fix(watercrawl): bound result download timeout (#37495)

api/core/rag/extractor/watercrawl/client.py

@@ -217,7 +217,7 @@ class WaterCrawlAPIClient(BaseAPIClient):
                 return event_data["data"]
 
     def download_result(self, result_object: dict[str, Any]):
-        response = httpx.get(result_object["result"], timeout=None)
+        response = httpx.get(result_object["result"], timeout=30)
         try:
             response.raise_for_status()
             result_object["result"] = response.json()

api/tests/unit_tests/core/rag/extractor/watercrawl/test_watercrawl.py

@@ -242,11 +242,18 @@ class TestWaterCrawlAPIClient:
         client = WaterCrawlAPIClient(api_key="k")
 
         response = _response(200, {"markdown": "body"})
-        monkeypatch.setattr(client_module.httpx, "get", lambda *args, **kwargs: response)
+        captured = {}
+
+        def fake_get(*args, **kwargs):
+            captured.update(kwargs)
+            return response
+
+        monkeypatch.setattr(client_module.httpx, "get", fake_get)
 
         result = client.download_result({"result": "https://example.com/result.json"})
 
         assert result["result"] == {"markdown": "body"}
+        assert captured["timeout"] is not None
         response.close.assert_called_once()