ref(boto3): Drop send_default_pii support (#7684)

### Description
- remove `send_default_pii` branching in `tests_s3.py`.
- Remove `send_default_pii` fallback from
`tracing_utils.get_url_attributes()`

#### Issues
Resolves #7658

Author: pabloDeputter

sentry_sdk/integrations/httpx.py

@@ -7,7 +7,7 @@
 from sentry_sdk.tracing_utils import (
     add_http_breadcrumb,
     add_http_request_source,
-    get_url_attributes,
+    get_url_attributes_legacy,
     propagate_trace_headers,
 )
 from sentry_sdk.utils import (
@@ -59,7 +59,7 @@ def send(self: "Client", request: "Request", **kwargs: "Any") -> "Response":
         with capture_internal_exceptions():
             parsed_url = parse_url(str(request.url), sanitize=False)
 
-        url_attributes = get_url_attributes(client, parsed_url)
+        url_attributes = get_url_attributes_legacy(client, parsed_url)
 
         if sentry_sdk.get_current_span() is None:
             span_ctx = nullcontext()
@@ -148,7 +148,7 @@ async def send(
                 },
             )
 
-        url_attributes = get_url_attributes(client, parsed_url)
+        url_attributes = get_url_attributes_legacy(client, parsed_url)
 
         with span_ctx as span:
             propagate_trace_headers(client, request)

sentry_sdk/integrations/httpx2.py

@@ -7,7 +7,7 @@
 from sentry_sdk.tracing_utils import (
     add_http_breadcrumb,
     add_http_request_source,
-    get_url_attributes,
+    get_url_attributes_legacy,
     propagate_trace_headers,
 )
 from sentry_sdk.utils import (
@@ -59,7 +59,7 @@ def send(self: "Client", request: "Request", **kwargs: "Any") -> "Response":
         with capture_internal_exceptions():
             parsed_url = parse_url(str(request.url), sanitize=False)
 
-        url_attributes = get_url_attributes(client, parsed_url)
+        url_attributes = get_url_attributes_legacy(client, parsed_url)
 
         if sentry_sdk.get_current_span() is None:
             span_ctx = nullcontext()
@@ -148,7 +148,7 @@ async def send(
                 },
             )
 
-        url_attributes = get_url_attributes(client, parsed_url)
+        url_attributes = get_url_attributes_legacy(client, parsed_url)
 
         with span_ctx as span:
             propagate_trace_headers(client, request)

sentry_sdk/integrations/pyreqwest.py

@@ -7,7 +7,7 @@
 from sentry_sdk.tracing_utils import (
     add_http_breadcrumb,
     add_http_request_source,
-    get_url_attributes,
+    get_url_attributes_legacy,
     propagate_trace_headers,
 )
 from sentry_sdk.utils import (
@@ -155,7 +155,7 @@ async def sentry_async_middleware(
         # after the request has been sent
         parsed_url = parse_url(str(request.url), sanitize=False)
 
-    url_attributes = get_url_attributes(sentry_sdk.get_client(), parsed_url)
+    url_attributes = get_url_attributes_legacy(sentry_sdk.get_client(), parsed_url)
 
     response = None
     with _sentry_pyreqwest_span(request, url_attributes) as span:
@@ -193,7 +193,7 @@ def sentry_sync_middleware(
         # after the request has been sent
         parsed_url = parse_url(str(request.url), sanitize=False)
 
-    url_attributes = get_url_attributes(sentry_sdk.get_client(), parsed_url)
+    url_attributes = get_url_attributes_legacy(sentry_sdk.get_client(), parsed_url)
 
     response = None
     with _sentry_pyreqwest_span(request, url_attributes) as span:

sentry_sdk/integrations/stdlib.py

@@ -14,7 +14,7 @@
     EnvironHeaders,
     add_http_breadcrumb,
     add_http_request_source,
-    get_url_attributes,
+    get_url_attributes_legacy,
     should_propagate_trace,
 )
 from sentry_sdk.utils import (
@@ -274,7 +274,7 @@ def putrequest(
         span: "Optional[Span]" = None
         breadcrumb: "dict[str, Any]" = {}
 
-        url_attributes = get_url_attributes(client, parsed_url)
+        url_attributes = get_url_attributes_legacy(client, parsed_url)
 
         breadcrumb[SPANDATA.HTTP_REQUEST_METHOD] = method
         breadcrumb.update(url_attributes)

sentry_sdk/tracing_utils.py

@@ -191,8 +191,9 @@ def add_http_breadcrumb(status_code: "Optional[int]", data: "dict[str, Any]") ->
     sentry_sdk.add_breadcrumb(**kwargs)
 
 
-def get_url_attributes(
-    client: "sentry_sdk.client.BaseClient", parsed_url: "Optional[ParsedUrl]"
+def get_url_attributes_legacy(
+    client: "sentry_sdk.client.BaseClient",
+    parsed_url: "Optional[ParsedUrl]",
 ) -> "Attributes":
     """Build the `url.*` span attributes for an outgoing HTTP request.
 
@@ -231,6 +232,40 @@ def get_url_attributes(
     return attributes
 
 
+def get_url_attributes(
+    client: "sentry_sdk.client.BaseClient",
+    parsed_url: "Optional[ParsedUrl]",
+) -> "Attributes":
+    """Build the `url.*` span attributes for an outgoing HTTP request.
+
+    The query string is filtered through the resolved `data_collection`
+    settings.
+    """
+    attributes: "Attributes" = {}
+    if parsed_url is None:
+        return attributes
+
+    query: "Optional[str]" = None
+    if parsed_url.query:
+        query = _apply_data_collection_filtering_to_query_string(
+            query_string=parsed_url.query,
+            behaviour=client.options["data_collection"]["url_query_params"],
+        )
+
+    url_full = parsed_url.url
+    if query:
+        attributes[SPANDATA.URL_QUERY] = query
+        url_full += "?" + query
+
+    if parsed_url.fragment:
+        attributes[SPANDATA.URL_FRAGMENT] = parsed_url.fragment
+        url_full += "#" + parsed_url.fragment
+
+    attributes[SPANDATA.URL_FULL] = url_full
+
+    return attributes
+
+
 def _get_frame_module_abs_path(frame: "FrameType") -> "Optional[str]":
     try:
         return frame.f_code.co_filename

tests/integrations/boto3/test_s3.py

@@ -50,17 +50,11 @@ def test_basic(
     assert span["name"] == "aws.s3.ListObjects"
 
 
-@pytest.mark.parametrize("send_default_pii", [True, False])
-def test_streaming(
-    sentry_init,
-    capture_events,
-    capture_items,
-    send_default_pii,
-):
+def test_streaming(sentry_init, capture_events, capture_items):
     sentry_init(
         traces_sample_rate=1.0,
         integrations=[Boto3Integration()],
-        send_default_pii=send_default_pii,
+        data_collection={},
     )
 
     s3 = session.resource("s3")
@@ -101,15 +95,11 @@ def test_streaming(
         "thread.id": mock.ANY,
         "thread.name": mock.ANY,
     }
-    if send_default_pii:
-        expected_attrs["url.full"] = "https://bucket.s3.amazonaws.com/foo.pdf"
+    expected_attrs["url.full"] = "https://bucket.s3.amazonaws.com/foo.pdf"
     assert span1["attributes"] == ApproxDict(expected_attrs)
 
     assert "url.fragment" not in span1["attributes"]
     assert "url.query" not in span1["attributes"]
-    if not send_default_pii:
-        assert "url.full" not in span1["attributes"]
-
     span2 = spans[1]
     assert span2["attributes"]["sentry.op"] == "http.client.stream"
     assert span2["name"] == "aws.s3.GetObject"
@@ -157,7 +147,7 @@ def test_omit_url_data_if_parsing_fails(
     sentry_init(
         traces_sample_rate=1.0,
         integrations=[Boto3Integration()],
-        send_default_pii=True,
+        data_collection={},
     )
 
     s3 = session.resource("s3")
@@ -228,12 +218,11 @@ def test_span_origin(
     assert spans[0]["attributes"]["sentry.origin"] == "auto.http.boto3"
 
 
-@pytest.mark.parametrize("send_default_pii", [True, False])
-def test_breadcrumb(sentry_init, capture_events, send_default_pii):
+def test_breadcrumb(sentry_init, capture_events):
     sentry_init(
         integrations=[Boto3Integration()],
         default_integrations=False,
-        send_default_pii=send_default_pii,
+        data_collection={},
     )
 
     s3 = session.resource("s3")
@@ -253,24 +242,14 @@ def test_breadcrumb(sentry_init, capture_events, send_default_pii):
     assert crumb["type"] == "http"
     assert crumb["category"] == "httplib"
 
-    if send_default_pii:
-        assert crumb["data"] == ApproxDict(
-            {
-                SPANDATA.URL_FULL: mock.ANY,
-                SPANDATA.HTTP_REQUEST_METHOD: "GET",
-                SPANDATA.URL_QUERY: mock.ANY,
-            }
-        )
-        assert SPANDATA.URL_FRAGMENT not in crumb["data"]
-    else:
-        assert crumb["data"] == ApproxDict(
-            {
-                SPANDATA.HTTP_REQUEST_METHOD: "GET",
-            }
-        )
-        assert SPANDATA.URL_FULL not in crumb["data"]
-        assert SPANDATA.URL_QUERY not in crumb["data"]
-        assert SPANDATA.URL_FRAGMENT not in crumb["data"]
+    assert crumb["data"] == ApproxDict(
+        {
+            SPANDATA.URL_FULL: mock.ANY,
+            SPANDATA.HTTP_REQUEST_METHOD: "GET",
+            SPANDATA.URL_QUERY: mock.ANY,
+        }
+    )
+    assert SPANDATA.URL_FRAGMENT not in crumb["data"]
 
 
 BUCKET_URL = "https://bucket.s3.amazonaws.com/"
@@ -279,19 +258,9 @@ def test_breadcrumb(sentry_init, capture_events, send_default_pii):
 # means the URL is recorded without a query string.
 # Structure of the parameters is "init_kwargs, expected_query"
 URL_QUERY_PARAMS = [
-    pytest.param(
-        {"send_default_pii": True},
-        "list-type=2&prefix=foo&continuation-token=abc&encoding-type=url",
-        id="send_default_pii_true",
-    ),
-    pytest.param(
-        {"send_default_pii": False},
-        None,
-        id="send_default_pii_false",
-    ),
     pytest.param(
         {},
-        None,
+        "list-type=2&prefix=foo&continuation-token=%5BFiltered%5D&encoding-type=url",
         id="defaults",
     ),
     pytest.param(
@@ -334,14 +303,6 @@ def test_breadcrumb(sentry_init, capture_events, send_default_pii):
         "",
         id="data_collection_off",
     ),
-    pytest.param(
-        {
-            "send_default_pii": True,
-            "data_collection": {"url_query_params": {"mode": "off"}},
-        },
-        "",
-        id="data_collection_wins_over_send_default_pii",
-    ),
 ]