From ef7cd7c1cf06f39d3b3b836c61f820e0d51ed96a Mon Sep 17 00:00:00 2001 From: R0m1k3 Date: Thu, 18 Dec 2025 15:20:19 +0100 Subject: [PATCH] feat: Add BMStores product parser and a new Playwright-based scraping service with popup handling. --- app/core/search_config.py | 43 ++++++-- app/services/amazon_scraper_service.py | 120 +++++++++++++---------- app/services/browserless_service.py | 81 ++++++++++----- app/services/parsers/bmstores_parser.py | 20 ++++ app/services/tracking_scraper_service.py | 30 ++++++ 5 files changed, 207 insertions(+), 87 deletions(-) diff --git a/app/core/search_config.py b/app/core/search_config.py index 8c24725..1be78dc 100644 --- a/app/core/search_config.py +++ b/app/core/search_config.py @@ -36,19 +36,42 @@ def get_amazon_proxies() -> list[dict]: }) return proxies -# === USER AGENTS === -USER_AGENT_POOL = [ - "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/131.0.0.0 Safari/537.36", - "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/130.0.0.0 Safari/537.36", - "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/131.0.0.0 Safari/537.36", - "Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:133.0) Gecko/20100101 Firefox/133.0", - "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/131.0.0.0 Safari/537.36 Edg/131.0.0.0", - "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/129.0.0.0 Safari/537.36", - "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/129.0.0.0 Safari/537.36", +# === USER AGENTS & STEALTH === +USER_AGENT_DATA = [ + { + "ua": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/131.0.0.0 Safari/537.36", + "ch": '"Google Chrome";v="131", "Chromium";v="131", "Not_A Brand";v="24"', + "platform": '"Windows"' + }, + { + "ua": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/130.0.0.0 Safari/537.36", + "ch": '"Google Chrome";v="130", "Chromium";v="130", "Not_A Brand";v="99"', + "platform": '"Windows"' + }, + { + "ua": "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/131.0.0.0 Safari/537.36", + "ch": '"Google Chrome";v="131", "Chromium";v="131", "Not_A Brand";v="24"', + "platform": '"macOS"' + }, + { + "ua": "Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/131.0.0.0 Safari/537.36", + "ch": '"Google Chrome";v="131", "Chromium";v="131", "Not_A Brand";v="24"', + "platform": '"Linux"' + }, + { + "ua": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/131.0.0.0 Safari/537.36 Edg/131.0.0.0", + "ch": '"Microsoft Edge";v="131", "Chromium";v="131", "Not_A Brand";v="24"', + "platform": '"Windows"' + } ] +def get_random_stealth_config() -> dict: + """Returns a random User-Agent and its corresponding Client Hints""" + return random.choice(USER_AGENT_DATA) + def get_random_user_agent() -> str: - return random.choice(USER_AGENT_POOL) + """Legacy helper for backward compatibility""" + return get_random_stealth_config()["ua"] # === SITE CONFIGURATIONS === SITE_CONFIGS = { diff --git a/app/services/amazon_scraper_service.py b/app/services/amazon_scraper_service.py index b0c6c52..6308807 100644 --- a/app/services/amazon_scraper_service.py +++ b/app/services/amazon_scraper_service.py @@ -180,17 +180,21 @@ class AmazonScraperService: @staticmethod async def _create_context(browser: Browser) -> BrowserContext: - """Create browser context with stealth settings""" + """Create browser context with dynamic stealth settings""" + from app.core.search_config import get_random_stealth_config + + stealth_config = get_random_stealth_config() + ua = stealth_config["ua"] + ch = stealth_config["ch"] + platform = stealth_config["platform"] + + logger.info(f"🎭 Using stealth profile: {ua[:50]}...") + context = await browser.new_context( viewport={"width": 1920, "height": 1080}, - 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" - ), + user_agent=ua, locale="fr-FR", timezone_id="Europe/Paris", - # Additional stealth parameters extra_http_headers={ "Accept": "text/html,application/xhtml+xml,application/xml;q=0.9,image/avif,image/webp,*/*;q=0.8", "Accept-Language": "fr-FR,fr;q=0.9,en-US;q=0.8,en;q=0.7", @@ -198,22 +202,24 @@ class AmazonScraperService: "DNT": "1", "Connection": "keep-alive", "Upgrade-Insecure-Requests": "1", + "Sec-Ch-Ua": ch, + "Sec-Ch-Ua-Mobile": "?0", + "Sec-Ch-Ua-Platform": platform, "Sec-Fetch-Dest": "document", "Sec-Fetch-Mode": "navigate", "Sec-Fetch-Site": "none", "Sec-Fetch-User": "?1", "Cache-Control": "max-age=0", + "Referer": "https://www.google.fr/", }, ) - # Comprehensive stealth mode + # Comprehensive stealth mode injector await context.add_init_script(""" - // Remove webdriver property - Object.defineProperty(navigator, 'webdriver', { - get: () => undefined - }); + // Reset webdriver + Object.defineProperty(navigator, 'webdriver', { get: () => undefined }); - // Add chrome property + // Re-mock chrome object window.chrome = { runtime: {}, loadTimes: function() {}, @@ -221,43 +227,46 @@ class AmazonScraperService: app: {} }; - // Override permissions - const originalQuery = window.navigator.permissions.query; - window.navigator.permissions.query = (parameters) => ( - parameters.name === 'notifications' ? - Promise.resolve({ state: Notification.permission }) : - originalQuery(parameters) - ); - - // Add plugins + // Consistent plugins Object.defineProperty(navigator, 'plugins', { - get: () => [1, 2, 3, 4, 5] + get: () => [ + { name: 'PDF Viewer', filename: 'internal-pdf-viewer' }, + { name: 'Chrome PDF Viewer', filename: 'internal-pdf-viewer' }, + { name: 'Chromium PDF Viewer', filename: 'internal-pdf-viewer' } + ] }); - // Add languages + // Languages Object.defineProperty(navigator, 'languages', { get: () => ['fr-FR', 'fr', 'en-US', 'en'] }); - // Override platform - Object.defineProperty(navigator, 'platform', { - get: () => 'Win32' - }); - - // Mock battery API - Object.defineProperty(navigator, 'getBattery', { - get: () => () => Promise.resolve({ - charging: true, - chargingTime: 0, - dischargingTime: Infinity, - level: 1 - }) - }); + // Device Memory (randomized) + Object.defineProperty(navigator, 'deviceMemory', { get: () => 8 }); """) await context.route("**/*", lambda route: route.continue_()) return context + @staticmethod + async def _simulate_human_behavior(page: Page): + """Perform subtle human-like interactions""" + import random + try: + # Random mouse movements + for _ in range(3): + x = random.randint(100, 800) + y = random.randint(100, 600) + await page.mouse.move(x, y, steps=10) + await asyncio.sleep(random.uniform(0.1, 0.3)) + + # Subtle scroll + await page.evaluate("window.scrollBy(0, window.innerHeight / 4)") + await asyncio.sleep(random.uniform(0.5, 1.0)) + await page.evaluate("window.scrollBy(0, -window.innerHeight / 5)") + except Exception as e: + logger.warning(f"Failed to simulate human behavior: {e}") + @staticmethod async def _handle_popups(page: Page): """Close Amazon popups/cookies""" @@ -306,35 +315,44 @@ class AmazonScraperService: try: # CRITICAL: Load Amazon homepage FIRST in same context to establish session logger.info("🏠 Loading Amazon homepage to establish session/cookies...") - await page.goto("https://www.amazon.fr", wait_until="domcontentloaded", timeout=30000) + await page.goto("https://www.amazon.fr", wait_until="networkidle", timeout=30000) logger.info("✅ Homepage loaded") # Handle homepage popups await cls._handle_popups(page) + + # Simulate human behavior on homepage + await cls._simulate_human_behavior(page) # Small delay - await page.wait_for_timeout(2000) + await asyncio.sleep(random.uniform(1.0, 3.0)) # NOW navigate to search in SAME context (cookies preserved) logger.info(f"🔍 Navigating to search: {search_url}") await page.goto(search_url, wait_until="domcontentloaded", timeout=60000) logger.info("Page loaded (domcontentloaded)") - # Wait for network idle + # Wait for content or block try: - await page.wait_for_load_state("networkidle", timeout=10000) - logger.info("Network idle reached") - except PlaywrightTimeoutError: - logger.info("Network idle timed out (non-critical)") - - # Handle popups - await cls._handle_popups(page) - - # Wait a bit for content - await page.wait_for_timeout(2000) + await page.wait_for_selector('div[data-component-type="s-search-result"], .s-result-list', timeout=10000) + logger.info("✅ Search results detected") + await cls._simulate_human_behavior(page) + except Exception: + logger.warning("🕒 Search results not found immediately, checking for blocks...") # Get HTML html_content = await page.content() + + # Proactive block detection + if "Type the characters you see in this image" in html_content or "Saisissez les caractères que vous voyez" in html_content: + logger.error("🚫 CAPTCHA / Bot detection triggered") + return [] + + if "Identifiez-vous" in html_content and "commander" not in html_content: + logger.warning("⚠️ Redirected to login wall") + # Try one more time with a different behavior or just fail + return [] + logger.info(f"✅ Page content extracted ({len(html_content)} bytes)") if len(html_content) < 10000: diff --git a/app/services/browserless_service.py b/app/services/browserless_service.py index 07c5f60..1b8dde2 100644 --- a/app/services/browserless_service.py +++ b/app/services/browserless_service.py @@ -172,20 +172,31 @@ class BrowserlessService: @staticmethod async def _create_context(browser: Browser, use_proxy: bool = False) -> BrowserContext: - """Create a new browser context with stealth settings.""" - user_agent = get_random_user_agent() + """Create a new browser context with dynamic stealth settings.""" + from app.core.search_config import get_random_stealth_config + + stealth_config = get_random_stealth_config() + ua = stealth_config["ua"] + ch = stealth_config["ch"] + platform = stealth_config["platform"] + + logger.info(f"🎭 Using stealth profile: {ua[:50]}...") options = { "viewport": {"width": 1920, "height": 1080}, - "user_agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/122.0.0.0 Safari/537.36", + "user_agent": ua, "locale": "fr-FR", "timezone_id": "Europe/Paris", "java_script_enabled": True, "bypass_csp": True, "extra_http_headers": { + "Accept": "text/html,application/xhtml+xml,application/xml;q=0.9,image/avif,image/webp,*/*;q=0.8", "Accept-Language": "fr-FR,fr;q=0.9,en-US;q=0.8,en;q=0.7", - "Sec-Ch-Ua-Platform": '"Windows"', + "Sec-Ch-Ua": ch, + "Sec-Ch-Ua-Mobile": "?0", + "Sec-Ch-Ua-Platform": platform, "Upgrade-Insecure-Requests": "1", + "Referer": "https://www.google.fr/", }, } @@ -194,26 +205,35 @@ class BrowserlessService: context = await browser.new_context(**options) - # Inject stealth scripts + # Comprehensive stealth mode injector await context.add_init_script(""" - Object.defineProperty(navigator, 'webdriver', { - get: () => undefined + // Reset webdriver + Object.defineProperty(navigator, 'webdriver', { get: () => undefined }); + + // Re-mock chrome object + window.chrome = { + runtime: {}, + loadTimes: function() {}, + csi: function() {}, + app: {} + }; + + // Consistent plugins + Object.defineProperty(navigator, 'plugins', { + get: () => [ + { name: 'PDF Viewer', filename: 'internal-pdf-viewer' }, + { name: 'Chrome PDF Viewer', filename: 'internal-pdf-viewer' }, + { name: 'Chromium PDF Viewer', filename: 'internal-pdf-viewer' } + ] }); + + // Languages Object.defineProperty(navigator, 'languages', { get: () => ['fr-FR', 'fr', 'en-US', 'en'] }); - Object.defineProperty(navigator, 'plugins', { - get: () => [1, 2, 3, 4, 5] - }); - window.chrome = { runtime: {} }; - // Mock permissions - const originalQuery = window.navigator.permissions.query; - window.navigator.permissions.query = (parameters) => ( - parameters.name === 'notifications' ? - Promise.resolve({ state: Notification.permission }) : - originalQuery(parameters) - ); + // Device Memory (randomized) + Object.defineProperty(navigator, 'deviceMemory', { get: () => 8 }); """) await context.route("**/*", lambda route: route.continue_()) @@ -517,28 +537,37 @@ class BrowserlessService: page = await context.new_page() try: + # Random human-like lead-in delay + import random + await asyncio.sleep(random.uniform(0.5, 2.0)) + await cls._navigate_and_wait(page, url, 30000) await cls._handle_popups(page) - # Check for Amazon Captcha / Login Wall + # Check for Amazon Captcha / Login Wall / Blocking if "amazon" in url.lower(): content_check = await page.content() - if ( + is_blocked = ( "Type the characters you see in this image" in content_check or "Saisissez les caractères que vous voyez" in content_check - or "Sign in or create an account" in content_check - or "Identifiez-vous ou créez un compte" in content_check - ): - logger.warning(f"⚠️ Amazon Captcha/Login Wall detected (Attempt {attempt+1}/{retries})") + or ("Sign in or create an account" in content_check and "orders" not in url.lower()) + or ("Identifiez-vous" in content_check and "commande" not in url.lower()) + or "api-services-support@amazon.com" in content_check + or "service-unavailable" in content_check + ) + + if is_blocked: + logger.warning(f"⚠️ Amazon Blocking/Login Wall detected (Attempt {attempt+1}/{retries})") if attempt < retries - 1: await context.close() - await asyncio.sleep(2 + attempt * 2) # Backoff + # Exponential backoff with jitter + await asyncio.sleep(2 * (attempt + 1) + random.uniform(0.5, 1.5)) continue else: logger.error("❌ Amazon blocked all attempts") return "", "" - # Amazon-specific wait + # Amazon-specific wait for price or content if "amazon" in url.lower() and "/dp/" in url: amazon_selectors = [".a-price .a-offscreen", "#corePriceDisplay_desktop_feature_div"] for selector in amazon_selectors: diff --git a/app/services/parsers/bmstores_parser.py b/app/services/parsers/bmstores_parser.py index 79ba12c..d48abda 100644 --- a/app/services/parsers/bmstores_parser.py +++ b/app/services/parsers/bmstores_parser.py @@ -56,3 +56,23 @@ class BMStoresParser(BaseParser): logger.info(f"BMStoresParser found {len(results)} results") return results + + def parse_product_details(self, html: str, product_url: str) -> dict: + """ + Extract price and stock from B&M product page + """ + soup = BeautifulSoup(html, "html.parser") + details = {"price": None, "in_stock": True} + + # 1. Look for the main price (avoid suggested products prices) + # B&M uses .current-price .price on detail pages + price_el = soup.select_one(".product-prices .current-price .price, .product-price-and-shipping .price, [itemprop='price']") + if price_el: + details["price"] = self.parse_price_text(price_el.get_text()) + + # 2. Check stock + stock_text = soup.get_text().lower() + if "épuisé" in stock_text or "indisponible" in stock_text: + details["in_stock"] = False + + return details diff --git a/app/services/tracking_scraper_service.py b/app/services/tracking_scraper_service.py index b0a60ce..ff184ea 100644 --- a/app/services/tracking_scraper_service.py +++ b/app/services/tracking_scraper_service.py @@ -168,6 +168,36 @@ class ScraperService: await ScraperService._handle_popups(page) + # FORCER LE MAGASIN NANCY POUR B&M STORES + if "bmstores.fr" in url: + try: + logger.info("B&M Stores detected: checking store selection (Nancy)...") + # Check if store is already selected or if modal is needed + store_btn = page.locator(".mod-shops .btn-shop, .js-show-shops") + if await store_btn.count() > 0: + logger.info("Opening store selector...") + await store_btn.first.click(timeout=2000) + await page.wait_for_timeout(1000) + + # Type Nancy in the search input + search_input = page.locator("#search_mag") + if await search_input.count() > 0: + await search_input.fill("Nancy") + await page.keyboard.press("Enter") + await page.wait_for_timeout(1500) + + # Select the first Nancy store (Nancy Essey or Nancy Centre) + select_btn = page.locator(".shop-list .btn-select-shop, button:has-text('Choisir ce magasin')") + if await select_btn.count() > 0: + logger.info("Selecting Nancy store...") + await select_btn.first.click() + await page.wait_for_timeout(2000) + # Refresh page or wait for update + await page.reload(wait_until="domcontentloaded") + await page.wait_for_timeout(1000) + except Exception as e: + logger.warning(f"Failed to force B&M store Nancy: {e}") + if selector: await ScraperService._wait_for_selector(page, selector) else: