init

crawler/scenarios/fresh_session.py

@@ -0,0 +1,52 @@
+"""Scenario 2 — fresh cookie jar per run so the site treats the visitor as a new user."""
+from __future__ import annotations
+
+import undetected_chromedriver as uc
+from selenium.common.exceptions import TimeoutException
+from selenium.webdriver.support.ui import WebDriverWait
+
+from config import config
+from crawler.driver import human_delay, make_driver
+
+
+class FreshSessionScenario:
+    """
+    Opens the target URL in a brand-new browser profile (no stored cookies,
+    localStorage, or cache) so each run appears as a first-time visitor.
+
+    Because undetected-chromedriver already creates an isolated temp profile
+    per instance, simply spinning up a new driver is sufficient — but we also
+    explicitly delete all cookies after loading to be safe.
+    """
+
+    def __init__(self, driver: uc.Chrome, timeout: int = 30) -> None:
+        self.driver = driver
+        self.wait = WebDriverWait(driver, timeout)
+
+    def run(self) -> dict[str, object]:
+        """Navigate to the target, clear any residual state, return page info."""
+        self.driver.delete_all_cookies()
+
+        # Clear localStorage / sessionStorage via JS
+        self.driver.execute_script(
+            "try { window.localStorage.clear(); window.sessionStorage.clear(); } catch(e) {}"
+        )
+
+        self.driver.get(config.target_url)
+        human_delay(2.0, 4.0)
+
+        # Wait for the page to reach a ready state
+        self.wait.until(
+            lambda d: d.execute_script("return document.readyState") == "complete"
+        )
+
+        title = self.driver.title
+        current_url = self.driver.current_url
+        cookies = self.driver.get_cookies()
+
+        return {
+            "title": title,
+            "url": current_url,
+            "cookie_count": len(cookies),
+            "cookies": cookies,
+        }