Update app.py

app.py

@@ -1,5 +1,5 @@
-# ====================    El Detective de Alimentos (Versión 8.1 - Diagnóstico Diferencial)  =====================================
-# Mejoras: Presentación de un "Top 3" de resultados con justificación de síntomas.
+# ====================    El Detective de Alimentos (Versión 9.0 - Maestra)  =====================================
+# Mejoras: Corrección de errores de sintaxis, expansión masiva de diccionarios y refinamientos de UI.
 
 import streamlit as st
 import google.generativeai as genai
@@ -18,7 +18,7 @@ st.set_page_config(
     layout="wide"
 )
 
-# ... (Configuración de Gemini y carga de datos no cambia) ...
+# --- CONFIGURACIÓN Y CARGA DE DATOS (SIN CAMBIOS) ---
 # Configurar logging
 logging.basicConfig(level=logging.INFO, format='%(asctime)s - %(levelname)s - %(message)s')
 logger = logging.getLogger("food_detective_app")
@@ -74,39 +74,43 @@ def load_data():
 alimentos_data, lista_condiciones, foodb_index = load_data()
 
 
-# --- DICCIONARIOS Y FUNCIONES AUXILIARES (CON CAMBIOS) ---
-# ... (Los diccionarios no cambian) ...
+# --- DICCIONARIOS DE MAPEADO EXPANDIDOS ---
 FOOD_TO_COMPOUND_MAP = {
-    "pan": ["gluten"], "trigo": ["gluten"], "harina de trigo": ["gluten"], "cebada": ["gluten"], "centeno": ["gluten"], "pasta": ["gluten"], "galletas": ["gluten"], "avena": ["gluten"], "pizza": ["gluten"], "torta": ["gluten"],
+    "pan": ["gluten"], "trigo": ["gluten"], "harina": ["gluten"], "cebada": ["gluten"], "centeno": ["gluten"], "pasta": ["gluten"], "galletas": ["gluten"], "avena": ["gluten"], "pizza": ["gluten"], "torta": ["gluten"], "pastel": ["gluten"], "cerveza": ["gluten", "histamina", "tiramina", "purinas"],
     "leche": ["lácteos", "caseína", "lactosa"], "queso": ["lácteos", "caseína", "lactosa", "histamina", "tiramina"], "yogur": ["lácteos", "caseína", "lactosa"], "mantequilla": ["lácteos", "caseína", "lactosa"], "crema": ["lácteos"], "helado": ["lácteos"],
-    "manzana": ["salicilatos", "fructosa"], "almendras": ["salicilatos", "arginina"], "uvas": ["salicilatos"], "pasas": ["salicilatos"], "naranja": ["salicilatos"], "brócoli": ["salicilatos", "goitrógenos", "fodmaps"], "cúrcuma": ["salicilatos"],
+    "manzana": ["salicilatos", "fructosa"], "pera": ["fructosa", "polioles"], "mango": ["fructosa"], "cereza": ["fructosa", "salicilatos"], "sandía": ["fructosa"], "almendras": ["salicilatos", "arginina", "oxalatos"], "uvas": ["salicilatos", "fructosa"], "pasas": ["salicilatos", "fructosa"], "naranja": ["salicilatos", "ácidos"], "limón": ["salicilatos", "ácidos"], "fresa": ["salicilatos"], "arándano": ["salicilatos"],
+    "brócoli": ["salicilatos", "goitrógenos", "fodmaps"], "coliflor": ["goitrógenos", "fodmaps"], "repollo": ["goitrógenos", "fodmaps"], "col": ["goitrógenos", "fodmaps"], "cúrcuma": ["salicilatos"],
     "azucar": ["azúcar", "fructosa"], "dulces": ["azúcar"], "refrescos": ["azúcar", "fructosa"], "gaseosas": ["azúcar", "fructosa"], "miel": ["fructosa", "fodmaps"], "jarabe de maiz": ["fructosa"],
-    "vino tinto": ["histamina", "tiramina", "sulfitos"], "vino rojo": ["histamina", "tiramina", "sulfitos"], "vino": ["histamina", "tiramina", "sulfitos"], "cerveza": ["histamina", "tiramina", "purinas", "gluten"], "chocolate": ["cafeína", "tiramina", "níquel", "arginina"],
-    "embutidos": ["histamina", "tiramina", "nitritos"], "pescado enlatado": ["histamina"], "tomate": ["histamina", "solaninas", "ácidos"],
-    "aguacate": ["fodmaps", "polioles"], "cebolla": ["fodmaps"], "ajo": ["fodmaps"], "legumbres": ["fodmaps", "gos"],
-    "carne": ["alfa-gal", "purinas", "hierro"], "carnes rojas": ["purinas", "alfa-gal", "hierro"],
-    "mariscos": ["purinas", "sulfitos", "alérgenos", "yodo"], "huevo": ["alérgenos"], "soya": ["alérgenos"],
-    "café": ["cafeína", "ácidos"], "nueces": ["arginina", "salicilatos", "níquel"]
+    "vino tinto": ["histamina", "tiramina", "sulfitos"], "vino rojo": ["histamina", "tiramina", "sulfitos"], "vino": ["histamina", "tiramina", "sulfitos"], "chocolate": ["cafeína", "tiramina", "níquel", "arginina", "oxalatos"],
+    "embutidos": ["histamina", "tiramina", "nitritos"], "pescado enlatado": ["histamina"], "atún en lata": ["histamina"], "sardinas": ["histamina", "purinas"], "tomate": ["histamina", "solaninas", "ácidos", "lectinas"],
+    "aguacate": ["fodmaps", "polioles", "histamina"], "cebolla": ["fodmaps", "fructanos"], "ajo": ["fodmaps", "fructanos"], "legumbres": ["fodmaps", "gos", "lectinas", "fitatos"], "lentejas": ["fodmaps", "gos", "lectinas"], "garbanzos": ["fodmaps", "gos", "lectinas"], "frijoles": ["fodmaps", "gos", "lectinas"],
+    "carne": ["alfa-gal", "purinas", "hierro", "histamina"], "carnes rojas": ["purinas", "alfa-gal", "hierro"], "hígado": ["purinas", "hierro", "vitamina a"],
+    "mariscos": ["purinas", "sulfitos", "alérgenos", "yodo", "níquel"], "huevo": ["alérgenos"], "soya": ["alérgenos", "fitatos", "goitrógenos"], "soja": ["alérgenos", "fitatos", "goitrógenos"],
+    "café": ["cafeína", "ácidos"], "té": ["cafeína", "taninos", "oxalatos"], "nueces": ["arginina", "salicilatos", "níquel", "oxalatos"], "maní": ["alérgenos", "arginina", "lectinas", "aflatoxinas"], "cacahuetes": ["alérgenos", "arginina", "lectinas", "aflatoxinas"],
+    "pimiento": ["solaninas", "lectinas"], "berenjena": ["solaninas", "lectinas"], "patata": ["solaninas", "lectinas"]
 }
 CONDITION_SYNONYMS = {
     "síndrome del intestino irritable (sii).": ["intolerancia a los fodmaps", "intolerancia a los gos (fodmap)", "intolerancia a gomas fermentables"],
     "gota / hiperuricemia.": ["acido urico aumentado"], "intolerancia a la lactosa.": ["déficit de lactasa"],
     "enfermedad celíaca (clásica).": ["dermatitis herpetiforme"], "migraña.": ["dolor de cabeza", "cefalea"]
 }
+
+# --- DICCIONARIO TRADUCTOR CORREGIDO Y EXPANDIDO ---
 FOOD_NAME_TO_FOODB_KEY = {
     "pan": ["bread"], "pasta": ["pasta"], "galleta": ["cookie"], "pizza": ["pizza"], "cebada": ["barley"], "centeno": ["rye"],
     "leche": ["milk"], "queso": ["cheese"], "huevo": ["egg"], "carne": ["beef", "pork", "lamb", "meat"], "ternera": ["beef"], "cerdo": ["pork"], "cordero": ["lamb"], "pollo": ["chicken"],
-    "manzana": ["apple"], "naranja": ["orange"], "uva": ["grape"], "plátano": ["banana"], "aguacate": ["avocado"], "limón": ["lemon"],
-    "tomate": ["tomato"], "patata": ["potato"], "cebolla": ["onion"], "ajo": ["garlic"], "espinaca": ["spinach"],
-    "vino": ["wine"], "cerveza": ["beer"], "café": ["coffee"], "chocolate": ["chocolate"], "miel": ["honey"],
-    "almendra": ["almond"], "nuez": ["walnut"], "cacahuete": ["peanut"], "arroz": ["rice"], "maíz": ["corn"],
-    "pescado": ["fish"], "atún": ["tuna", "salmón": "salmon", "marisco": "shellfish", "camarón": "shrimp"]
+    "manzana": ["apple"], "naranja": ["orange"], "uva": ["grape"], "plátano": ["banana"], "aguacate": ["avocado"], "limón": ["lemon"], "fresa": ["strawberry"], "pera": ["pear"], "mango": ["mango"],
+    "tomate": ["tomato"], "patata": ["potato"], "cebolla": ["onion"], "ajo": ["garlic"], "espinaca": ["spinach"], "zanahoria": ["carrot"], "pimiento": ["bell pepper"], "brócoli": ["broccoli"],
+    "vino": ["wine", "red wine", "white wine"], "cerveza": ["beer"], "café": ["coffee"], "chocolate": ["chocolate"], "miel": ["honey"], "té": ["tea"],
+    "almendra": ["almond"], "nuez": ["walnut"], "cacahuete": ["peanut"], "arroz": ["rice"], "maíz": ["corn"], "lenteja": ["lentil"], "frijol": ["bean"],
+    "pescado": ["fish"], "atún": ["tuna"], "salmón": ["salmon"], "marisco": ["shellfish"], "camarón": ["shrimp"], "sardina": ["sardine"]
 }
+
+# --- LÓGICA DE BÚSQUEDA Y ANÁLISIS (SIN CAMBIOS) ---
 def sanitize_text(text):
     if not text: return ""
     return re.sub(r'[.,;()]', '', text).lower().strip()
 def extract_and_infer_with_gemini(query, condiciones):
-    # ... (Sin cambios)
     if not model: return None
     condiciones_str = "\n".join([f"- {c}" for c in condiciones])
     system_prompt = f"""
@@ -131,8 +135,6 @@ def extract_and_infer_with_gemini(query, condiciones):
         logger.error(f"Error en la extracción/inferencia con Gemini: {e}")
         st.error(f"Hubo un problema al interpretar tu consulta con la IA.")
         return None
-
-# --- FUNCIÓN DE BÚSQUEDA MODIFICADA PARA DEVOLVER MÁS INFO ---
 def find_best_matches_hybrid(entities, data):
     if not entities or not data: return []
     user_symptoms = set(sanitize_text(s) for s in entities.get("sintomas", []))
@@ -160,7 +162,7 @@ def find_best_matches_hybrid(entities, data):
     results = []
     for entry in target_data:
         score_details = {'condition': 0, 'food': 0, 'symptoms': 0, 'total': 0}
-        matched_symptoms = [] # <-- NUEVO: Guardamos las pistas
+        matched_symptoms = []
         if inferred_condition_raw and sanitize_text(entry.get("condicion_asociada", "")) == inferred_condition_raw:
             score_details['condition'] = 100
         entry_compounds_text = sanitize_text(entry.get("compuesto_alimento", ""))
@@ -172,7 +174,7 @@ def find_best_matches_hybrid(entities, data):
             for key in entry_symptoms_keys:
                 if key in user_symptom or user_symptom in key:
                     symptom_score += 10
-                    matched_symptoms.append(key) # <-- NUEVO: Añadimos la pista
+                    matched_symptoms.append(key)
                     break
         score_details['symptoms'] = symptom_score
         total_score = sum(score_details.values())
@@ -181,7 +183,6 @@ def find_best_matches_hybrid(entities, data):
     if not results: return []
     sorted_results = sorted(results, key=lambda x: x['score']['total'], reverse=True)
     return sorted_results
-# ... (El resto de funciones auxiliares no cambia) ...
 def generate_detailed_analysis(query, match):
     if not model: return "Error: El modelo de IA no está disponible."
     prompt_parts = [
@@ -232,26 +233,12 @@ def create_relevance_chart(results):
     ).properties(title='Principales Coincidencias según tu Caso').configure_axis(labelFontSize=12, titleFontSize=14).configure_title(fontSize=16, anchor='start')
     return chart
 def log_feedback(query, result, feedback):
-    log_entry = {
-        "timestamp": datetime.now().isoformat(),
-        "query": query,
-        "best_match_condition": result['entry']['condicion_asociada'],
-        "score": result['score']['total'],
-        "feedback": feedback
-    }
+    log_entry = {"timestamp": datetime.now().isoformat(), "query": query, "best_match_condition": result['entry']['condicion_asociada'], "score": result['score']['total'], "feedback": feedback}
     os.makedirs("logs", exist_ok=True)
     with open(os.path.join("logs", "feedback_log.txt"), "a", encoding="utf-8") as f:
         f.write(json.dumps(log_entry, ensure_ascii=False) + "\n")
-def extract_foods_from_string(food_string):
-    match = re.search(r'\(([^)]+)\)$', food_string)
-    if match:
-        foods_part = match.group(1)
-        return [food.strip().lower() for food in foods_part.split(',') if food.strip()]
-    else:
-        main_part = re.sub(r'^\w+\s*\(.*?\)\s*', '', food_string).strip()
-        return [food.strip().lower() for food in main_part.split(',') if food.strip()]
 
-# --- INTERFAZ DE USUARIO Y LÓGICA PRINCIPAL (CON CAMBIOS) ---
+# --- INTERFAZ DE USUARIO Y LÓGICA PRINCIPAL ---
 col_img, col_text = st.columns([1, 4], gap="medium")
 with col_img:
     if os.path.exists("imagen.png"):
@@ -277,7 +264,7 @@ with st.form(key="search_form"):
     submitted = st.form_submit_button("Analizar mi caso", type="primary")
 
 if submitted:
-    clear_search_state() # Limpiamos el estado en cada nueva búsqueda
+    clear_search_state()
     if not query:
         st.warning("Por favor, describe lo que sientes y lo que comiste.")
     elif alimentos_data is None:
@@ -312,14 +299,47 @@ if st.session_state.search_results is not None:
         chart = create_relevance_chart(results)
         st.altair_chart(chart, use_container_width=True)
 
-        # --- ANÁLISIS DEL RESULTADO PRINCIPAL ---
         best_match_data = results[0]
         best_match = best_match_data['entry']
         with st.expander(f"**Análisis Detallado de la Principal Coincidencia: {best_match.get('condicion_asociada')}**", expanded=True):
-            # ... (código del desglose de puntuación y popover de FoodB no cambia)
-            # (El código para mostrar el análisis detallado del #1 es el mismo que antes)
+            col1, col2 = st.columns([3, 1])
+            with col1:
+                st.markdown("##### Desglose de la Puntuación de Relevancia:")
+                score_col1, score_col2, score_col3, score_col4 = st.columns(4)
+                score_col1.metric("Puntos por Condición", f"{best_match_data['score']['condition']}")
+                score_col2.metric("Puntos por Alimento", f"{best_match_data['score']['food']}")
+                score_col3.metric("Puntos por Síntomas", f"{best_match_data['score']['symptoms']}")
+                score_col4.metric("PUNTUACIÓN TOTAL", f"{best_match_data['score']['total']}", delta="Máxima coincidencia")
+            with col2:
+                st.write("")
+                if foodb_index:
+                    with st.popover("🔬 Principales componentes moleculares"):
+                        user_foods_mentioned = st.session_state.entities.get("alimentos", [])
+                        if not user_foods_mentioned:
+                            st.info("El usuario no especificó un alimento, no se puede realizar la búsqueda molecular.")
+                        else:
+                            found_data = False
+                            displayed_foodb_keys = set()
+                            for alimento_es in user_foods_mentioned:
+                                search_terms_en = []
+                                for key_es, value_en_list in FOOD_NAME_TO_FOODB_KEY.items():
+                                    if key_es in alimento_es.lower():
+                                        search_terms_en.extend(value_en_list)
+                                for term in set(search_terms_en):
+                                    for foodb_key, foodb_data in foodb_index.items():
+                                        if term in foodb_key and foodb_key not in displayed_foodb_keys:
+                                            found_data = True
+                                            displayed_foodb_keys.add(foodb_key)
+                                            with st.container(border=True):
+                                                st.subheader(f"Análisis de: {foodb_key.capitalize()}")
+                                                for item in foodb_data[:3]:
+                                                    st.write(f"**Compuesto:** {item['compound']}")
+                                                    st.write(f"**Efectos reportados:** {', '.join(item['effects'])}")
+                                            st.markdown("---")
+                            if not found_data:
+                                st.warning("Sin datos moleculares para este alimento.")
+            st.markdown("---")
             with st.spinner("✍️ Generando un análisis personalizado con IA..."):
-                # Usamos un caché para no regenerar el análisis si ya existe
                 if 'best_match_analysis' not in st.session_state.analysis_cache:
                     st.session_state.analysis_cache['best_match_analysis'] = generate_detailed_analysis(st.session_state.user_query, best_match)
                 st.markdown(st.session_state.analysis_cache['best_match_analysis'])
@@ -333,27 +353,24 @@ if st.session_state.search_results is not None:
                 log_feedback(st.session_state.user_query, best_match_data, "no_util")
                 st.warning("Gracias. Usaremos tu feedback para mejorar.")
 
-        # --- NUEVA SECCIÓN PARA "DIAGNÓSTICO DIFERENCIAL" ---
         if len(results) > 1:
             with st.expander("**Otras Posibilidades Relevantes (Diagnóstico Diferencial)**"):
-                # Mostramos las siguientes 2 o 3 posibilidades
                 for i, result in enumerate(results[1:4]):
-                    entry = result['entry']
-                    score = result['score']
-                    st.subheader(f"{i+2}. {entry.get('condicion_asociada')}")
-                    col1, col2 = st.columns([2,1])
-                    with col1:
-                        st.write(f"**Puntuación Total de Relevancia:** {score['total']}")
-                        if result.get('matched_symptoms'):
-                            st.write(f"**Pistas Clave (Síntomas Coincidentes):** {', '.join(result['matched_symptoms']).capitalize()}")
-                        st.write(f"**Alimentos Típicos Asociados:** {entry.get('compuesto_alimento')}")
-                    
-                    with col2:
-                        analysis_key = f"analysis_{i+2}"
-                        if st.button(f"Generar análisis para esta opción", key=analysis_key):
-                            with st.spinner("Generando análisis..."):
-                                st.session_state.analysis_cache[analysis_key] = generate_detailed_analysis(st.session_state.user_query, entry)
-                        
-                        if analysis_key in st.session_state.analysis_cache:
-                            st.info(st.session_state.analysis_cache[analysis_key])
-                    st.markdown("---")
+                    with st.container(border=True):
+                        entry = result['entry']
+                        score = result['score']
+                        st.subheader(f"{i+2}. {entry.get('condicion_asociada')}")
+                        col1, col2 = st.columns([2,1])
+                        with col1:
+                            st.write(f"**Puntuación Total de Relevancia:** {score['total']}")
+                            if result.get('matched_symptoms'):
+                                st.write(f"**Pistas Clave (Síntomas Coincidentes):** {', '.join(result['matched_symptoms']).capitalize()}")
+                            st.write(f"**Alimentos Típicos Asociados:** {entry.get('compuesto_alimento')}")
+                        with col2:
+                            analysis_key = f"analysis_{i+2}"
+                            if st.button(f"Generar análisis para esta opción", key=analysis_key):
+                                with st.spinner("Generando análisis..."):
+                                    st.session_state.analysis_cache[analysis_key] = generate_detailed_analysis(st.session_state.user_query, entry)
+                            if analysis_key in st.session_state.analysis_cache:
+                                st.info(st.session_state.analysis_cache[analysis_key])
+