análisis | |
if analyze_button and uploaded_file is not None: | |
# En lugar de realizar el análisis inmediatamente, establecer el flag | |
st.session_state.semantic_state['pending_analysis'] = True | |
# Forzar la recarga de la aplicación | |
st.rerun() | |
# 6. Mostrar resultados previos o mensaje inicial | |
elif 'semantic_result' in st.session_state and st.session_state.semantic_result is not None: | |
# Mostrar mensaje sobre el análisis actual | |
st.info( | |
semantic_t.get('current_analysis_message', | |
'Mostrando análisis del archivo: {}. Para realizar un nuevo análisis, cargue otro archivo.' | |
).format(st.session_state.semantic_state["current_file"]) | |
) | |
display_semantic_results( | |
st.session_state.semantic_result, | |
lang_code, | |
semantic_t | |
) | |
else: | |
st.info(semantic_t.get('upload_prompt', 'Cargue un archivo para comenzar el análisis')) | |
except Exception as e: | |
logger.error(f"Error general en interfaz semántica: {str(e)}") | |
st.error(semantic_t.get('general_error', "Se produjo un error. Por favor, intente de nuevo.")) | |
####################################### | |
def display_semantic_results(semantic_result, lang_code, semantic_t): | |
""" | |
Muestra los resultados del análisis semántico de conceptos clave. | |
Versión simplificada que muestra el gráfico directamente. | |
""" | |
if semantic_result is None or not semantic_result['success']: | |
st.warning(semantic_t.get('no_results', 'No results available')) | |
return | |
analysis = semantic_result['analysis'] | |
# Mostrar conceptos clave | |
st.subheader(semantic_t.get('key_concepts', 'Key Concepts')) | |
if 'key_concepts' in analysis and analysis['key_concepts']: | |
df = pd.DataFrame( | |
analysis['key_concepts'], | |
columns=[ | |
semantic_t.get('concept', 'Concept'), | |
semantic_t.get('frequency', 'Frequency') | |
] | |
) | |
# Mostrar conceptos como chips | |
cols = st.columns(4) # 4 columnas para distribuir los conceptos | |
for i, (concept, freq) in enumerate(df.values): | |
with cols[i % 4]: | |
st.markdown( | |
f""" | |
<div style=" | |
background-color: #f0f2f6; | |
border-radius: 20px; | |
padding: 8px 12px; | |
margin: 5px 0; | |
text-align: center; | |
"> | |
<b>{concept}</b><br> | |
<small>{freq:.2f}</small> | |
</div> | |
""", | |
unsafe_allow_html=True | |
) | |
else: | |
st.info(semantic_t.get('no_concepts', 'No key concepts found')) | |
# Mostrar gráfico de conceptos directamente | |
if 'concept_graph' in analysis and analysis['concept_graph'] is not None: | |
try: | |
# Mostrar el gráfico directamente con st.image() | |
st.image( | |
analysis['concept_graph'], | |
use_column_width=True, | |
caption=semantic_t.get('graph_description', 'Visualización de relaciones entre conceptos clave') | |
) | |
# Sección de interpretación | |
with st.expander("📊 " + semantic_t.get('semantic_graph_interpretation', "Interpretación del gráfico")): | |
st.markdown(f""" | |
- 🔀 {semantic_t.get('semantic_arrow_meaning', 'Flechas: dirección de la relación')} | |
- 🎨 {semantic_t.get('semantic_color_meaning', 'Color: centralidad del concepto')} | |
- ⭕ {semantic_t.get('semantic_size_meaning', 'Tamaño: frecuencia del concepto')} | |
- ↔️ {semantic_t.get('semantic_thickness_meaning', 'Grosor: fuerza de la conexión')} | |
""") | |
# Botón de descarga | |
st.download_button( | |
label="📥 " + semantic_t.get('download_graph', "Descargar gráfico"), | |
data=analysis['concept_graph'], | |
file_name="semantic_network.png", | |
mime="image/png" | |
) | |
except Exception as e: | |
logger.error(f"Error al mostrar el gráfico: {str(e)}") | |
st.error(semantic_t.get('graph_error', 'Error al visualizar el gráfico')) | |
else: | |
st.info(semantic_t.get('no_graph', 'No se generó el gráfico de conceptos')) | |