app.py

import streamlit as st
import pandas as pd
import re
from datetime import datetime
import fitz  # PyMuPDF
import plotly.express as px
import io

def extract_data_from_pdf(pdf_bytes):
    # Open the PDF file from bytes
    pdf_document = fitz.open(stream=pdf_bytes, filetype="pdf")
    extracted_texts = [page.get_text() for page in pdf_document]
    pdf_document.close()

    # Define a regular expression pattern to match the lines with temperature data
    pattern = re.compile(r'(\+\s\d+\sMin\.|\Début)\s*(\d+°C)\s*(\d+°C)\s*(\d{2},\d{2})')
    structured_data = []

    # Process the extracted text from each page
    for page_text in extracted_texts:
        # Extract metadata
        company_match = re.search(r'Entreprise::\s*(.+)', page_text)
        date_match = re.search(r'Date:\s*(\d{2}[./]\d{2}[./]\d{4})', page_text)
        user_match = re.search(r'Utilisateur:\s*(.+)', page_text)
        product_match = re.search(r'Produit:\s*(.+)', page_text)

        if not all([company_match, date_match, user_match, product_match]):
            continue  # Skip pages without metadata

        company = company_match.group(1).strip()
        date = date_match.group(1).replace('/', '.').strip()
        user = user_match.group(1).strip()
        product = product_match.group(1).strip()

        # Find all matches of the pattern in the text
        matches = pattern.findall(page_text)
        for match in matches:
            time, sterilizer_temp, core_temp, f_value = match
            structured_data.append({
                'Date': datetime.strptime(date, '%d.%m.%Y').strftime('%Y-%m-%d'),
                'Entreprise': company,
                'Utilisateur': user,
                'Produit': product,
                'Déroulement': time.strip(),
                'Temp. du stérilisateur': float(sterilizer_temp.replace('°C', '')),
                'Temp. à coeur': float(core_temp.replace('°C', '')),
                'Valeur F': float(f_value.replace(',', '.'))
            })

    return pd.DataFrame(structured_data)

def analyze_sterilization(data):
    results = []
    
    for product, group in data.groupby('Produit'):
        # Determine product type and required temperature
        is_nutabreizh = 'NutaBreizh' in product
        required_temp = 108 if is_nutabreizh else 103
        
        # Filter rows where core temperature is above or equal to the required temperature
        above_required_temp = group[group['Temp. à coeur'] >= required_temp]
        
        # Calculate the duration (in minutes) at the required temperature
        minutes_at_temp = len(above_required_temp)
        
        # Calculate max temperatures
        max_temp_sterilisateur = group['Temp. du stérilisateur'].max()
        max_temp_coeur = group['Temp. à coeur'].max()
        
        # Determine if criteria met (at least 30 minutes at required temperature)
        criteria_met = minutes_at_temp >= 30
        
        results.append({
            'Date': group['Date'].iloc[0],
            'Produit': product,
            'Utilisateur': group['Utilisateur'].iloc[0],
            'Temperature_Requise': required_temp,
            'Minutes_Temperature_Requise': minutes_at_temp,
            'Temperature_Max_Sterilisateur': max_temp_sterilisateur,
            'Temperature_Max_Coeur': max_temp_coeur,
            'Criteres_Respectes': criteria_met
        })
    
    return pd.DataFrame(results)

def main():
    st.title("Analyse des Protocoles de Stérilisation")
    
    uploaded_file = st.file_uploader("Choisir un fichier PDF", type="pdf")
    
    if uploaded_file is not None:
        # Read the uploaded file as bytes
        pdf_bytes = uploaded_file.getvalue()
        
        # Process data
        data = extract_data_from_pdf(pdf_bytes)
        results_df = analyze_sterilization(data)
        
        # Display results
        st.subheader("Résultats de l'analyse")
        st.dataframe(results_df)
        
        # Check if all criteria are met
        if not results_df['Criteres_Respectes'].all():
            st.warning("Attention : Certains produits n'ont pas respecté les critères de stérilisation.")
        
        # Display failed products
        failed_products = results_df[results_df['Criteres_Respectes'] == False]
        if not failed_products.empty:
            st.subheader("Produits n'ayant pas respecté les critères")
            st.dataframe(failed_products)
        
        # Create visualization
        if not results_df.empty:
            fig = px.scatter(results_df, 
                            x='Date', 
                            y='Minutes_Temperature_Requise',
                            color='Criteres_Respectes',
                            hover_data=['Produit', 'Temperature_Requise'],
                            title="Minutes à température requise par production")
            st.plotly_chart(fig)
        else:
            st.warning("Aucune donnée valide à afficher.")
        
        # Export button
        if st.button("Exporter en Excel"):
            output = io.BytesIO()
            with pd.ExcelWriter(output, engine='xlsxwriter') as writer:
                results_df.to_excel(writer, index=False)
            
            output.seek(0)
            st.download_button(
                label="Télécharger l'analyse",
                data=output,
                file_name="analyse_sterilisation.xlsx",
                mime="application/vnd.openxmlformats-officedocument.spreadsheetml.sheet"
            )

if __name__ == "__main__":
    main()