04/23 Enhance app functionalities 2

views/admin_websites_view.py

@@ -427,6 +427,21 @@ class WebsiteDialog(tk.Toplevel):
         if not name or not url:
             show_error("Name and URL are required.")
             return
+
+        # Normalise and validate URL before writing to the database.
+        # Reject schemes that could be executed in-browser (javascript:, data:, etc.)
+        # and enforce that a host is present so the health-check thread and the
+        # "open in browser" action both receive a usable address.
+        url = _validate_and_normalise_url(url)
+        if url is None:
+            show_error(
+                "The URL entered is not valid.\n\n"
+                "Please enter a full web address, e.g.:\n"
+                "  https://www.example.com\n"
+                "  http://intranet.local/app"
+            )
+            return
+
         if visibility == "assigned" and not assigned_user_ids:
             show_error("Please assign at least one user, or set visibility to All Users.")
             return
@@ -463,3 +478,42 @@ class WebsiteDialog(tk.Toplevel):
             self.destroy()
         except Exception as e:
             show_error(f"Save failed:\n{e}")
+
+
+# ─── URL validation helper ────────────────────────────────────────────────────
+
+def _validate_and_normalise_url(raw: str) -> "str | None":
+    """
+    Validate and normalise a user-supplied URL string.
+
+    Rules:
+      - If no scheme is present, prepend 'https://'.
+      - Only 'http' and 'https' schemes are accepted.
+      - A non-empty netloc (host) must be present.
+      - Returns the normalised URL string on success, None on failure.
+
+    This prevents dangerous schemes (javascript:, data:, file:, etc.) from
+    reaching the database, the health-check thread, or webbrowser.open().
+    """
+    import urllib.parse
+
+    if not raw:
+        return None
+
+    # Prepend https:// if no scheme is given so urlparse can parse the host
+    if "://" not in raw:
+        raw = "https://" + raw
+
+    try:
+        parts = urllib.parse.urlparse(raw)
+    except Exception:
+        return None
+
+    if parts.scheme.lower() not in ("http", "https"):
+        return None
+
+    if not parts.netloc:
+        return None
+
+    # Reconstruct a clean URL (strips any leading/trailing whitespace artefacts)
+    return urllib.parse.urlunparse(parts)