Update app.py

app.py

@@ -33,7 +33,7 @@ def user(text_prompt: str, chatbot: CHAT_HISTORY):
 
 def bot(
     model_choice: str,
-    system_instruction: Optional[str],  # Sistema de instrucciones opcional
+    system_instruction: Optional[str],
     chatbot: CHAT_HISTORY
 ):
     if not GOOGLE_API_KEY:
@@ -47,9 +47,8 @@ def bot(
         top_p=0.9
     )
 
-    # Usar el valor por defecto para system_instruction si está vacío
     if not system_instruction:
-        system_instruction = "1"  # O puedes poner un valor predeterminado como "No system instruction provided."
+        system_instruction = "1"
 
     text_prompt = [chatbot[-1][0]] if chatbot and chatbot[-1][0] and isinstance(chatbot[-1][0], str) else []
     
@@ -69,59 +68,69 @@ def bot(
             time.sleep(0.01)
             yield chatbot
 
-# Componente para el acordeón que contiene el cuadro de texto para la instrucción del sistema
-system_instruction_component = gr.Textbox(
-    placeholder="Enter system instruction...",
-    show_label=True,
-    scale=8
-)
+def multimodal(file, chatbot: CHAT_HISTORY):
+    """
+    Procesa un archivo multimodal (imagen, PDF, texto) y genera resultados
+    utilizando Gemini Vision y Pro.
+    """
+    if not GOOGLE_API_KEY:
+        raise ValueError("GOOGLE_API_KEY is not set.")
+
+    genai.configure(api_key=GOOGLE_API_KEY)
+    file_type = file.name.split(".")[-1].lower()
+
+    if file_type in ["png", "jpg", "jpeg"]:
+        # Procesar imágenes
+        image = Image.open(file.name)
+        chatbot.append(("Image uploaded for analysis.", None))
+        result = genai.generate_images(
+            prompt="Analyze this image.",
+            image=image,
+            model="gemini-vision"
+        )
+        chatbot[-1] = ("Image uploaded for analysis.", result[0]["description"])
+
+    elif file_type == "txt":
+        # Procesar texto
+        content = file.read().decode("utf-8")
+        chatbot.append((content, None))
+        response = genai.generate_text(prompt=content, model="gemini-pro")
+        chatbot[-1] = (content, response.result)
+
+    elif file_type == "pdf":
+        # Procesar PDFs (extraer texto de la primera página)
+        import fitz  # PyMuPDF
+        doc = fitz.open(file.name)
+        page_text = doc[0].get_text() if len(doc) > 0 else "PDF vacío."
+        chatbot.append((page_text, None))
+        response = genai.generate_text(prompt=page_text, model="gemini-pro")
+        chatbot[-1] = (page_text, response.result)
+
+    else:
+        chatbot.append(("Unsupported file type.", "Please upload a valid file."))
+    
+    return chatbot
 
-# Definir los componentes de entrada y salida
+# Componentes de interfaz actualizados
 chatbot_component = gr.Chatbot(label='Gemini', bubble_full_width=False, scale=2, height=300)
-text_prompt_component = gr.Textbox(placeholder="Message...", show_label=False, autofocus=True, scale=8)
-run_button_component = gr.Button(value="Run", variant="primary", scale=1)
-model_choice_component = gr.Dropdown(
-    choices=["gemini-1.5-flash", "gemini-2.0-flash-exp", "gemini-1.5-pro"],
-    value="gemini-1.5-flash",
-    label="Select Model",
-    scale=2
-)
-
-user_inputs = [text_prompt_component, chatbot_component]
-bot_inputs = [model_choice_component, system_instruction_component, chatbot_component]
-
-# Definir la interfaz de usuario
+file_upload_component = gr.File(label="Upload File (Image, PDF, or TXT)")
+run_button_component = gr.Button(value="Analyze File", variant="primary", scale=1)
+
+user_inputs = [file_upload_component, chatbot_component]
+
+# Interfaz actualizada
 with gr.Blocks() as demo:
     gr.HTML(TITLE)
     gr.HTML(SUBTITLE)
     with gr.Column():
-        # Campo de selección de modelo arriba
-        model_choice_component.render()
         chatbot_component.render()
-        with gr.Row():
-            text_prompt_component.render()
-            run_button_component.render()
-
-        # Crear el acordeón para la instrucción del sistema al final
-        with gr.Accordion("System Instruction", open=False):  # Acordeón cerrado por defecto
-            system_instruction_component.render()
+        file_upload_component.render()
+        run_button_component.render()
 
     run_button_component.click(
-        fn=user,
-        inputs=user_inputs,
-        outputs=[text_prompt_component, chatbot_component],
-        queue=False
-    ).then(
-        fn=bot, inputs=bot_inputs, outputs=[chatbot_component],
-    )
-
-    text_prompt_component.submit(
-        fn=user,
+        fn=multimodal,
         inputs=user_inputs,
-        outputs=[text_prompt_component, chatbot_component],
-        queue=False
-    ).then(
-        fn=bot, inputs=bot_inputs, outputs=[chatbot_component],
+        outputs=[chatbot_component],
     )
 
 # Lanzar la aplicación