feat: Add initial application structure, core data models, and a new catalog module with Bonial scraping capabilities.

app/services/bonial_scraper.py

@@ -0,0 +1,400 @@
+"""
+Bonial.fr Scraper Service
+
+Scrapes promotional catalogs from Bonial.fr for configured enseignes.
+Uses Playwright for browser automation.
+"""
+
+import asyncio
+import hashlib
+import logging
+import re
+from datetime import datetime
+from typing import Any
+
+from playwright.async_api import async_playwright, Page, Browser
+from sqlalchemy.orm import Session
+
+from app.models import Enseigne, Catalogue, CataloguePage, ScrapingLog
+
+logger = logging.getLogger(__name__)
+
+# Bonial selectors based on manual analysis
+BONIAL_SELECTORS = {
+    "catalog_card": "section:has(img[src*='content-media.bonial.biz'])",
+    "catalog_title": "div.text-dark.text-lg.font-bold, h3, h2",
+    "catalog_image": "img[src*='content-media.bonial.biz']",
+    "catalog_link": "a:has-text('Ouvrir le catalogue')",
+    "cookie_accept": "button:has-text('OK'), button:has-text('Accepter')",
+}
+
+# Date pattern: "mar. 25/11 - lun. 08/12/2025"
+DATE_PATTERN = r"(\d{2}/\d{2})\s*-\s*(\d{2}/\d{2}/\d{4})"
+
+
+def parse_bonial_dates(date_str: str) -> tuple[datetime, datetime]:
+    """
+    Parse Bonial date format to datetime objects.
+    
+    Args:
+        date_str: Date string like "mar. 25/11 - lun. 08/12/2025"
+    
+    Returns:
+        Tuple of (date_debut, date_fin)
+    
+    Raises:
+        ValueError: If date format is invalid
+    """
+    match = re.search(DATE_PATTERN, date_str)
+    
+    if not match:
+        raise ValueError(f"Invalid date format: {date_str}")
+    
+    debut_str, fin_str = match.groups()
+    
+    # Extract year from end date
+    year = fin_str.split("/")[2]
+    
+    # Add year to start date
+    debut_full = f"{debut_str}/{year}"
+    
+    # Parse dates
+    debut = datetime.strptime(debut_full, "%d/%m/%Y")
+    fin = datetime.strptime(fin_str, "%d/%m/%Y")
+    
+    return debut, fin
+
+
+def compute_catalog_hash(enseigne_id: int, titre: str, date_debut: datetime) -> str:
+    """
+    Compute SHA256 hash for duplicate detection.
+    
+    Args:
+        enseigne_id: ID of the enseigne
+        titre: Catalog title
+        date_debut: Start date
+    
+    Returns:
+        SHA256 hash string
+    """
+    content = f"{enseigne_id}|{titre}|{date_debut.isoformat()}"
+    return hashlib.sha256(content.encode()).hexdigest()
+
+
+async def accept_cookies(page: Page) -> None:
+    """Accept cookie consent banner if present."""
+    try:
+        await page.click(BONIAL_SELECTORS["cookie_accept"], timeout=3000)
+        logger.debug("Accepted cookies")
+        await asyncio.sleep(1)
+    except Exception:
+        logger.debug("No cookie banner found or already accepted")
+
+
+async def scrape_catalog_list(
+    page: Page,
+    enseigne: Enseigne
+) -> list[dict[str, Any]]:
+    """
+    Scrape the list of catalogs for an ens enseigne from Bonial.
+    
+    Args:
+        page: Playwright page object
+        enseigne: Enseigne object
+    
+    Returns:
+        List of catalog data dictionaries
+    """
+    url = f"https://www.bonial.fr/Enseignes/{enseigne.slug_bonial}"
+    logger.info(f"Scraping catalog list for {enseigne.nom} from {url}")
+    
+    await page.goto(url, wait_until="networkidle")
+    await accept_cookies(page)
+    
+    # Wait for catalog cards to load
+    try:
+        await page.wait_for_selector(BONIAL_SELECTORS["catalog_card"], timeout=10000)
+    except Exception as e:
+        logger.error(f"No catalog cards found for {enseigne.nom}: {e}")
+        return []
+    
+    # Extract all catalog cards
+    cards = await page.query_selector_all(BONIAL_SELECTORS["catalog_card"])
+    logger.info(f"Found {len(cards)} catalog cards for {enseigne.nom}")
+    
+    catalogues = []
+    
+    for idx, card in enumerate(cards):
+        try:
+            # Extract title
+            title_el = await card.query_selector(BONIAL_SELECTORS["catalog_title"])
+            titre = await title_el.inner_text() if title_el else None
+            
+            # Extract dates - look for any element containing date pattern
+            dates_text = await card.inner_text()
+            date_match = re.search(DATE_PATTERN, dates_text)
+            
+            # Extract image
+            image_el = await card.query_selector(BONIAL_SELECTORS["catalog_image"])
+            image_url = await image_el.get_attribute("src") if image_el else None
+            
+            # Extract link - try multiple strategies
+            link_el = await card.query_selector(BONIAL_SELECTORS["catalog_link"])
+            if not link_el:
+                # Fallback: click on image
+                link_el = image_el
+            
+            catalogue_url = await link_el.get_attribute("href") if link_el else None
+            
+            # Make URL absolute if relative
+            if catalogue_url and not catalogue_url.startswith("http"):
+                catalogue_url = f"https://www.bonial.fr{catalogue_url}"
+            
+            if all([titre, date_match, image_url, catalogue_url]):
+                # Parse dates
+                try:
+                    date_debut, date_fin = parse_bonial_dates(dates_text)
+                except ValueError as e:
+                    logger.warning(f"Failed to parse dates for catalog {idx}: {e}")
+                    continue
+                
+                catalogues.append({
+                    "titre": titre.strip(),
+                    "date_debut": date_debut,
+                    "date_fin": date_fin,
+                    "image_couverture_url": image_url,
+                    "catalogue_url": catalogue_url,
+                })
+                
+                logger.debug(f"Extracted catalog: {titre}")
+            else:
+                logger.warning(f"Incomplete data for catalog {idx}: title={titre}, dates={bool(date_match)}, image={bool(image_url)}, url={bool(catalogue_url)}")
+        
+        except Exception as e:
+            logger.error(f"Error extracting catalog {idx} for {enseigne.nom}: {e}")
+            continue
+    
+    logger.info(f"Successfully extracted {len(catalogues)} catalogs for {enseigne.nom}")
+    return catalogues
+
+
+async def scrape_catalog_pages(page: Page, catalogue_url: str) -> list[dict[str, Any]]:
+    """
+    Scrape all pages of a catalog from the Bonial viewer.
+    
+    Args:
+        page: Playwright page object
+        catalogue_url: URL of the catalog viewer
+    
+    Returns:
+        List of page data dictionaries
+    """
+    logger.info(f"Scraping catalog pages from {catalogue_url}")
+    
+    await page.goto(catalogue_url, wait_until="networkidle")
+    await asyncio.sleep(2)  # Wait for viewer to initialize
+    
+    # Try to close any popup/modal
+    try:
+        await page.click("button:has-text('Continuer'), button:has-text('Fermer')", timeout=2000)
+        await asyncio.sleep(1)
+    except Exception:
+        pass
+    
+    # Find all images from content-media.bonial.biz
+    images = await page.query_selector_all("img[src*='content-media.bonial.biz']")
+    
+    if not images:
+        logger.warning(f"No images found in catalog viewer: {catalogue_url}")
+        return []
+    
+    # Extract unique image URLs (filter duplicates and thumbnails)
+    seen_urls = set()
+    pages = []
+    
+    for idx, img in enumerate(images):
+        try:
+            src = await img.get_attribute("src")
+            
+            if src and src not in seen_urls:
+                # Filter out blurred/low-res images if possible
+                # Full res images are typically larger
+                width = await img.evaluate("el => el.naturalWidth")
+                height = await img.evaluate("el => el.naturalHeight")
+                
+                # Skip very small images (likely thumbnails)
+                if width < 200 or height < 200:
+                    continue
+                
+                seen_urls.add(src)
+                pages.append({
+                    "numero_page": len(pages) + 1,
+                    "image_url": src,
+                    "largeur": width,
+                    "hauteur": height,
+                })
+        
+        except Exception as e:
+            logger.debug(f"Error processing image {idx}: {e}")
+            continue
+    
+    logger.info(f"Found {len(pages)} pages in catalog")
+    return pages
+
+
+async def scrape_enseigne(
+    enseigne: Enseigne,
+    db: Session,
+    browser: Browser | None = None
+) -> ScrapingLog:
+    """
+    Scrape all catalogs for a specific enseigne.
+    
+    Args:
+        enseigne: Enseigne to scrape
+        db: Database session
+        browser: Optional existing browser instance
+    
+    Returns:
+        ScrapingLog object with execution details
+    """
+    start_time = datetime.now()
+    log = ScrapingLog(
+        enseigne_id=enseigne.id,
+        statut="error",  # Will be updated
+        catalogues_trouves=0,
+        catalogues_nouveaux=0,
+        catalogues_mis_a_jour=0,
+    )
+    
+    own_browser = browser is None
+    
+    try:
+        # Create browser if not provided
+        if own_browser:
+            p = await async_playwright().start()
+            browser = await p.chromium.launch(headless=True)
+        
+        page = await browser.new_page()
+        
+        # Set realistic user agent
+        await page.set_extra_http_headers({
+            "User-Agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/131.0.0.0 Safari/537.36"
+        })
+        
+        # Scrape catalog list
+        catalogues_data = await scrape_catalog_list(page, enseigne)
+        log.catalogues_trouves = len(catalogues_data)
+        
+        if not catalogues_data:
+            log.statut = "success"
+            log.message_erreur = "No catalogs found (may be normal)"
+            await page.close()
+            return log
+        
+        # Process each catalog
+        for cat_data in catalogues_data:
+            try:
+                # Compute hash for duplicate detection
+                content_hash = compute_catalog_hash(
+                    enseigne.id,
+                    cat_data["titre"],
+                    cat_data["date_debut"]
+                )
+                
+                # Check if catalog already exists
+                existing = db.query(Catalogue).filter_by(content_hash=content_hash).first()
+                
+                if existing:
+                    logger.debug(f"Catalog already exists: {cat_data['titre']}")
+                    continue
+                
+                # Scrape catalog pages
+                await asyncio.sleep(3)  # Respectful delay
+                pages_data = await scrape_catalog_pages(page, cat_data["catalogue_url"])
+                
+                # Create new catalog
+                catalogue = Catalogue(
+                    enseigne_id=enseigne.id,
+                    titre=cat_data["titre"],
+                    date_debut=cat_data["date_debut"],
+                    date_fin=cat_data["date_fin"],
+                    image_couverture_url=cat_data["image_couverture_url"],
+                    catalogue_url=cat_data["catalogue_url"],
+                    nombre_pages=len(pages_data),
+                    content_hash=content_hash,
+                    statut="actif" if cat_data["date_fin"] >= datetime.now() else "termine",
+                )
+                
+                db.add(catalogue)
+                db.flush()  # Get catalogue.id
+                
+                # Add pages
+                for page_data in pages_data:
+                    cat_page = CataloguePage(
+                        catalogue_id=catalogue.id,
+                        **page_data
+                    )
+                    db.add(cat_page)
+                
+                log.catalogues_nouveaux += 1
+                logger.info(f"Created catalog: {cat_data['titre']} ({len(pages_data)} pages)")
+            
+            except Exception as e:
+                logger.error(f"Error processing catalog {cat_data.get('titre')}: {e}")
+                continue
+        
+        await page.close()
+        
+        # Update status
+        log.statut = "success" if log.catalogues_nouveaux > 0 else "partial"
+        db.commit()
+    
+    except Exception as e:
+        logger.error(f"Error scraping enseigne {enseigne.nom}: {e}")
+        log.statut = "error"
+        log.message_erreur = str(e)
+        db.rollback()
+    
+    finally:
+        if own_browser and browser:
+            await browser.close()
+        
+        # Record duration
+        log.duree_secondes = (datetime.now() - start_time).total_seconds()
+        db.add(log)
+        db.commit()
+    
+    return log
+
+
+async def scrape_all_enseignes(db: Session) -> list[ScrapingLog]:
+    """
+    Scrape all active enseignes.
+    
+    Args:
+        db: Database session
+    
+    Returns:
+        List of ScrapingLog objects
+    """
+    enseignes = db.query(Enseigne).filter_by(is_active=True).all()
+    logger.info(f"Starting scraping for {len(enseignes)} active enseignes")
+    
+    logs = []
+    
+    async with async_playwright() as p:
+        browser = await p.chromium.launch(headless=True)
+        
+        for enseigne in enseignes:
+            try:
+                await asyncio.sleep(2)  # Delay between enseignes
+                log = await scrape_enseigne(enseigne, db, browser)
+                logs.append(log)
+            except Exception as e:
+                logger.error(f"Failed to scrape {enseigne.nom}: {e}")
+        
+        await browser.close()
+    
+    logger.info(f"Scraping complete: {len(logs)} enseignes processed")
+    return logs

app/services/scheduler.py

@@ -0,0 +1,62 @@
+"""
+Scheduler for automated catalog scraping
+
+Runs scraping jobs twice daily (6h and 18h) for all active enseignes.
+"""
+
+import logging
+from apscheduler.schedulers.asyncio import AsyncIOScheduler
+from apscheduler.triggers.cron import CronTrigger
+
+from app.database import SessionLocal
+from app.services.bonial_scraper import scrape_all_enseignes
+
+logger = logging.getLogger(__name__)
+
+# Global scheduler instance
+scheduler = AsyncIOScheduler()
+
+
+async def scheduled_scrape_job():
+    """Scheduled job to scrape all active enseignes"""
+    logger.info("Starting scheduled catalog scraping job")
+    
+    db = SessionLocal()
+    try:
+        logs = await scrape_all_ense ignes(db)
+        
+        total_new = sum(log.catalogues_nouveaux for log in logs)
+        total_found = sum(log.catalogues_trouves for log in logs)
+        
+        logger.info(
+            f"Scheduled scraping complete: "
+            f"{len(logs)} enseignes processed, "
+            f"{total_found} catalogues found, "
+            f"{total_new} new catalogues created"
+        )
+    except Exception as e:
+        logger.error(f"Error in scheduled scraping job: {e}")
+    finally:
+        db.close()
+
+
+def start_scheduler():
+    """Start the scheduler with configured jobs"""
+    # Schedule scraping twice daily at 6:00 and 18:00
+    scheduler.add_job(
+        scheduled_scrape_job,
+        CronTrigger(hour="6,18", minute=0),
+        id="catalog_scraping",
+        name="Scrape Bonial catalogs",
+        replace_existing=True,
+    )
+    
+    scheduler.start()
+    logger.info("Catalog scraping scheduler started (runs at 6:00 and 18:00)")
+
+
+def stop_scheduler():
+    """Stop the scheduler"""
+    if scheduler.running:
+        scheduler.shutdown()
+        logger.info("Catalog scraping scheduler stopped")

app/services/seed_enseignes.py

@@ -0,0 +1,128 @@
+"""
+Seed default enseignes (stores) for the catalog module
+"""
+
+import logging
+from sqlalchemy.orm import Session
+
+from app.models import Enseigne
+
+logger = logging.getLogger(__name__)
+
+# Data for the 9 enseignes
+ENSEIGNES_DATA = [
+    {
+        "nom": "Gifi",
+        "slug_bonial": "Gifi",
+        "couleur": "#E30613",
+        "site_url": "https://www.gifi.fr",
+        "description": "Décoration, maison, bazar",
+        "ordre_affichage": 1,
+    },
+    {
+        "nom": "Action",
+        "slug_bonial": "Action",
+        "couleur": "#0066B3",
+        "site_url": "https://www.action.com/fr-fr/",
+        "description": "Discount non-alimentaire",
+        "ordre_affichage": 2,
+    },
+    {
+        "nom": "Centrakor",
+        "slug_bonial": "Centrakor",
+        "couleur": "#E94E1B",
+        "site_url": "https://www.centrakor.com",
+        "description": "Décoration, maison",
+        "ordre_affichage": 3,
+    },
+    {
+        "nom": "La Foir'Fouille",
+        "slug_bonial": "La-Foir-Fouille",
+        "couleur": "#009639",
+        "site_url": "https://www.lafoirfouille.fr",
+        "description": "Bazar, décoration",
+        "ordre_affichage": 4,
+    },
+    {
+        "nom": "Stokomani",
+        "slug_bonial": "Stokomani",
+        "couleur": "#FF6600",
+        "site_url": "https://www.stokomani.fr",
+        "description": "Déstockage textile et maison",
+        "ordre_affichage": 5,
+    },
+    {
+        "nom": "B&M",
+        "slug_bonial": "BM",
+        "couleur": "#D4145A",
+        "site_url": "https://bmstores.fr",
+        "description": "Discount britannique",
+        "ordre_affichage": 6,
+    },
+    {
+        "nom": "L'Incroyable",
+        "slug_bonial": "L-incroyable",
+        "couleur": "#8B0000",
+        "site_url": "https://www.lincroyable.fr",
+        "description": "Décoration et mobilier discount (Groupe Althys, siège à Denain)",
+        "ordre_affichage": 7,
+    },
+    {
+        "nom": "Bazarland",
+        "slug_bonial": "Bazarland",
+        "couleur": "#FFCC00",
+        "site_url": None,
+        "description": "Bazar discount",
+        "ordre_affichage": 8,
+    },
+    {
+        "nom": "Noz",
+        "slug_bonial": "Noz",
+        "couleur": "#003366",
+        "site_url": "https://www.noz.fr",
+        "description": "Déstockage généraliste",
+        "ordre_affichage": 9,
+    },
+]
+
+
+def seed_enseignes(db: Session) -> int:
+    """
+    Seed the default enseignes if they don't exist.
+    
+    Returns:
+        Number of enseignes created
+    """
+    created = 0
+    
+    for data in ENSEIGNES_DATA:
+        # Check if enseigne already exists
+        existing = db.query(Enseigne).filter_by(slug_bonial=data["slug_bonial"]).first()
+        
+        if not existing:
+            enseigne = Enseigne(**data)
+            db.add(enseigne)
+            created += 1
+            logger.info(f"Created enseigne: {data['nom']}")
+        else:
+            logger.debug(f"Enseigne already exists: {data['nom']}")
+    
+    db.commit()
+    logger.info(f"Seeding complete: {created} enseigne(s) created")
+    
+    return created
+
+
+def get_enseigne_by_slug(db: Session, slug: str) -> Enseigne | None:
+    """Get an enseigne by its Bonial slug"""
+    return db.query(Enseigne).filter_by(slug_bonial=slug).first()
+
+
+def get_all_active_enseignes(db: Session) -> list[Enseigne]:
+    """Get all active enseignes ordered by display order"""
+    return (
+        db.query(Enseigne)
+        .filter_by(is_active=True)
+        .order_by(Enseigne.ordre_affichage)
+        .all()
+    )