app.py

import json
import pandas as pd
import streamlit as st
from io import BytesIO
import requests

# Set Streamlit to wide mode
st.set_page_config(layout="wide")

# Function to flatten the nested JSON structure
def flatten_json_safe(nested_json, parent_key='', sep='_'):
    """Flatten a nested JSON dictionary, safely handling strings and primitives."""
    items = []
    if isinstance(nested_json, dict):
        for k, v in nested_json.items():
            new_key = f'{parent_key}{sep}{k}' if parent_key else k
            if isinstance(v, dict):
                items.extend(flatten_json_safe(v, new_key, sep=sep).items())
            elif isinstance(v, list):
                for i, item in enumerate(v):
                    items.extend(flatten_json_safe(item, f'{new_key}{sep}{i}', sep=sep).items())
            else:
                items.append((new_key, v))
    else:
        items.append((parent_key, nested_json))
    return dict(items)

# Function to extract data from the flattened JSON
def extract_from_flattened(flattened_data, mapping, selected_fields):
    extracted_data = {}
    for label, flat_path in mapping.items():
        if label in selected_fields:
            extracted_data[label] = flattened_data.get(flat_path, 'N/A')
    return extracted_data

# Custom CSS for the table display
def apply_table_css():
    st.markdown(
        """
        <style>
        table {
            width: 100%;
            border-collapse: collapse;
            background-color: #f9f9f9;
        }
        th, td {
            border: 1px solid #ddd;
            padding: 10px;
            text-align: left;
        }
        th {
            background-color: #f2f2f2;
        }
        </style>
        """, unsafe_allow_html=True
    )

# Load the CSV mapping for UUIDs corresponding to NUM from a URL
def load_uuid_mapping_from_url(url):
    response = requests.get(url)
    if response.status_code == 200:
        from io import StringIO
        csv_data = StringIO(response.text)
        uuid_mapping_df = pd.read_csv(csv_data)
        
        # Check if the columns 'UUID', 'Num', 'Chapitre', 'Theme', and 'SSTheme' exist and have non-empty values
        required_columns = ['UUID', 'Num', 'Chapitre', 'Theme', 'SSTheme']
        for column in required_columns:
            if column not in uuid_mapping_df.columns:
                st.error(f"Le fichier CSV doit contenir une colonne '{column}' avec des valeurs valides.")
                return {}
        
        uuid_mapping_df = uuid_mapping_df.dropna(subset=['UUID', 'Num'])  # Drop rows with empty 'UUID' or 'Num' values
        uuid_mapping_df['Chapitre'] = uuid_mapping_df['Chapitre'].astype(str).str.strip()
        uuid_mapping_df = uuid_mapping_df.drop_duplicates(subset=['Chapitre', 'Num'])  # Remove duplicate rows based on 'Chapitre' and 'Num'
        return uuid_mapping_df
    else:
        st.error("Impossible de charger le fichier CSV des UUID depuis l'URL fourni.")
        return pd.DataFrame()

# URL for the UUID CSV
UUID_MAPPING_URL = "https://raw.githubusercontent.com/M00N69/Gemini-Knowledge/refs/heads/main/IFSV8listUUID.csv"

UUID_MAPPING_DF = load_uuid_mapping_from_url(UUID_MAPPING_URL)

# Complete mapping based on your provided field names and JSON structure
FLATTENED_FIELD_MAPPING = {
    "Nom du site à auditer": "data_modules_food_8_questions_companyName_answer",
    "N° COID du portail": "data_modules_food_8_questions_companyCoid_answer",
    "Code GLN": "data_modules_food_8_questions_companyGln_answer_0_rootQuestions_companyGlnNumber_answer",
    "Rue": "data_modules_food_8_questions_companyStreetNo_answer",
    "Code postal": "data_modules_food_8_questions_companyZip_answer",
    "Nom de la ville": "data_modules_food_8_questions_companyCity_answer",
    "Pays": "data_modules_food_8_questions_companyCountry_answer",
    "Téléphone": "data_modules_food_8_questions_companyTelephone_answer",
    "Latitude": "data_modules_food_8_questions_companyGpsLatitude_answer",
    "Longitude": "data_modules_food_8_questions_companyGpsLongitude_answer",
    "Email": "data_modules_food_8_questions_companyEmail_answer",
    "Nom du siège social": "data_modules_food_8_questions_headquartersName_answer",
    "Rue (siège social)": "data_modules_food_8_questions_headquartersStreetNo_answer",
    "Nom de la ville (siège social)": "data_modules_food_8_questions_headquartersCity_answer",
    "Code postal (siège social)": "data_modules_food_8_questions_headquartersZip_answer",
    "Pays (siège social)": "data_modules_food_8_questions_headquartersCountry_answer",
    "Téléphone (siège social)": "data_modules_food_8_questions_headquartersTelephone_answer",
    "Surface couverte de l'entreprise (m²)": "data_modules_food_8_questions_productionAreaSize_answer",
    "Nombre de bâtiments": "data_modules_food_8_questions_numberOfBuildings_answer",
    "Nombre de lignes de production": "data_modules_food_8_questions_numberOfProductionLines_answer",
    "Nombre d'étages": "data_modules_food_8_questions_numberOfFloors_answer",
    "Nombre maximum d'employés dans l'année, au pic de production": "data_modules_food_8_questions_numberOfEmployeesForTimeCalculation_answer",
    "Langue parlée et écrite sur le site": "data_modules_food_8_questions_workingLanguage_answer",
    "Périmètre de l'audit": "data_modules_food_8_questions_scopeCertificateScopeDescription_en_answer",
    "Process et activités": "data_modules_food_8_questions_scopeProductGroupsDescription_answer",
    "Activité saisonnière ? (O/N)": "data_modules_food_8_questions_seasonalProduction_answer",
    "Une partie du procédé de fabrication est-elle sous traitée? (OUI/NON)": "data_modules_food_8_questions_partlyOutsourcedProcesses_answer",
    "Si oui lister les procédés sous-traités": "data_modules_food_8_questions_partlyOutsourcedProcessesDescription_answer",
    "Avez-vous des produits totalement sous-traités? (OUI/NON)": "data_modules_food_8_questions_fullyOutsourcedProducts_answer",
    "Si oui, lister les produits totalement sous-traités": "data_modules_food_8_questions_fullyOutsourcedProductsDescription_answer",
    "Avez-vous des produits de négoce? (OUI/NON)": "data_modules_food_8_questions_tradedProductsBrokerActivity_answer",
    "Si oui, lister les produits de négoce": "data_modules_food_8_questions_tradedProductsBrokerActivityDescription_answer",
    "Produits à exclure du champ d'audit (OUI/NON)": "data_modules_food_8_questions_exclusions_answer",
    "Préciser les produits à exclure": "data_modules_food_8_questions_exclusionsDescription_answer"
}

# Streamlit app
st.sidebar.title("Menu de Navigation")
option = st.sidebar.radio("Choisissez une option:", ["Extraction des données", "Exigences de la checklist", "Modification des données", "Exportation", "Plan d'actions"])

st.title("IFS NEO Form Data Extractor")

# Step 1: Upload the JSON (.ifs) file
uploaded_json_file = st.file_uploader("Charger le fichier IFS de NEO", type="ifs")

if uploaded_json_file:
    try:
        # Step 2: Load the uploaded JSON file
        json_data = json.load(uploaded_json_file)

        # Step 3: Flatten the JSON data
        flattened_json_data_safe = flatten_json_safe(json_data)

        if option == "Extraction des données":
            st.subheader("Champs disponibles pour l'extraction")
            select_all = st.checkbox("Sélectionner tous les champs")
            if select_all:
                selected_fields = list(FLATTENED_FIELD_MAPPING.keys())
            else:
                selected_fields = st.multiselect("Sélectionnez les champs que vous souhaitez extraire", list(FLATTENED_FIELD_MAPPING.keys()))
            if selected_fields:
                # Step 4: Extract the required data based on the selected fields
                extracted_data = extract_from_flattened(flattened_json_data_safe, FLATTENED_FIELD_MAPPING, selected_fields)

                # Step 5: Display the extracted data using Streamlit widgets for real editing
                st.subheader("Données extraites")
                edit_mode = st.checkbox("Modifier les données")
                updated_data = extracted_data.copy()

                if edit_mode:
                    for field, value in extracted_data.items():
                        if field in ["Périmètre de l'audit", "Process et activités", "Si oui lister les procédés sous-traités", "Si oui, lister les produits totalement sous-traités", "Si oui, lister les produits de négoce", "Préciser les produits à exclure"]:
                            updated_data[field] = st.text_area(f"{field}", value=value, height=150)
                        else:
                            updated_data[field] = st.text_input(f"{field}", value=value)
                else:
                    # Display in read-only table format
                    apply_table_css()
                    table_html = "<table><thead><tr><th>Field</th><th>Value</th></tr></thead><tbody>"
                    for field, value in extracted_data.items():
                        table_html += f"<tr><td>{field}</td><td>{value}</td></tr>"
                    table_html += "</tbody></table>"
                    st.markdown(table_html, unsafe_allow_html=True)

                # Step 6: Option to download the extracted data as an Excel file with formatting and COID in the name
                df = pd.DataFrame(list(updated_data.items()), columns=["Field", "Value"])

                # Extract the COID number to use in the file name
                numero_coid = updated_data.get("N° COID du portail", "inconnu")

                # Create the Excel file with column formatting
                output = BytesIO()

                # Create Excel writer and adjust column widths
                with pd.ExcelWriter(output, engine='openpyxl') as writer:
                    df.to_excel(writer, index=False, sheet_name="Données extraites")
                    
                    # Access the worksheet to modify the formatting
                    worksheet = writer.sheets["Données extraites"]
                    
                    # Adjust the width of each column based on the longest entry
                    for col in worksheet.columns:
                        max_length = max(len(str(cell.value)) for cell in col)
                        col_letter = col[0].column_letter  # Get the column letter
                        worksheet.column_dimensions[col_letter].width = max_length + 5  # Adjust column width

                # Reset the position of the output to the start
                output.seek(0)

                # Provide the download button with the COID number in the filename
                st.download_button(
                    label="Télécharger le fichier Excel",
                    data=output,
                    file_name=f'extraction_{numero_coid}.xlsx',
                    mime="application/vnd.openxmlformats-officedocument.spreadsheetml.sheet"
                )

        elif option == "Exigences de la checklist":
            st.subheader("Exigences de la checklist")
            if not UUID_MAPPING_DF.empty:
                # Filtering options with linked filtering
                chapitre_options = ["Tous"] + sorted(UUID_MAPPING_DF['Chapitre'].dropna().unique())
                chapitre_filter = st.selectbox("Filtrer par Chapitre", options=chapitre_options)
                filtered_df = UUID_MAPPING_DF

                if chapitre_filter != "Tous":
                    filtered_df = filtered_df[filtered_df['Chapitre'] == chapitre_filter]
                    theme_options = ["Tous"] + sorted(filtered_df['Theme'].dropna().unique())
                else:
                    theme_options = ["Tous"] + sorted(UUID_MAPPING_DF['Theme'].dropna().unique())
                theme_filter = st.selectbox("Filtrer par Thème", options=theme_options)

                if theme_filter != "Tous":
                    filtered_df = filtered_df[filtered_df['Theme'] == theme_filter]
                    sstheme_options = ["Tous"] + sorted(filtered_df['SSTheme'].dropna().unique())
                else:
                    sstheme_options = ["Tous"] + sorted(UUID_MAPPING_DF['SSTheme'].dropna().unique())
                sstheme_filter = st.selectbox("Filtrer par Sous-Thème", options=sstheme_options)

                if sstheme_filter != "Tous":
                    filtered_df = filtered_df[filtered_df['SSTheme'] == sstheme_filter]

                # Extracting checklist requirements from flattened JSON data
                checklist_requirements = []
                for _, row in filtered_df.iterrows():
                    key = row['Num']
                    uuid = row['UUID']
                    prefix = f"data_modules_food_8_checklists_checklistFood8_resultScorings_{uuid}"
                    explanation_text = flattened_json_data_safe.get(f"{prefix}_answers_englishExplanationText", "N/A")
                    detailed_explanation = flattened_json_data_safe.get(f"{prefix}_answers_explanationText", "N/A")
                    score_label = flattened_json_data_safe.get(f"{prefix}_score_label", "N/A")
                    response = flattened_json_data_safe.get(f"{prefix}_answers_fieldAnswers", "N/A")
                    checklist_requirements.append({
                        "Num": key,
                        "Explanation": explanation_text,
                        "Detailed Explanation": detailed_explanation,
                        "Score": score_label,
                        "Response": response
                    })

                # Convert to filtered table display
                apply_table_css()
                table_html = "<table><thead><tr><th>Numéro d'exigence</th><th>Explication</th><th>Explication Détaillée</th><th>Note</th><th>Réponse</th></tr></thead><tbody>"
                for req in checklist_requirements:
                    table_html += f"<tr><td>{req['Num']}</td><td>{req['Explanation']}</td><td>{req['Detailed Explanation']}</td><td>{req['Score']}</td><td>{req['Response']}</td></tr>"
                table_html += "</tbody></table>"
                st.markdown(table_html, unsafe_allow_html=True)
            else:
                st.error("Impossible de charger les données des UUID. Veuillez vérifier l'URL.")

    except json.JSONDecodeError:
        st.error("Erreur lors du décodage du fichier JSON. Veuillez vous assurer qu'il est au format correct.")
else:
    st.write("Le fichier de NEO doit être un (.ifs)")