fix(stdlib): Gate url.full and http.query behind send_default_pii (#6666)

`url.full`, `url.query`, and `url.fragment` span attributes set by the
stdlib (httplib) integration are now gated behind `send_default_pii`,
consistent with the same fix already applied to the aiohttp and wsgi
integrations. These attributes can contain sensitive query parameters or
path fragments, so they should only be captured when the user has
explicitly opted in.

`http.fragment` on the legacy span data path is unconditionally omitted
since fragments are never sent to the server and carry no useful
diagnostic value regardless of PII setting.

Fixes #PY-2557
Fixes #6665

Author: ericapisani

sentry_sdk/integrations/stdlib.py

@@ -8,7 +8,7 @@
 import sentry_sdk
 from sentry_sdk.consts import OP, SPANDATA
 from sentry_sdk.integrations import Integration
-from sentry_sdk.scope import add_global_event_processor
+from sentry_sdk.scope import add_global_event_processor, should_send_default_pii
 from sentry_sdk.traces import StreamedSpan
 from sentry_sdk.tracing import Span
 from sentry_sdk.tracing_utils import (
@@ -124,10 +124,10 @@ def putrequest(
                 },
             )
 
-            if parsed_url is not None:
+            if parsed_url is not None and should_send_default_pii():
+                span.set_attribute(SPANDATA.URL_FRAGMENT, parsed_url.fragment)
                 span.set_attribute(SPANDATA.URL_FULL, parsed_url.url)
                 span.set_attribute(SPANDATA.URL_QUERY, parsed_url.query)
-                span.set_attribute(SPANDATA.URL_FRAGMENT, parsed_url.fragment)
 
             set_on_span = span.set_attribute
 
@@ -141,9 +141,9 @@ def putrequest(
 
             span.set_data(SPANDATA.HTTP_METHOD, method)
             if parsed_url is not None:
+                span.set_data(SPANDATA.HTTP_FRAGMENT, parsed_url.fragment)
                 span.set_data("url", parsed_url.url)
                 span.set_data(SPANDATA.HTTP_QUERY, parsed_url.query)
-                span.set_data(SPANDATA.HTTP_FRAGMENT, parsed_url.fragment)
 
             set_on_span = span.set_data
 

tests/integrations/requests/test_requests.py

@@ -13,7 +13,7 @@
 
 
 def test_crumb_capture(sentry_init, capture_events):
-    sentry_init(integrations=[StdlibIntegration()])
+    sentry_init(integrations=[StdlibIntegration()], send_default_pii=True)
     events = capture_events()
 
     url = f"http://localhost:{PORT}/hello-world"  # noqa:E231
@@ -51,7 +51,7 @@ def test_crumb_capture(sentry_init, capture_events):
     ],
 )
 def test_crumb_capture_client_error(sentry_init, capture_events, status_code, level):
-    sentry_init(integrations=[StdlibIntegration()])
+    sentry_init(integrations=[StdlibIntegration()], send_default_pii=True)
 
     events = capture_events()
 

tests/integrations/stdlib/test_httplib.py

@@ -77,7 +77,7 @@ def create_chunked_server():
 
 
 def test_crumb_capture(sentry_init, capture_events):
-    sentry_init(integrations=[StdlibIntegration()])
+    sentry_init(integrations=[StdlibIntegration()], send_default_pii=True)
     events = capture_events()
 
     url = "http://localhost:{}/some/random/url".format(PORT)
@@ -113,7 +113,7 @@ def test_crumb_capture(sentry_init, capture_events):
     ],
 )
 def test_crumb_capture_client_error(sentry_init, capture_events, status_code, level):
-    sentry_init(integrations=[StdlibIntegration()])
+    sentry_init(integrations=[StdlibIntegration()], send_default_pii=True)
     events = capture_events()
 
     url = f"http://localhost:{PORT}/status/{status_code}"  # noqa:E231
@@ -151,7 +151,11 @@ def before_breadcrumb(crumb, hint):
         crumb["data"]["extra"] = "foo"
         return crumb
 
-    sentry_init(integrations=[StdlibIntegration()], before_breadcrumb=before_breadcrumb)
+    sentry_init(
+        integrations=[StdlibIntegration()],
+        before_breadcrumb=before_breadcrumb,
+        send_default_pii=True,
+    )
     events = capture_events()
 
     url = "http://localhost:{}/some/random/url".format(PORT)
@@ -203,7 +207,7 @@ def test_httplib_misuse(sentry_init, capture_events, request):
     wrongly.
     """
 
-    sentry_init()
+    sentry_init(send_default_pii=True)
     events = capture_events()
 
     conn = HTTPConnection("localhost", PORT)
@@ -1136,15 +1140,18 @@ def test_http_timeout(
 
 @pytest.mark.parametrize("tunnel_port", [8080, None])
 @pytest.mark.parametrize("span_streaming", [True, False])
+@pytest.mark.parametrize("send_default_pii", [True, False])
 def test_proxy_http_tunnel(
     sentry_init,
     capture_events,
     capture_items,
     tunnel_port,
     span_streaming,
+    send_default_pii,
 ):
     sentry_init(
         traces_sample_rate=1.0,
+        send_default_pii=send_default_pii,
         _experiments={"trace_lifecycle": "stream" if span_streaming else "static"},
     )
 
@@ -1154,7 +1161,7 @@ def test_proxy_http_tunnel(
         with sentry_sdk.traces.start_span(name="custom parent"):
             conn = HTTPConnection("localhost", PROXY_PORT)
             conn.set_tunnel("api.example.com", tunnel_port)
-            conn.request("GET", "/foo")
+            conn.request("GET", "/foo?bar=1")
             conn.getresponse()
 
         sentry_sdk.flush()
@@ -1167,31 +1174,38 @@ def test_proxy_http_tunnel(
 
         port_modifier = f":{tunnel_port}" if tunnel_port else ""
         assert span["name"] == f"GET http://api.example.com{port_modifier}/foo"
-        assert (
-            span["attributes"][SPANDATA.URL_FULL]
-            == f"http://api.example.com{port_modifier}/foo"
-        )
         assert span["attributes"][SPANDATA.HTTP_REQUEST_METHOD] == "GET"
         assert span["attributes"][SPANDATA.NETWORK_PEER_ADDRESS] == "localhost"
         assert span["attributes"][SPANDATA.NETWORK_PEER_PORT] == PROXY_PORT
+
+        if send_default_pii:
+            assert (
+                span["attributes"][SPANDATA.URL_FULL]
+                == f"http://api.example.com{port_modifier}/foo"
+            )
+            assert span["attributes"][SPANDATA.URL_QUERY] == "bar=1"
+        else:
+            assert SPANDATA.URL_FULL not in span["attributes"]
+            assert SPANDATA.URL_QUERY not in span["attributes"]
     else:
         events = capture_events()
 
         with start_transaction(name="test_transaction"):
             conn = HTTPConnection("localhost", PROXY_PORT)
             conn.set_tunnel("api.example.com", tunnel_port)
-            conn.request("GET", "/foo")
+            conn.request("GET", "/foo?bar=1")
             conn.getresponse()
 
         (event,) = events
         (span,) = event["spans"]
 
         port_modifier = f":{tunnel_port}" if tunnel_port else ""
         assert span["description"] == f"GET http://api.example.com{port_modifier}/foo"
-        assert span["data"]["url"] == f"http://api.example.com{port_modifier}/foo"
         assert span["data"][SPANDATA.HTTP_METHOD] == "GET"
         assert span["data"][SPANDATA.NETWORK_PEER_ADDRESS] == "localhost"
         assert span["data"][SPANDATA.NETWORK_PEER_PORT] == PROXY_PORT
+        assert span["data"]["url"] == f"http://api.example.com{port_modifier}/foo"
+        assert span["data"][SPANDATA.HTTP_QUERY] == "bar=1"
 
 
 @pytest.mark.parametrize("span_streaming", [True, False])