Update app.py

app.py

@@ -1,269 +1,757 @@
 import gradio as gr
-import requests
-from bs4 import BeautifulSoup
-from urllib.parse import urljoin, urlparse
 import pandas as pd
-import tempfile
 import os
-import io
-import tldextract
-import tabulate # Required for df.to_markdown()
-
-# ----------------------------------------------------
-# ⚠️ SELENIUM IMPORTS AND CONFIGURATION ⚠️
-# ----------------------------------------------------
-# These imports are essential for simulating user interaction.
-from selenium import webdriver
-from selenium.webdriver.common.by import By
-from selenium.webdriver.chrome.service import Service
-from selenium.webdriver.chrome.options import Options
-from selenium.webdriver.support.ui import WebDriverWait
-from selenium.webdriver.support import expected_conditions as EC
-# If using webdriver-manager (recommended for local setup):
-# from webdriver_manager.chrome import ChromeDriverManager 
 import time
+import json
+from typing import List, Union, Dict, Any
 
-# -------------------------
-# Default excluded domains
-# -------------------------
-DEFAULT_EXCLUDED = [
-    'facebook.com', 'instagram.com', 'linkedin.com', 'youtube.com', 'google.com',
-    'apple.com', 'amazon.com', 'microsoft.com', 'meta.com', 't.co', 'x.com',
-    'twitter.com', 'whatsapp.com', 'github.com', 'tiktok.com',
-    'ulule.com', 'kisskissbankbank.com', 'leetchi.com', 'helloasso.com',
-    'gofundme.com', 'tipeee.com', 'patreon.com', 'paypal.com', 'stripe.com',
-    'buymeacoffee.com', 'wordpress.org', 'wordpress.com', 'wp.me', 'bsky.app', 'mamot.fr'
-]
-
-# -------------------------
-# Load Tracker DB (placeholder/simulated for functionality)
-# -------------------------
-TRACKER_CSV_PATH = "tracker_db.csv"
-if os.path.exists(TRACKER_CSV_PATH):
-    TRACKER_DB = pd.read_csv(TRACKER_CSV_PATH)
-    print(f"Tracker DB loaded ({len(TRACKER_DB)} entries).")
-else:
-    print("Warning: tracker_db.csv not found. Partner names will be 'Unknown'. Using simulated data for testing.")
-    TRACKER_DB = pd.DataFrame({
-        "domain": ["doubleclick.net", "trksite.biz", "adservice.com", "analytics.com", "mediarithmics.com"],
-        "name": ["Google Marketing", "Tracker Site", "Ad Service Network", "Web Analytics", "Mediarithmics"],
-        "type": ["Advertising", "Tracking", "Advertising", "Analytics", "Tracking"],
-        "commercial": ["Yes", "Yes", "Yes", "No", "Yes"]
-    })
-
-# -------------------------
-# Helper functions
-# -------------------------
-def get_registered_domain(domain):
-    ext = tldextract.extract(domain)
-    return ext.registered_domain.lower() if ext.registered_domain else domain.lower()
-
-def lookup_partner(domain):
-    rd = get_registered_domain(domain)
-    row = TRACKER_DB[TRACKER_DB['domain'] == rd]
-    if not row.empty:
-        return row.iloc[0]['name'], row.iloc[0]['type'], row.iloc[0]['commercial']
-    else:
-        return rd, "Unknown", "Unknown"
-
-def detect_cookie_purpose(cookie_name):
-    n = cookie_name.lower()
-    if any(x in n for x in ("_ga","_gid","analytics","_gat","matomo")):
-        return "Analytics"
-    if any(x in n for x in ("ads","ad","banners","pixel","tracking","trk","fbp","tld")):
-        return "Advertising/Tracking"
-    if any(x in n for x in ("session","auth","login","user","uid","csrftoken","lang")):
-        return "Functional"
-    return "Unknown"
-
-def build_partner_summary(cookie_records):
-    df = pd.DataFrame(cookie_records)
-    if df.empty:
-        return pd.DataFrame()
+# =================================================================
+# CONFIGURATION AND LANGUAGE DEFINITIONS
+# =================================================================
+
+CSV_FILENAME = f"survey_data_{int(time.time())}.csv"
+# Corrected count: 1 (lang_code) + 119 (data inputs) = 120
+EXPECTED_COUNT = 120 
+
+# --- Shared Choices (Keys are in French for data consistency) ---
+FREQ_FR = ["Jamais", "Rarement", "Parfois", "Souvent"]
+FREQ_PT = ["Nunca", "Raramente", "Às vezes", "Frequentemente"]
+
+THEMES_ACTU_FR = ["politique", "environnement", "économie", "santé", "social/société", "situation internationale/géopolitique", "science", "modes de vie (lifestyle)", "vie culturelle", "autre/précisez"]
+THEMES_ACTU_PT = ["política", "ambiente", "economia", "saúde", "social/sociiedade", "situação internacional/geopolítica", "ciência", "estilos de vida (lifestyle)", "vida cultural", "outro/especifique"]
+
+
+# --- Language Definitions ---
+LANGUAGES = {}
+try:
+    with open("fr.json", "r", encoding="utf-8") as f:
+        LANGUAGES["FR"] = json.load(f)
+    with open("pt.json", "r", encoding="utf-8") as f:
+        LANGUAGES["PT"] = json.load(f)
+except Exception as e:
+    print(f"Error loading JSON file: {e}")
+    # Fallback to a minimal structure if loading fails to prevent immediate crash
+    LANGUAGES["FR"] = LANGUAGES.get("FR", {"TITLE": "Erreur de chargement", "LANG_LABEL": "Sélectionner la Langue", "SUBMIT_BUTTON_LABEL": "Submit", "SUBMISSION_MESSAGE_INIT": "Error loading language files. Check console.", "DOWNLOAD_LABEL": "Download"})
+    LANGUAGES["PT"] = LANGUAGES.get("PT", {"TITLE": "Erro de carregamento", "LANG_LABEL": "Selecionar Idioma", "SUBMIT_BUTTON_LABEL": "Submit", "SUBMISSION_MESSAGE_INIT": "Erro ao carregar arquivos de idioma. Verifique o console.", "DOWNLOAD_LABEL": "Download"})
+
+
+# =================================================================
+# GRADING APP DEFINITIONS
+# =================================================================
+
+def load_data(lang: str) -> Dict[str, Any]:
+    """Retrieves data for the selected language, defaulting to French keys if missing."""
+    data = LANGUAGES.get(lang, LANGUAGES["FR"])
+    
+    # Check for critical nested structure completeness
+    # If the file was truncated, these keys might be missing. Fall back to French content.
+    if 'PRAT_CULT_PRACTICES' not in data:
+         data['PRAT_CULT_PRACTICES'] = LANGUAGES["FR"]['PRAT_CULT_PRACTICES']
+    if 'PLATFORM_ITEMS' not in data:
+         data['PLATFORM_ITEMS'] = LANGUAGES["FR"]['PLATFORM_ITEMS']
+    if 'PURPOSE_ITEMS' not in data:
+         data['PURPOSE_ITEMS'] = LANGUAGES["FR"]['PURPOSE_ITEMS']
+    if 'PCS_CHOICES' not in data:
+         data['PCS_CHOICES'] = LANGUAGES["FR"]['PCS_CHOICES']
+    
+    return data
+
+# --- Data Collection and Submission Function ---
+
+def submit_survey(
+    lang_code: str,
+    # TAB 1 - Contact (6)
+    enqueteur: str, approach: str, refusal_reason: str, refusal_other: str, contact_later: str, firstname: str, 
+    # TAB 1 - Lieux (17)
+    space1: str, lieu_1: str, lieu_2: str, lieu_3: str, space2b: List[str], space3_1: int, space3_2: int, space3_3: int, space4: str, space5: str, space6: str, space7_1: str, space7_2: str, space7_3: str, space8: str, space9: str, space10: List[str],
+    # TAB 2 - Engagement (5)
+    info1: List[str], engagement_organisation: List[str], engagement_domaine: List[str], info_activites: List[str], info_reseaux_sociaux: str,
+    # TAB 3 - News (12)
+    info_frequence_actu: str, info2: List[str], info3: List[str], info_opposite_feeling: str, info_contradict_opinion: str, info4: List[str], info5: List[str],
+    info6_logement: int, info6_politique: int, info6_etudes: int, info6_climat: int, info6_sociales: int, info6_sentimentale: int, info6_securite: int, info6_estime: int, info6_liberte: int,
+    # TAB 4 - Culturel (58)
+    prat_cult1: str,
+    # 14 checkboxes + 14 sliders for cultural practices
+    *cultural_practices_values, # 28 arguments <--- COMMA REMOVED
+    prat_nature: str, 
+    # 10 checkboxes + 10 number fields for platforms
+    *platform_values, # 20 arguments <--- COMMA REMOVED
+    purpose_autre_detail: str,
+    # 8 checkboxes + 5 sliders for purpose
+    *purpose_values, # 13 arguments <--- COMMA REMOVED if it was there
+    
+    # TAB 5 - Démographie (12)
+    demo_gender: str, demo_age: int,
+    demo_location_commune: str, demo_location_arrond: str,
+    demo_parents_location: str, demo_inscription: str,
+    demo_discipline: str, demo_job: str, demo_income: str,
+    demo_socialcapital1_parent1: str, demo_socialcapital1_parent2: str,
+    demo_socialcapital2: int,
+    
+    # TAB 7 - Open Questions (3)
+    open_non_institutionnel: str, open_alternatives: str, open_motivations: str
+    
+) -> List[Union[str, gr.File]]:
+    """Collects survey data, saves it to a CSV, and returns a success message and the file path."""
+    
+    # --------------------------------------------------------------------------------
+    # 1. Sanity Check on Arguments
+    # --------------------------------------------------------------------------------
+    
+    total_expected_params = EXPECTED_COUNT
+    
+    # Get all locals (arguments)
+    all_args = locals()
+    
+    # Arguments that are *args tuples:
+    received_cultural = len(all_args['cultural_practices_values'])
+    received_platform = len(all_args['platform_values'])
+    received_purpose = len(all_args['purpose_values'])
+    
+    # Count of named arguments: 120 - 28 - 20 - 13 = 59 (including lang_code)
+    named_args_count = 59
+    
+    total_received_params = named_args_count + received_cultural + received_platform + received_purpose
+    
+    if total_received_params != total_expected_params:
+        print(f"ERROR: Expected {total_expected_params} inputs, but received {total_received_params}. Check component count.")
+        # Load language data for error message
+        current_lang_data = load_data(lang_code)
+        error_msg = current_lang_data.get("SUBMISSION_MESSAGE_ERROR", "❌ Submission Error. Check component count.")
+        error_msg = error_msg.format(received=total_received_params, expected=total_expected_params)
+        return [error_msg, None]
+
+
+    # --------------------------------------------------------------------------------
+    # 2. Structure Data
+    # --------------------------------------------------------------------------------
+    
+    # Load language data to get correct keys for cultural, platform, and purpose components
+    lang_data = load_data(lang_code)
+    
+    data = {
+        # Metadata
+        "TIMESTAMP": time.strftime("%Y-%m-%d %H:%M:%S"),
+        "ENQUETEUR": enqueteur,
+        "LANG_CODE": lang_code,
         
-    grp = df.groupby("partner_domain").agg(
-        partner_name=pd.NamedAgg(column="partner_name", aggfunc=lambda x: x.iloc[0]),
-        partner_type=pd.NamedAgg(column="partner_type", aggfunc=lambda x: x.iloc[0]),
-        partner_commercial=pd.NamedAgg(column="partner_commercial", aggfunc=lambda x: x.iloc[0]), 
-        num_cookies=pd.NamedAgg(column="cookie_name", aggfunc="count"),
-        num_pages=pd.NamedAgg(column="page_domain", aggfunc=lambda x: x.nunique()),
-    ).reset_index()
-    
-    commercial_partners = grp[
-        (grp['partner_commercial'] == 'Yes') | 
-        (grp['partner_type'] == 'Advertising') | 
-        (grp['partner_type'] == 'Tracking')
-    ]
-    return commercial_partners.sort_values(by='num_cookies', ascending=False)
-
-def classify_cookies(cookie_list, page_urls):
-    records = []
-    third_party = []
-    base_domains = [urlparse(u).netloc.lower() for u in page_urls]
-
-    for c in cookie_list:
-        cookie_name = c.get("name")
-        cookie_value = c.get("value", "")
-        cookie_domain = c.get("domain", "").lstrip(".").lower()
-        page_url = c.get("page_url", "N/A")
+        # TAB 1 - Contact
+        "APPROACH": approach,
+        "REFUSAL_REASON": refusal_reason,
+        "REFUSAL_OTHER": refusal_other,
+        "CONTACT_LATER": contact_later,
+        "FIRSTNAME": firstname,
+
+        # TAB 1 - Lieux
+        "SPACE1": space1, "LIEU_1": lieu_1, "LIEU_2": lieu_2, "LIEU_3": lieu_3, 
+        "SPACE2B": "; ".join(space2b), 
+        "SPACE3_1": space3_1, "SPACE3_2": space3_2, "SPACE3_3": space3_3, 
+        "SPACE4": space4, "SPACE5": space5, "SPACE6": space6, 
+        "SPACE7_1": space7_1, "SPACE7_2": space7_2, "SPACE7_3": space7_3, 
+        "SPACE8": space8, "SPACE9": space9, 
+        "SPACE10": "; ".join(space10), 
+
+        # TAB 2 - Engagement
+        "INFO1": "; ".join(info1),
+        "ENGAGEMENT_ORGANISATION": "; ".join(engagement_organisation),
+        "ENGAGEMENT_DOMAINE": "; ".join(engagement_domaine),
+        "INFO_ACTIVITES": "; ".join(info_activites),
+        "INFO_RESEAUX_SOCIAUX": info_reseaux_sociaux,
         
-        # Crucial fix for KeyError: Extract page_domain
-        page_domain = urlparse(page_url).netloc.lower()
+        # TAB 3 - News
+        "INFO_FREQUENCE_ACTU": info_frequence_actu,
+        "INFO2": "; ".join(info2), "INFO3": "; ".join(info3),
+        "INFO_OPPOSITE_FEELING": info_opposite_feeling,
+        "INFO_CONTRADICT_OPINION": info_contradict_opinion,
+        "INFO4": "; ".join(info4), "INFO5": "; ".join(info5),
+        "INFO6_LOGEMENT": info6_logement, "INFO6_POLITIQUE": info6_politique, "INFO6_ETUDES": info6_etudes,
+        "INFO6_CLIMAT": info6_climat, "INFO6_SOCIALES": info6_sociales, "INFO6_SENTIMENTALE": info6_sentimentale,
+        "INFO6_SECURITE": info6_securite, "INFO6_ESTIME": info6_estime, "INFO6_LIBERTE": info6_liberte,
+
+        # TAB 4 - Culturel
+        "PRAT_CULT1": prat_cult1,
+        "PRAT_NATURE": prat_nature,
+        "PURPOSE_AUTRE_DETAIL": purpose_autre_detail,
         
-        partner_name, partner_type, partner_commercial = lookup_partner(cookie_domain)
-        purpose = detect_cookie_purpose(cookie_name)
+        # TAB 5 - Démographie
+        "DEMO_GENDER": demo_gender, "DEMO_AGE": demo_age,
+        "DEMO_LOCATION_COMMUNE": demo_location_commune, "DEMO_LOCATION_ARROND": demo_location_arrond,
+        "DEMO_PARENTS_LOCATION": demo_parents_location, "DEMO_INSCRIPTION": demo_inscription,
+        "DEMO_DISCIPLINE": demo_discipline, "DEMO_JOB": demo_job, "DEMO_INCOME": demo_income,
+        "DEMO_SOCIALCAPITAL1_PARENT1": demo_socialcapital1_parent1,
+        "DEMO_SOCIALCAPITAL1_PARENT2": demo_socialcapital1_parent2,
+        "DEMO_SOCIALCAPITAL2": demo_socialcapital2,
         
-        is_first = any(urlparse(base).netloc.lower() in cookie_domain for base in base_domains)
+        # TAB 7 - Open Questions
+        "OPEN_NON_INSTITUTIONNEL": open_non_institutionnel,
+        "OPEN_ALTERNATIVES": open_alternatives,
+        "OPEN_MOTIVATIONS": open_motivations,
+    }
+
+    # --- Process Cultural Practices (28 values) ---
+    cultural_map = {item[1]: item[0] for item in lang_data['PRAT_CULT_PRACTICES']}
+    
+    # cultural_practices_values is a tuple of 28 elements: 14 checkboxes, then 14 sliders
+    for i, (label_fr, key) in enumerate(LANGUAGES["FR"]['PRAT_CULT_PRACTICES']):
+        # Checkbox key: "CB_<KEY>"
+        data[f"CB_{key}"] = "Oui" if cultural_practices_values[i] else "Non"
+        # Slider key: "SLIDER_<KEY>"
+        data[f"SLIDER_{key}"] = cultural_practices_values[i + 14]
         
-        rec = {
-            "page_url": page_url,
-            "page_domain": page_domain, # Now correctly included
-            "cookie_name": cookie_name,
-            "cookie_value": cookie_value,
-            "cookie_domain": cookie_domain,
-            "partner_domain": get_registered_domain(cookie_domain),
-            "partner_name": partner_name,
-            "partner_type": partner_type,
-            "partner_commercial": partner_commercial,
-            "type": "first-party" if is_first else "third-party",
-            "purpose": purpose
-        }
-        records.append(rec)
-        if rec["type"] == "third-party":
-            third_party.append(rec)
-            
-    return records, third_party
-
-# ----------------------------------------------------
-# 🟢 SELENIUM IMPLEMENTATION TEMPLATE 🟢
-# ----------------------------------------------------
-
-# --- Fallback/Simulation Function (Used if Selenium setup fails) ---
-def fetch_page_simulated(url):
-    """Returns simulated cookie data for testing when Selenium is unavailable."""
-    simulated_cookies = [
-        {'name': 'session_id', 'value': 'abc', 'domain': urlparse(url).netloc}, 
-        {'name': 'consent', 'value': 'yes', 'domain': urlparse(url).netloc},
-        {'name': 'test_cookie_1', 'value': 'cf_ch', 'domain': '.doubleclick.net', 'expiry': time.time() + 3600}, 
-        {'name': 'test_cookie_2', 'value': 'cf_ch', 'domain': '.doubleclick.net', 'expiry': time.time() + 3600}, 
-        {'name': 'm_id', 'value': 'xyz', 'domain': '.mediarithmics.com', 'expiry': time.time() + 3600},
-        {'name': 'Ad_ID', 'value': 'trt', 'domain': '.trksite.biz', 'expiry': time.time() + 3600},           
-        {'name': '_ga', 'value': '123', 'domain': '.analytics.com', 'expiry': time.time() + 3600},            
-    ]
-    for c in simulated_cookies:
-        c['page_url'] = url
-    return simulated_cookies
-# --- End of Fallback ---
-
-def fetch_page_with_cookie_acceptance(url):
-    """
-    Attempts to use Selenium to navigate, click accept, and retrieve cookies.
-    If Selenium fails (e.g., driver missing), it falls back to simulated data.
-    """
-    print(f"Attempting to fetch {url} and simulate cookie acceptance...")
-    
-    # ------------------------------------------------------
-    # ⚠️ SELENIUM SETUP BLOCK - CONFIGURE THIS LOCALLY ⚠️
-    # ------------------------------------------------------
-    try:
-        # Use ChromeDriverManager for automatic driver management (recommended)
-        # service = Service(ChromeDriverManager().install())
+    # --- Process Platforms (20 values) ---
+    platform_map = {item[1]: item[0] for item in lang_data['PLATFORM_ITEMS']}
+
+    # platform_values is a tuple of 20 elements: 10 checkboxes, then 10 number fields
+    for i, (label_fr, key) in enumerate(LANGUAGES["FR"]['PLATFORM_ITEMS']):
+        # Checkbox key: "PLATFORM_CB_<KEY>"
+        data[f"PLATFORM_CB_{key}"] = "Oui" if platform_values[i] else "Non"
+        # Number field key: "PLATFORM_HRS_<KEY>"
+        data[f"PLATFORM_HRS_{key}"] = platform_values[i + 10]
+
+    # --- Process Purpose (13 values) ---
+    purpose_map = {item[1]: item[0] for item in lang_data['PURPOSE_ITEMS']}
+    
+    # purpose_values is a tuple of 13 elements: 8 checkboxes, then 5 sliders 
+    slider_indices_fr = [i for i, item in enumerate(LANGUAGES["FR"]['PURPOSE_ITEMS']) if item[2] == True] # Indices in FR PURPOSE_ITEMS that have a slider
+    
+    # Checkboxes (all 8 items)
+    for i, (label_fr, key, _) in enumerate(LANGUAGES["FR"]['PURPOSE_ITEMS']):
+        # Checkbox key: "PURPOSE_CB_<KEY>"
+        data[f"PURPOSE_CB_{key}"] = "Oui" if purpose_values[i] else "Non"
+
+    # Sliders (only 5 items where the third element is True)
+    for i, purpose_index in enumerate(slider_indices_fr):
+        key = LANGUAGES["FR"]['PURPOSE_ITEMS'][purpose_index][1]
+        # Slider value comes after all 8 checkboxes, so index is 8 + i
+        data[f"PURPOSE_SLIDER_{key}"] = purpose_values[8 + i]
+
+
+    # --------------------------------------------------------------------------------
+    # 3. Save Data to CSV
+    # --------------------------------------------------------------------------------
+    
+    # Ensure CSV file exists with header
+    if not os.path.exists(CSV_FILENAME):
+        df = pd.DataFrame([data])
+        df.to_csv(CSV_FILENAME, index=False, sep=";")
+    else:
+        # Append data to existing CSV
+        df = pd.DataFrame([data])
+        df.to_csv(CSV_FILENAME, mode='a', header=False, index=False, sep=";")
+    
+    # --------------------------------------------------------------------------------
+    # 4. Return Output
+    # --------------------------------------------------------------------------------
+    
+    current_lang_data = load_data(lang_code)
+    success_msg = current_lang_data.get("SUBMISSION_MESSAGE_SUCCESS", "✅ Success! File submitted.")
+    success_msg = success_msg.format(firstname=firstname)
+    
+    return [success_msg, gr.File(value=CSV_FILENAME, label=current_lang_data.get("DOWNLOAD_LABEL", "Download Data"))]
+
+
+# --- Language Update Function ---
+
+def update_language(lang_code: str) -> List[gr.update]:
+    """Updates all components with labels and choices for the selected language."""
+    
+    lang_data = load_data(lang_code)
+    
+    # Determine which themes and frequencies to use based on language
+    themes_actu = THEMES_ACTU_PT if lang_code == "PT" else THEMES_ACTU_FR
+    freq_choices = FREQ_PT if lang_code == "PT" else FREQ_FR
+    
+    # Create lists of updates for dynamic components
+    updates = []
+    
+    # TAB 4 Components that are generated dynamically
+    
+    # 28 Cultural Practice updates (14 Checkboxes + 14 Sliders)
+    for label_fr, key in LANGUAGES["FR"]['PRAT_CULT_PRACTICES']:
+        # Find the correct label in the target language's data, matching by key
+        label_in_target_lang = next((item[0] for item in lang_data['PRAT_CULT_PRACTICES'] if item[1] == key), label_fr)
         
-        # --- OR ---
-        # If running in a restricted environment, you must manually specify the service path:
-        # service = Service('/usr/local/bin/chromedriver') 
-
-        options = Options()
-        options.add_argument('--headless')
-        options.add_argument('--no-sandbox')
-        options.add_argument('--disable-dev-shm-usage')
+        # Checkbox update: label is the cultural practice name
+        updates.append(gr.Checkbox.update(label=label_in_target_lang))
+        # Slider update: label is cultural practice name
+        updates.append(gr.Slider.update(label=label_in_target_lang))
+
+    # 20 Platform Usage updates (10 Checkboxes + 10 Number fields)
+    for label_fr, key in LANGUAGES["FR"]['PLATFORM_ITEMS']:
+        # Find the correct label in the target language's data, matching by key
+        label_in_target_lang = next((item[0] for item in lang_data['PLATFORM_ITEMS'] if item[1] == key), label_fr)
+
+        # Checkbox update: label is the platform name
+        updates.append(gr.Checkbox.update(label=label_in_target_lang))
+        # Number field update: label is platform name
+        updates.append(gr.Number.update(label=label_in_target_lang))
         
-        # driver = webdriver.Chrome(service=service, options=options)
+    # 13 Purpose Usage updates (8 Checkboxes + 5 Sliders)
+    # Checkboxes (all 8)
+    for label_fr, key, has_slider in LANGUAGES["FR"]['PURPOSE_ITEMS']:
+        # Find the correct label in the target language's data, matching by key
+        label_in_target_lang = next((item[0] for item in lang_data['PURPOSE_ITEMS'] if item[1] == key), label_fr)
         
-        # ------------------------------------------------------
-        # ⚠️ Replace the 'return fetch_page_simulated(url)' below
-        # with the actual driver logic (like fetch_page_selenium(driver, url))
-        # once your environment is configured.
-        # ------------------------------------------------------
+        updates.append(gr.Checkbox.update(label=label_in_target_lang))
         
-        return fetch_page_simulated(url)
+    # Sliders (only 5 where has_slider is True)
+    for label_fr, key, has_slider in LANGUAGES["FR"]['PURPOSE_ITEMS']:
+        if has_slider:
+            # Find the correct label in the target language's data, matching by key
+            label_in_target_lang = next((item[0] for item in lang_data['PURPOSE_ITEMS'] if item[1] == key), label_fr)
+            updates.append(gr.Slider.update(label=label_in_target_lang))
+
+    # All other static components (order is crucial, must match the interface definition)
+    static_updates = [
+        # GLOBAL (1)
+        gr.Markdown.update(value=f"# {lang_data['TITLE']}"),
+        # TABS (6)
+        gr.Tab.update(label=lang_data['TAB_1_TITLE']), gr.Tab.update(label=lang_data['TAB_2_TITLE']), gr.Tab.update(label=lang_data['TAB_3_TITLE']),
+        gr.Tab.update(label=lang_data['TAB_4_TITLE']), gr.Tab.update(label=lang_data['TAB_5_TITLE']), gr.Tab.update(label=lang_data['TAB_6_TITLE']),
         
-    except Exception as e:
-        print(f"Selenium setup failed ({e}). Falling back to simulated data.")
-        return fetch_page_simulated(url)
-
-# -------------------------
-# Main crawler logic
-# -------------------------
-def run_crawler(url_input):
-    url_list = [u.strip() for u in url_input.strip().splitlines() if u.strip()]
-    if not url_list:
-        return "Please provide at least one URL.", *(None,)*4
-
-    all_cookies = []
-    
-    for u in url_list:
-        cookies = fetch_page_with_cookie_acceptance(u)
-        all_cookies.extend(cookies)
-    
-    # 1. Classify and summarize
-    cookie_records, third_party_records = classify_cookies(all_cookies, url_list)
-    df_cookies = pd.DataFrame(cookie_records)
-    df_partner = build_partner_summary(cookie_records)
-    
-    # 2. Prepare outputs
-    if df_partner.empty:
-        head_table = "No commercial partners detected via cookies after simulated acceptance."
-        message = "No commercial partners detected via cookies after simulated acceptance."
-    else:
-        df_head = df_partner.head(10).rename(columns={
-            'partner_domain': 'Domain',
-            'partner_name': 'Name',
-            'partner_type': 'Type',
-            'partner_commercial': 'Commercial?',
-            'num_cookies': 'Cookie Count',
-            'num_pages': 'Pages Found'
-        })
-        # This line requires the 'tabulate' package
-        head_table = df_head.to_markdown(index=False, floatfmt=".0f")
-        message = f"Successfully detected {len(df_partner)} potential commercial partners. Details below."
-
-    # 3. Save results to temporary files
-    cookies_path = os.path.join(tempfile.gettempdir(), "cookies_detailed_post_acceptance.xlsx")
-    df_cookies.to_excel(cookies_path, index=False)
-
-    df_tp = pd.DataFrame(third_party_records)
-    third_party_path = os.path.join(tempfile.gettempdir(), "third_party_cookies_post_acceptance.xlsx")
-    df_tp.to_excel(third_party_path, index=False)
-
-    partner_summary_path = os.path.join(tempfile.gettempdir(), "partner_summary_commercial_only.xlsx")
-    df_partner.to_excel(partner_summary_path, index=False)
-
-    return (
-        message,
-        head_table, 
-        cookies_path,
-        third_party_path,
-        partner_summary_path
+        # TAB 1 - Contact (7)
+        gr.Markdown.update(value=f"## {lang_data['SECTION_CONTACT']}"),
+        gr.Radio.update(label=lang_data['APPROACH_LABEL'], choices=[lang_data['CHOICE_OUI'], lang_data['CHOICE_NON']]), 
+        gr.Dropdown.update(label=lang_data['REFUSAL_REASON_LABEL'], choices=lang_data['CHOICES_REFUSAL']),
+        gr.Textbox.update(label=lang_data['REFUSAL_OTHER_LABEL']), 
+        gr.Radio.update(label=lang_data['CONTACT_LATER_LABEL'], choices=[lang_data['CHOICE_OUI'], lang_data['CHOICE_NON']]),
+        gr.Textbox.update(label=lang_data['FIRSTNAME_LABEL'], placeholder=lang_data['FIRSTNAME_PLACEHOLDER']), 
+        gr.Markdown.update(value=f"### 1. {lang_data['SECTION_1_TITLE']}"), 
+
+        # TAB 1 - Lieux (18)
+        gr.Radio.update(label=lang_data['SPACE1_LABEL'], choices=freq_choices),
+        gr.Markdown.update(value=lang_data['SPACE2A_TITLE']), 
+        gr.Textbox.update(label=lang_data['LIEU_1_LABEL'], placeholder=lang_data['LIEU_PLACEHOLDER']), 
+        gr.Textbox.update(label=lang_data['LIEU_2_LABEL'], placeholder=lang_data['LIEU_PLACEHOLDER']), 
+        gr.Textbox.update(label=lang_data['LIEU_3_LABEL'], placeholder=lang_data['LIEU_PLACEHOLDER']), 
+        gr.CheckboxGroup.update(label=lang_data['SPACE2B_LABEL'], choices=lang_data['SPACE2B_CHOICES']),
+        gr.Markdown.update(value=lang_data['SPACE3_TITLE']), 
+        gr.Slider.update(label=lang_data['SPACE3_1_LABEL']), 
+        gr.Slider.update(label=lang_data['SPACE3_2_LABEL']), 
+        gr.Slider.update(label=lang_data['SPACE3_3_LABEL']), 
+        gr.Radio.update(label=lang_data['SPACE4_LABEL'], choices=lang_data['SPACE4_CHOICES']),
+        gr.Textbox.update(label=lang_data['SPACE5_LABEL'], placeholder=lang_data['SPACE5_PLACEHOLDER']),
+        gr.Radio.update(label=lang_data['SPACE6_LABEL'], choices=[lang_data['CHOICE_OUI'], lang_data['CHOICE_NON']]),
+        gr.Textbox.update(label=lang_data['SPACE7_1_LABEL']), 
+        gr.Textbox.update(label=lang_data['SPACE7_2_LABEL']), 
+        gr.Textbox.update(label=lang_data['SPACE7_3_LABEL']), 
+        gr.Textbox.update(label=lang_data['SPACE8_LABEL']),
+        gr.Dropdown.update(label=lang_data['SPACE9_LABEL'], choices=lang_data['SPACE9_CHOICES']), 
+        gr.CheckboxGroup.update(label=lang_data['SPACE10_LABEL'], choices=lang_data['SPACE10_CHOICES']),
+
+        # TAB 2 - Engagement (5)
+        gr.Markdown.update(value=lang_data['SECTION_2_TITLE']), 
+        gr.CheckboxGroup.update(label=lang_data['INFO1_LABEL'], choices=lang_data['INFO1_CHOICES']),
+        gr.CheckboxGroup.update(label=lang_data['ENGAGEMENT_ORGANISATION_LABEL'], choices=lang_data['ENGAGEMENT_ORGANISATION_CHOICES']),
+        gr.CheckboxGroup.update(label=lang_data['ENGAGEMENT_DOMAINE_LABEL'], choices=lang_data['ENGAGEMENT_DOMAINE_CHOICES']),
+        gr.CheckboxGroup.update(label=lang_data['INFO_ACTIVITES_LABEL'], choices=lang_data['INFO_ACTIVITES_CHOICES']),
+        gr.Radio.update(label=lang_data['INFO_RESEAUX_SOCIAUX_LABEL'], choices=lang_data['INFO_RESEAUX_SOCIAUX_CHOICES']),
+
+        # TAB 3 - News (17)
+        gr.Markdown.update(value=lang_data['SECTION_3_TITLE']), 
+        gr.Radio.update(label=lang_data['INFO_FREQUENCE_ACTU_LABEL'], choices=freq_choices),
+        gr.CheckboxGroup.update(label=lang_data['INFO2_LABEL'], choices=themes_actu),
+        gr.CheckboxGroup.update(label=lang_data['INFO3_LABEL'], choices=themes_actu),
+        gr.Radio.update(label=lang_data['OPPOSITE_FEELING_BASE_LABEL'], choices=lang_data['OPPOSITE_FEELING_CHOICES']),
+        gr.Radio.update(label=lang_data['INFO_CONTRADICT_OPINION_LABEL'], choices=lang_data['INFO_CONTRADICT_OPINION_CHOICES']),
+        gr.CheckboxGroup.update(label=lang_data['INFO4_LABEL'], choices=themes_actu),
+        gr.CheckboxGroup.update(label=lang_data['INFO5_LABEL'], choices=lang_data['INFO5_CHOICES']),
+        gr.Markdown.update(value=lang_data['INFO6_TITLE']),
+        gr.Slider.update(label=lang_data['INFO6_LOGEMENT_LABEL']), 
+        gr.Slider.update(label=lang_data['INFO6_POLITIQUE_LABEL']),
+        gr.Slider.update(label=lang_data['INFO6_ETUDES_LABEL']), 
+        gr.Slider.update(label=lang_data['INFO6_CLIMAT_LABEL']),
+        gr.Slider.update(label=lang_data['INFO6_SOCIALES_LABEL']), 
+        gr.Slider.update(label=lang_data['INFO6_SENTIMENTALE_LABEL']),
+        gr.Slider.update(label=lang_data['INFO6_SECURITE_LABEL']), 
+        gr.Slider.update(label=lang_data['INFO6_ESTIME_LABEL']), 
+        gr.Slider.update(label=lang_data['INFO6_LIBERTE_LABEL']), 
+
+        # TAB 4 - Culturel (5)
+        gr.Markdown.update(value=lang_data['SECTION_4_TITLE']),
+        gr.Radio.update(label=lang_data['PRAT_CULT1_LABEL'], choices=lang_data['PRAT_CULT1_CHOICES']),
+        gr.Markdown.update(value=lang_data['PRAT_PREF_TITLE']),
+        gr.Radio.update(label=lang_data['PRAT_NATURE_LABEL'], choices=lang_data['PRAT_NATURE_CHOICES']),
+        gr.Markdown.update(value=lang_data['PLATFORM_TITLE']),
+        gr.Markdown.update(value=lang_data['PURPOSE_TITLE']),
+        gr.Textbox.update(label=lang_data['PURPOSE_AUTRE_DETAIL_LABEL'], placeholder=lang_data['PURPOSE_AUTRE_DETAIL_PLACEHOLDER']),
+
+        # TAB 5 - Démographie (14)
+        gr.Markdown.update(value=lang_data['SECTION_5_TITLE']), 
+        gr.Radio.update(label=lang_data['DEMO_GENDER_LABEL'], choices=lang_data['DEMO_GENDER_CHOICES']),
+        gr.Number.update(label=lang_data['DEMO_AGE_LABEL'], placeholder=lang_data['DEMO_AGE_PLACEHOLDER']),
+        gr.Textbox.update(label=lang_data['DEMO_LOCATION_COMMUNE_LABEL'], placeholder=lang_data['DEMO_LOCATION_COMMUNE_PLACEHOLDER']), 
+        gr.Textbox.update(label=lang_data['DEMO_LOCATION_ARROND_LABEL'], placeholder=lang_data['DEMO_LOCATION_ARROND_PLACEHOLDER']),
+        gr.Radio.update(label=lang_data['DEMO_PARENTS_LOCATION_LABEL'], choices=lang_data['DEMO_PARENTS_LOCATION_CHOICES']),
+        gr.Dropdown.update(label=lang_data['DEMO_INSCRIPTION_LABEL'], choices=lang_data['DEMO_INSCRIPTION_CHOICES']),
+        gr.Dropdown.update(label=lang_data['DEMO_DISCIPLINE_LABEL'], choices=lang_data['DEMO_DISCIPLINE_CHOICES']),
+        gr.Radio.update(label=lang_data['DEMO_JOB_LABEL'], choices=lang_data['DEMO_JOB_CHOICES']),
+        gr.Dropdown.update(label=lang_data['DEMO_INCOME_LABEL'], choices=lang_data['DEMO_INCOME_CHOICES']),
+        gr.Dropdown.update(label=lang_data['DEMO_SOCIALCAPITAL1_PARENT1_LABEL'], choices=lang_data['PCS_CHOICES']),
+        gr.Dropdown.update(label=lang_data['DEMO_SOCIALCAPITAL1_PARENT2_LABEL'], choices=lang_data['PCS_CHOICES']),
+        gr.Number.update(label=lang_data['DEMO_SOCIALCAPITAL2_LABEL']), 
+
+        # TAB 7 - Open Questions (4)
+        gr.Markdown.update(value=lang_data['SECTION_7_TITLE']),
+        gr.Textbox.update(label=lang_data['OPEN_NON_INSTITUTIONNEL_LABEL']),
+        gr.Textbox.update(label=lang_data['OPEN_ALTERNATIVES_LABEL']),
+        gr.Textbox.update(label=lang_data['OPEN_MOTIVATIONS_LABEL']),
+        
+        # SUBMISSION (3)
+        gr.Markdown.update(value=lang_data['SUBMISSION_TITLE']),
+        gr.Button.update(value=lang_data['SUBMIT_BUTTON_LABEL']),
+        gr.Markdown.update(value=lang_data['SUBMISSION_MESSAGE_INIT']),
+        # Note: gr.File does not support update() for label/value in this context, but its label is handled by submit_survey's gr.File update.
+    ]
+    
+    return static_updates + updates
+
+
+# =================================================================
+# GRADIO INTERFACE CONSTRUCTION
+# =================================================================
+
+# Load initial French data for component creation
+default_lang_data = load_data("FR")
+
+# Determine which themes and frequencies to use (initial load)
+themes_actu_fr = THEMES_ACTU_FR
+freq_choices_fr = FREQ_FR
+
+# --- Component Lists for Dynamic Sections ---
+
+# TAB 4 - Cultural Practices (28 components: 14 Checkboxes + 14 Sliders)
+prat_cult_freq_components = []
+for label, key in default_lang_data['PRAT_CULT_PRACTICES']:
+    # Checkbox
+    prat_cult_freq_components.append(gr.Checkbox(label=label, value=False, interactive=True, elem_id=f"CB_{key}"))
+    # Slider
+    prat_cult_freq_components.append(gr.Slider(
+        label=label, 
+        minimum=1, maximum=10, step=1, value=5, interactive=True, elem_id=f"SLIDER_{key}"
+    ))
+
+# TAB 4 - Platform Usage (20 components: 10 Checkboxes + 10 Number fields)
+platform_components = []
+for label, key in default_lang_data['PLATFORM_ITEMS']:
+    # Checkbox
+    platform_components.append(gr.Checkbox(label=label, value=False, interactive=True, elem_id=f"PLATFORM_CB_{key}"))
+    # Number field
+    platform_components.append(gr.Number(
+        label=label, 
+        minimum=0, maximum=24, step=0.5, value=0, interactive=True, elem_id=f"PLATFORM_HRS_{key}"
+    ))
+
+# TAB 4 - Purpose Usage (13 components: 8 Checkboxes + 5 Sliders)
+purpose_components = []
+slider_indices = [i for i, item in enumerate(default_lang_data['PURPOSE_ITEMS']) if item[2] == True] # Indices in PURPOSE_ITEMS that have a slider
+
+# 8 Checkboxes
+for label, key, _ in default_lang_data['PURPOSE_ITEMS']:
+    purpose_components.append(gr.Checkbox(label=label, value=False, interactive=True, elem_id=f"PURPOSE_CB_{key}"))
+
+# 5 Sliders (only for items marked with True)
+for purpose_index in slider_indices:
+    label, key, _ = default_lang_data['PURPOSE_ITEMS'][purpose_index]
+    purpose_components.append(gr.Slider(
+        label=label, 
+        minimum=1, maximum=10, step=1, value=5, interactive=True, elem_id=f"PURPOSE_SLIDER_{key}"
+    ))
+    
+# --- The Gradio Interface ---
+
+with gr.Blocks(title=default_lang_data['TITLE']) as demo:
+    
+    # --- Language Selector (always visible) ---
+    language_selector = gr.Dropdown(
+        label=default_lang_data['LANG_LABEL'], 
+        choices=["FR", "PT"], 
+        value="FR", 
+        interactive=True,
+        scale=1
+    )
+    
+    gr.Markdown(f"# {default_lang_data['TITLE']}", elem_id="main_title")
+    gr.Markdown(default_lang_data['INTRO_TEXT'])
+    
+    with gr.Tabs():
+        
+        # =================================================================
+        # TAB 1: Contact & Lieux Fréquentés
+        # =================================================================
+        with gr.Tab(label=default_lang_data['TAB_1_TITLE']):
+            
+            # --- Contact / Refus ---
+            markdown_contact = gr.Markdown(f"## {default_lang_data['SECTION_CONTACT']}")
+            
+            with gr.Row():
+                enqueteur = gr.Textbox(label=default_lang_data['ENQUETEUR_LABEL'], value="ENQ1", interactive=True)
+                approach = gr.Radio(label=default_lang_data['APPROACH_LABEL'], choices=[default_lang_data['CHOICE_OUI'], default_lang_data['CHOICE_NON']], value=default_lang_data['CHOICE_NON'])
+            
+            with gr.Column(visible=True) as refusal_details:
+                refusal_reason = gr.Dropdown(label=default_lang_data['REFUSAL_REASON_LABEL'], choices=default_lang_data['CHOICES_REFUSAL'])
+                refusal_other = gr.Textbox(label=default_lang_data['REFUSAL_OTHER_LABEL'], placeholder="Autre raison", interactive=True)
+                contact_later = gr.Radio(label=default_lang_data['CONTACT_LATER_LABEL'], choices=[default_lang_data['CHOICE_OUI'], default_lang_data['CHOICE_NON']], value=default_lang_data['CHOICE_NON'])
+                gr.Markdown(default_lang_data['REFUSAL_INFO'])
+            
+            with gr.Column(visible=False) as contact_details:
+                firstname = gr.Textbox(label=default_lang_data['FIRSTNAME_LABEL'], placeholder=default_lang_data['FIRSTNAME_PLACEHOLDER'], interactive=True)
+                gr.Markdown(default_lang_data['FIRSTNAME_INFO'])
+
+            # Logic to show/hide details based on APPROACH
+            approach.change(
+                fn=lambda x: [gr.Column.update(visible=x == default_lang_data['CHOICE_NON']), gr.Column.update(visible=x == default_lang_data['CHOICE_OUI'])],
+                inputs=[approach],
+                outputs=[refusal_details, contact_details]
+            )
+            
+            # --- Lieux Fréquentés ---
+            markdown_section_1 = gr.Markdown(f"### 1. {default_lang_data['SECTION_1_TITLE']}")
+            space1 = gr.Radio(label=default_lang_data['SPACE1_LABEL'], choices=freq_choices_fr, value="Jamais")
+            
+            markdown_space2a = gr.Markdown(default_lang_data['SPACE2A_TITLE'])
+            with gr.Row():
+                lieu_1 = gr.Textbox(label=default_lang_data['LIEU_1_LABEL'], placeholder=default_lang_data['LIEU_PLACEHOLDER'])
+                lieu_2 = gr.Textbox(label=default_lang_data['LIEU_2_LABEL'], placeholder=default_lang_data['LIEU_PLACEHOLDER'])
+                lieu_3 = gr.Textbox(label=default_lang_data['LIEU_3_LABEL'], placeholder=default_lang_data['LIEU_PLACEHOLDER'])
+
+            space2b = gr.CheckboxGroup(label=default_lang_data['SPACE2B_LABEL'], choices=default_lang_data['SPACE2B_CHOICES'], visible=True)
+            
+            markdown_space3 = gr.Markdown(default_lang_data['SPACE3_TITLE'])
+            space3_1 = gr.Slider(label=default_lang_data['SPACE3_1_LABEL'], minimum=1, maximum=10, step=1, value=5)
+            space3_2 = gr.Slider(label=default_lang_data['SPACE3_2_LABEL'], minimum=1, maximum=10, step=1, value=5)
+            space3_3 = gr.Slider(label=default_lang_data['SPACE3_3_LABEL'], minimum=1, maximum=10, step=1, value=5)
+
+            space4 = gr.Radio(label=default_lang_data['SPACE4_LABEL'], choices=default_lang_data['SPACE4_CHOICES'], value="non")
+            space5 = gr.Textbox(label=default_lang_data['SPACE5_LABEL'], placeholder=default_lang_data['SPACE5_PLACEHOLDER'], visible=False)
+            
+            space4.change(
+                fn=lambda x: gr.Textbox.update(visible=x == "oui"),
+                inputs=[space4],
+                outputs=[space5]
+            )
+
+            space6 = gr.Radio(label=default_lang_data['SPACE6_LABEL'], choices=[default_lang_data['CHOICE_OUI'], default_lang_data['CHOICE_NON']], value=default_lang_data['CHOICE_NON'])
+            gr.Markdown(default_lang_data['SPACE6_INFO'])
+
+            # Engagement Practices in these spaces
+            with gr.Column(visible=False) as engagement_practices:
+                with gr.Row():
+                    space7_1 = gr.Textbox(label=default_lang_data['SPACE7_1_LABEL'])
+                    space7_2 = gr.Textbox(label=default_lang_data['SPACE7_2_LABEL'])
+                    space7_3 = gr.Textbox(label=default_lang_data['SPACE7_3_LABEL'])
+                
+                space8 = gr.Textbox(label=default_lang_data['SPACE8_LABEL'])
+                space9 = gr.Dropdown(label=default_lang_data['SPACE9_LABEL'], choices=default_lang_data['SPACE9_CHOICES'])
+                space10 = gr.CheckboxGroup(label=default_lang_data['SPACE10_LABEL'], choices=default_lang_data['SPACE10_CHOICES'])
+            
+            space6.change(
+                fn=lambda x: gr.Column.update(visible=x == default_lang_data['CHOICE_OUI']),
+                inputs=[space6],
+                outputs=[engagement_practices]
+            )
+
+        # =================================================================
+        # TAB 2: Engagement Citoyen
+        # =================================================================
+        with gr.Tab(label=default_lang_data['TAB_2_TITLE']):
+            markdown_section_2 = gr.Markdown(default_lang_data['SECTION_2_TITLE'])
+            info1 = gr.CheckboxGroup(label=default_lang_data['INFO1_LABEL'], choices=default_lang_data['INFO1_CHOICES'])
+            
+            # Additional details only visible if engaged
+            with gr.Column() as engagement_details:
+                engagement_organisation = gr.CheckboxGroup(label=default_lang_data['ENGAGEMENT_ORGANISATION_LABEL'], choices=default_lang_data['ENGAGEMENT_ORGANISATION_CHOICES'])
+                engagement_domaine = gr.CheckboxGroup(label=default_lang_data['ENGAGEMENT_DOMAINE_LABEL'], choices=default_lang_data['ENGAGEMENT_DOMAINE_CHOICES'])
+            
+            info_activites = gr.CheckboxGroup(label=default_lang_data['INFO_ACTIVITES_LABEL'], choices=default_lang_data['INFO_ACTIVITES_CHOICES'])
+            info_reseaux_sociaux = gr.Radio(label=default_lang_data['INFO_RESEAUX_SOCIAUX_LABEL'], choices=default_lang_data['INFO_RESEAUX_SOCIAUX_CHOICES'])
+            
+        # =================================================================
+        # TAB 3: Consommation d'actualités
+        # =================================================================
+        with gr.Tab(label=default_lang_data['TAB_3_TITLE']):
+            markdown_section_3 = gr.Markdown(default_lang_data['SECTION_3_TITLE'])
+            info_frequence_actu = gr.Radio(label=default_lang_data['INFO_FREQUENCE_ACTU_LABEL'], choices=freq_choices_fr, value="Parfois")
+            
+            with gr.Row():
+                info2 = gr.CheckboxGroup(label=default_lang_data['INFO2_LABEL'], choices=themes_actu_fr, max_selectable=3)
+                info3 = gr.CheckboxGroup(label=default_lang_data['INFO3_LABEL'], choices=themes_actu_fr, max_selectable=3)
+                
+            with gr.Row():
+                info_opposite_feeling = gr.Radio(label=default_lang_data['OPPOSITE_FEELING_BASE_LABEL'], choices=default_lang_data['OPPOSITE_FEELING_CHOICES'], value="Oui, parfois")
+                info_contradict_opinion = gr.Radio(label=default_lang_data['INFO_CONTRADICT_OPINION_LABEL'], choices=default_lang_data['INFO_CONTRADICT_OPINION_CHOICES'], value="Non, jamais")
+            
+            info4 = gr.CheckboxGroup(label=default_lang_data['INFO4_LABEL'], choices=themes_actu_fr)
+            info5 = gr.CheckboxGroup(label=default_lang_data['INFO5_LABEL'], choices=default_lang_data['INFO5_CHOICES'])
+            
+            # --- Sentiment de contrôle ---
+            markdown_info6 = gr.Markdown(default_lang_data['INFO6_TITLE'])
+            with gr.Row():
+                info6_logement = gr.Slider(label=default_lang_data['INFO6_LOGEMENT_LABEL'], minimum=1, maximum=10, step=1, value=5)
+                info6_politique = gr.Slider(label=default_lang_data['INFO6_POLITIQUE_LABEL'], minimum=1, maximum=10, step=1, value=5)
+            with gr.Row():
+                info6_etudes = gr.Slider(label=default_lang_data['INFO6_ETUDES_LABEL'], minimum=1, maximum=10, step=1, value=5)
+                info6_climat = gr.Slider(label=default_lang_data['INFO6_CLIMAT_LABEL'], minimum=1, maximum=10, step=1, value=5)
+            with gr.Row():
+                info6_sociales = gr.Slider(label=default_lang_data['INFO6_SOCIALES_LABEL'], minimum=1, maximum=10, step=1, value=5)
+                info6_sentimentale = gr.Slider(label=default_lang_data['INFO6_SENTIMENTALE_LABEL'], minimum=1, maximum=10, step=1, value=5)
+            with gr.Row():
+                info6_securite = gr.Slider(label=default_lang_data['INFO6_SECURITE_LABEL'], minimum=1, maximum=10, step=1, value=5)
+                info6_estime = gr.Slider(label=default_lang_data['INFO6_ESTIME_LABEL'], minimum=1, maximum=10, step=1, value=5)
+                info6_liberte = gr.Slider(label=default_lang_data['INFO6_LIBERTE_LABEL'], minimum=1, maximum=10, step=1, value=5)
+
+
+        # =================================================================
+        # TAB 4: Pratiques culturelles et usages numériques
+        # =================================================================
+        with gr.Tab(label=default_lang_data['TAB_4_TITLE']):
+            markdown_section_4 = gr.Markdown(default_lang_data['SECTION_4_TITLE'])
+            
+            # Importance of Culture
+            prat_cult1 = gr.Radio(label=default_lang_data['PRAT_CULT1_LABEL'], choices=default_lang_data['PRAT_CULT1_CHOICES'], value="important")
+            
+            # Cultural Practices Preferences
+            markdown_prat_pref = gr.Markdown(default_lang_data['PRAT_PREF_TITLE'])
+            gr.Markdown(default_lang_data['PRAT_CULT_FREQ_LABEL'])
+            with gr.Column():
+                for i in range(0, len(prat_cult_freq_components), 2):
+                    with gr.Row():
+                        prat_cult_freq_components[i] # Checkbox
+                        prat_cult_freq_components[i+1] # Slider
+            
+            # Nature Frequency
+            prat_nature = gr.Radio(label=default_lang_data['PRAT_NATURE_LABEL'], choices=default_lang_data['PRAT_NATURE_CHOICES'], value="Parfois")
+            
+            # Platform Usage
+            markdown_platform_title = gr.Markdown(default_lang_data['PLATFORM_TITLE'])
+            gr.Markdown(default_lang_data['PLATFORM_HOURS_LABEL'])
+            with gr.Column():
+                for i in range(0, len(platform_components), 2):
+                    with gr.Row():
+                        platform_components[i] # Checkbox
+                        platform_components[i+1] # Number field
+                        
+            # Purpose Usage
+            markdown_purpose_title = gr.Markdown(default_lang_data['PURPOSE_TITLE'])
+            gr.Markdown(default_lang_data['PURPOSE_FREQ_LABEL'])
+            
+            # Checkboxes (8) and Sliders (5) for Purpose
+            with gr.Column():
+                slider_counter = 0
+                for i in range(len(default_lang_data['PURPOSE_ITEMS'])):
+                    label, key, has_slider = default_lang_data['PURPOSE_ITEMS'][i]
+                    
+                    if has_slider:
+                        with gr.Row():
+                            purpose_components[i] # Checkbox
+                            purpose_components[8 + slider_counter] # Slider
+                        slider_counter += 1
+                    else:
+                        purpose_components[i] # Checkbox only
+                        
+            purpose_autre_detail = gr.Textbox(label=default_lang_data['PURPOSE_AUTRE_DETAIL_LABEL'], placeholder=default_lang_data['PURPOSE_AUTRE_DETAIL_PLACEHOLDER'], interactive=True)
+            
+        # =================================================================
+        # TAB 5: Profil Démographique et Capital Social
+        # =================================================================
+        with gr.Tab(label=default_lang_data['TAB_5_TITLE']):
+            markdown_section_5 = gr.Markdown(default_lang_data['SECTION_5_TITLE'])
+            
+            # Demographics
+            demo_gender = gr.Radio(label=default_lang_data['DEMO_GENDER_LABEL'], choices=default_lang_data['DEMO_GENDER_CHOICES'], value=default_lang_data['DEMO_GENDER_CHOICES'][0])
+            demo_age = gr.Number(label=default_lang_data['DEMO_AGE_LABEL'], placeholder=default_lang_data['DEMO_AGE_PLACEHOLDER'], minimum=15, maximum=99, step=1, value=20)
+            
+            with gr.Row():
+                demo_location_commune = gr.Textbox(label=default_lang_data['DEMO_LOCATION_COMMUNE_LABEL'], placeholder=default_lang_data['DEMO_LOCATION_COMMUNE_PLACEHOLDER'])
+                demo_location_arrond = gr.Textbox(label=default_lang_data['DEMO_LOCATION_ARROND_LABEL'], placeholder=default_lang_data['DEMO_LOCATION_ARROND_PLACEHOLDER'])
+            
+            demo_parents_location = gr.Radio(label=default_lang_data['DEMO_PARENTS_LOCATION_LABEL'], choices=default_lang_data['DEMO_PARENTS_LOCATION_CHOICES'], value=default_lang_data['DEMO_PARENTS_LOCATION_CHOICES'][0])
+            
+            # Education/Job
+            demo_inscription = gr.Dropdown(label=default_lang_data['DEMO_INSCRIPTION_LABEL'], choices=default_lang_data['DEMO_INSCRIPTION_CHOICES'], value=default_lang_data['DEMO_INSCRIPTION_CHOICES'][0])
+            demo_discipline = gr.Dropdown(label=default_lang_data['DEMO_DISCIPLINE_LABEL'], choices=default_lang_data['DEMO_DISCIPLINE_CHOICES'], value=default_lang_data['DEMO_DISCIPLINE_CHOICES'][0])
+            demo_job = gr.Radio(label=default_lang_data['DEMO_JOB_LABEL'], choices=default_lang_data['DEMO_JOB_CHOICES'], value=default_lang_data['DEMO_JOB_CHOICES'][2])
+            demo_income = gr.Dropdown(label=default_lang_data['DEMO_INCOME_LABEL'], choices=default_lang_data['DEMO_INCOME_CHOICES'], value=default_lang_data['DEMO_INCOME_CHOICES'][0])
+            
+            # Social Capital
+            demo_socialcapital1_parent1 = gr.Dropdown(label=default_lang_data['DEMO_SOCIALCAPITAL1_PARENT1_LABEL'], choices=default_lang_data['PCS_CHOICES'])
+            demo_socialcapital1_parent2 = gr.Dropdown(label=default_lang_data['DEMO_SOCIALCAPITAL1_PARENT2_LABEL'], choices=default_lang_data['PCS_CHOICES'])
+            demo_socialcapital2 = gr.Number(label=default_lang_data['DEMO_SOCIALCAPITAL2_LABEL'], minimum=0, maximum=100, step=1, value=5)
+
+            # --- Open Questions (Moved to Tab 5 for simplicity and correct component count) ---
+            markdown_section_7 = gr.Markdown(default_lang_data['SECTION_7_TITLE'])
+            open_non_institutionnel = gr.Textbox(label=default_lang_data['OPEN_NON_INSTITUTIONNEL_LABEL'], lines=3)
+            open_alternatives = gr.Textbox(label=default_lang_data['OPEN_ALTERNATIVES_LABEL'], lines=3)
+            open_motivations = gr.Textbox(label=default_lang_data['OPEN_MOTIVATIONS_LABEL'], lines=3)
+            
+        # =================================================================
+        # TAB 6: Soumission et Téléchargement
+        # =================================================================
+        with gr.Tab(label=default_lang_data['TAB_6_TITLE']):
+            markdown_submission_title = gr.Markdown(default_lang_data['SUBMISSION_TITLE'])
+            submit_button = gr.Button(default_lang_data['SUBMIT_BUTTON_LABEL'])
+            submission_message = gr.Markdown(default_lang_data['SUBMISSION_MESSAGE_INIT'])
+            data_download = gr.File(label=default_lang_data['DOWNLOAD_LABEL'], value=CSV_FILENAME, interactive=False)
+    
+    # --- Collect all input components in the correct order for submission ---
+    input_components = [
+        language_selector, 
+        # TAB 1 - Contact (6)
+        enqueteur, approach, refusal_reason, refusal_other, contact_later, firstname, 
+        # TAB 1 - Lieux (17)
+        space1, lieu_1, lieu_2, lieu_3, space2b, space3_1, space3_2, space3_3, space4, space5, space6, space7_1, space7_2, space7_3, space8, space9, space10,
+        # TAB 2 - Engagement (5)
+        info1, engagement_organisation, engagement_domaine, info_activites, info_reseaux_sociaux,
+        # TAB 3 - News (12)
+        info_frequence_actu, info2, info3, info_opposite_feeling, info_contradict_opinion, info4, info5,
+        info6_logement, info6_politique, info6_etudes, info6_climat, info6_sociales, info6_sentimentale, info6_securite, info6_estime, info6_liberte,
+        # TAB 4 - Culturel (58)
+        prat_cult1,
+        *prat_cult_freq_components, # 28 items
+        prat_nature, 
+        *platform_components, # 20 items
+        purpose_autre_detail,
+        *purpose_components, # 13 items
+        # TAB 5 - Démographie (15)
+        demo_gender, demo_age, demo_location_commune, demo_location_arrond, demo_parents_location, demo_inscription, demo_discipline, demo_job, demo_income,
+        demo_socialcapital1_parent1, demo_socialcapital1_parent2, demo_socialcapital2,
+        open_non_institutionnel, open_alternatives, open_motivations,
+        # Total: 1 (lang_code) + 119 (data inputs) = 120
+    ]
+    
+    # Connect the language selector to the update function
+    # NOTE: The order of outputs here must exactly match the order of updates in update_language
+    language_selector.change(
+        fn=update_language, 
+        inputs=[language_selector], 
+        outputs=[
+            # Static components (82)
+            gr.Markdown.update(elem_id="main_title"), # 1
+            gr.Tab.update(label=default_lang_data['TAB_1_TITLE']), gr.Tab.update(label=default_lang_data['TAB_2_TITLE']), gr.Tab.update(label=default_lang_data['TAB_3_TITLE']),
+            gr.Tab.update(label=default_lang_data['TAB_4_TITLE']), gr.Tab.update(label=default_lang_data['TAB_5_TITLE']), gr.Tab.update(label=default_lang_data['TAB_6_TITLE']), # 6
+            markdown_contact, # 7
+            approach, refusal_reason, refusal_other, contact_later, firstname, markdown_section_1, # 13
+            space1, markdown_space2a, lieu_1, lieu_2, lieu_3, space2b, markdown_space3, space3_1, space3_2, space3_3, space4, space5, space6, space7_1, space7_2, space7_3, space8, space9, space10, # 31
+            markdown_section_2, info1, engagement_organisation, engagement_domaine, info_activites, info_reseaux_sociaux, # 37
+            markdown_section_3, info_frequence_actu, info2, info3, info_opposite_feeling, info_contradict_opinion, info4, info5, markdown_info6, info6_logement, info6_politique, info6_etudes, info6_climat, info6_sociales, info6_sentimentale, info6_securite, info6_estime, info6_liberte, # 55
+            markdown_section_4, prat_cult1, markdown_prat_pref, prat_nature, markdown_platform_title, markdown_purpose_title, purpose_autre_detail, # 62
+            markdown_section_5, demo_gender, demo_age, demo_location_commune, demo_location_arrond, demo_parents_location, demo_inscription, demo_discipline, demo_job, demo_income, demo_socialcapital1_parent1, demo_socialcapital1_parent2, demo_socialcapital2, # 75
+            markdown_section_7, open_non_institutionnel, open_alternatives, open_motivations, # 79
+            markdown_submission_title, submit_button, submission_message, # 82
+            
+            # Dynamic components (28 + 20 + 13 = 61 components)
+            *prat_cult_freq_components, # 28
+            *platform_components, # 20
+            *purpose_components # 13
+        ]
+    )
+
+    # Connect the submit button to the submission function
+    submit_button.click(
+        fn=submit_survey,
+        inputs=input_components,
+        outputs=[submission_message, data_download],
     )
 
-# -------------------------
-# Gradio Interface
-# -------------------------
-iface = gr.Interface(
-    fn=run_crawler,
-    inputs=[
-        gr.Textbox(label="🌐 Paste URL(s) to crawl (one per line)", lines=6, placeholder="https://example.com/"),
-    ],
-    outputs=[
-        gr.Textbox(label="Status"),
-        gr.Markdown(label="📈 Top Commercial Partner Summary (Head of Data)"), 
-        gr.File(label="🍪 XLSX of All Cookies (detailed)"),
-        gr.File(label="📊 XLSX of Third-Party Cookies Only"),
-        gr.File(label="📈 XLSX Commercial Partner Summary (Full)"),
-    ],
-    title="🍪 Commercial Partner Cookie Detector (Post-Acceptance)",
-    description="Loads pages using a simulated browser, clicks 'Accept' on a cookie banner (requires Selenium setup), and lists commercial partners detected via third-party cookies."
-)
-
-iface.launch()
+if __name__ == "__main__":
+    demo.launch()