Open target URL during Chrome startup

crawler/scenarios/fresh_session.py

@@ -13,9 +13,6 @@ from logger import get_logger
 
 log = get_logger(__name__)
 
-_BLANK_URLS = {"", "about:blank", "data:,"}
-
-
 class FreshSessionError(RuntimeError):
     """The browser could not leave its initial blank page."""
 
@@ -32,6 +29,11 @@ def _normalise_url(url: str) -> str:
     return value
 
 
+def _is_web_url(url: str) -> bool:
+    parsed = urlsplit(url)
+    return parsed.scheme in {"http", "https"} and bool(parsed.netloc)
+
+
 class FreshSessionScenario:
     """
     Opens the target URL in a brand-new browser profile (no stored cookies,
@@ -76,10 +78,18 @@ class FreshSessionScenario:
                 )
 
         try:
-            self.wait.until(lambda d: d.current_url not in _BLANK_URLS)
+            # Chrome's internal New Tab URLs (chrome://newtab and
+            # chrome://new-tab-page) are not successful navigation. Wait until
+            # the address bar actually contains an HTTP(S) page or redirect.
+            self.wait.until(lambda d: _is_web_url(d.current_url))
         except TimeoutException as exc:
+            try:
+                current_url = self.driver.current_url
+            except WebDriverException:
+                current_url = "<unavailable>"
             raise FreshSessionError(
-                f"navigation to {self.url} never left Chrome's blank page"
+                f"navigation to {self.url} never reached a web page "
+                f"(Chrome remained at {current_url})"
             ) from exc
 
         human_delay(1.0, 2.0)