app.py

import streamlit as st
import google.generativeai as genai
import os
import tempfile
import PIL.Image
import time
import ssl
from dotenv import load_dotenv

load_dotenv()

# Configure the API key
genai.configure(api_key=os.getenv("GOOGLE_API_KEY"))

safety_settings = [
    {"category": "HARM_CATEGORY_HARASSMENT", "threshold": "BLOCK_NONE"},
    {"category": "HARM_CATEGORY_HATE_SPEECH", "threshold": "BLOCK_NONE"},
    {"category": "HARM_CATEGORY_SEXUALLY_EXPLICIT", "threshold": "BLOCK_NONE"},
    {"category": "HARM_CATEGORY_DANGEROUS_CONTENT", "threshold": "BLOCK_NONE"},
]

model = genai.GenerativeModel('gemini-1.5-flash-002', safety_settings=safety_settings,
                              system_instruction="Tu es un assistant intelligent. ton but est d'assister au mieux que tu peux. tu as été créé par Aenir et tu t'appelles Mariam")

# Function to get response from the model
# Gemini uses 'model' for assistant; Streamlit uses 'assistant'
def role_to_streamlit(role):
    if role == "model":
        return "assistant"
    else:
        return role

# Add a Gemini Chat history object to Streamlit session state
if "chat" not in st.session_state:
    st.session_state.chat = model.start_chat(history=[])

# Display Form Title
st.title("Mariam AI!")

# File uploader
uploaded_files = st.file_uploader("Choose a file", accept_multiple_files=True)

# Display chat messages from history above current input box
for message in st.session_state.chat.history:
    with st.chat_message(role_to_streamlit(message.role)):
        st.markdown(message.parts[0].text)

def upload_and_process_file(file_path):
    """Upload et traite un fichier avec l'API Gemini avec gestion des erreurs améliorée"""
    max_retries = 3
    retry_delay = 2  # secondes

    for attempt in range(max_retries):
        try:
            print(f"Tentative d'upload {attempt + 1}/{max_retries} pour {file_path}")

            # Vérification du fichier
            if not os.path.exists(file_path):
                raise FileNotFoundError(f"Le fichier {file_path} n'existe pas")

            file_size = os.path.getsize(file_path)
            if file_size == 0:
                raise ValueError(f"Le fichier {file_path} est vide")

            # Upload du fichier
            uploaded_file = genai.upload_file(path=file_path)
            print(f"Upload réussi: {uploaded_file.uri}")

            # Attente du traitement
            timeout = 300  # 5 minutes
            start_time = time.time()

            while uploaded_file.state.name == "PROCESSING":
                if time.time() - start_time > timeout:
                    raise TimeoutError("Timeout pendant le traitement du fichier")

                print(
                    f"En attente du traitement... Temps écoulé: {int(time.time() - start_time)}s")
                time.sleep(10)
                uploaded_file = genai.get_file(uploaded_file.name)

            if uploaded_file.state.name == "FAILED":
                raise ValueError(
                    f"Échec du traitement: {uploaded_file.state.name}")

            print(f"Traitement terminé avec succès: {uploaded_file.uri}")
            return uploaded_file

        except ssl.SSLError as e:
            print(
                f"Erreur SSL lors de l'upload (tentative {attempt + 1}): {e}")
            if attempt < max_retries - 1:
                time.sleep(retry_delay * (attempt + 1))
            else:
                raise

        except Exception as e:
            print(
                f"Erreur lors de l'upload (tentative {attempt + 1}): {e}")
            if attempt < max_retries - 1:
                time.sleep(retry_delay * (attempt + 1))
            else:
                raise

# Accept user's next message, add to context, resubmit context to Gemini
if prompt := st.chat_input("Hey?"):
    # Display user's last message
    st.chat_message("user").markdown(prompt)

    content = [prompt]
    temp_files = []

    try:
        # Process uploaded files
        for uploaded_file in uploaded_files:
            with tempfile.NamedTemporaryFile(delete=False, suffix=os.path.splitext(uploaded_file.name)[1]) as temp_file:
                temp_file.write(uploaded_file.getvalue())
                temp_files.append(temp_file.name)
                if uploaded_file.name.lower().endswith(('.png', '.jpg', '.jpeg', '.gif')):
                    content.append(PIL.Image.open(temp_file.name))
                else:
                    processed_file = upload_and_process_file(temp_file.name)
                    content.append(processed_file)

        # Send user entry to Gemini and read the response
        response = model.generate_content(content)

        # Add the response to the chat history
        st.session_state.chat.history.extend([
            genai.types.Part(text=prompt, role="user"),
            genai.types.Part(text=response.text, role="model"),
        ])
        # Display last
        with st.chat_message("assistant"):
            st.markdown(response.text)

    except Exception as e:
        st.error(f"An error occurred: {e}")

    finally:
        # Cleanup temporary files
        for temp_file in temp_files:
            try:
                os.unlink(temp_file)
            except Exception as e:
                print(
                    f"Error deleting temporary file {temp_file}: {e}")