Unpack WSGI environ into span attrs accessible in traces sampler (#3775)

Author: sentrivana

MIGRATION_GUIDE.md

@@ -21,13 +21,27 @@ Looking to upgrade from Sentry SDK 2.x to 3.x? Here's a comprehensive list of wh
 - clickhouse-driver integration: The query is now available under the `db.query.text` span attribute (only if `send_default_pii` is `True`).
 - `sentry_sdk.init` now returns `None` instead of a context manager.
 - The `sampling_context` argument of `traces_sampler` now additionally contains all span attributes known at span start.
-- The `sampling_context` argument of `traces_sampler` doesn't contain the `asgi_scope` object anymore for ASGI frameworks. Instead, the individual properties on the scope, if available, are accessible as follows:
+- The `sampling_context` argument of `traces_sampler` doesn't contain the `wsgi_environ` object anymore for WSGI frameworks. Instead, the individual properties of the environment are accessible, if available, as follows:
+
+  | Env property      | Sampling context key(s)                           |
+  | ----------------- | ------------------------------------------------- |
+  | `PATH_INFO`       | `url.path`                                        |
+  | `QUERY_STRING`    | `url.query`                                       |
+  | `REQUEST_METHOD`  | `http.request.method`                             |
+  | `SERVER_NAME`     | `server.address`                                  |
+  | `SERVER_PORT`     | `server.port`                                     |
+  | `SERVER_PROTOCOL` | `server.protocol.name`, `server.protocol.version` |
+  | `wsgi.url_scheme` | `url.scheme`                                      |
+  | full URL          | `url.full`                                        |
+
+- The `sampling_context` argument of `traces_sampler` doesn't contain the `asgi_scope` object anymore for ASGI frameworks. Instead, the individual properties of the scope, if available, are accessible as follows:
 
   | Scope property | Sampling context key(s)         |
   | -------------- | ------------------------------- |
   | `type`         | `network.protocol.name`         |
   | `scheme`       | `url.scheme`                    |
   | `path`         | `url.path`                      |
+  | `query`        | `url.query`                     |
   | `http_version` | `network.protocol.version`      |
   | `method`       | `http.request.method`           |
   | `server`       | `server.address`, `server.port` |

sentry_sdk/integrations/asgi.py

@@ -356,6 +356,7 @@ def _prepopulate_attributes(scope):
         full_url = _get_url(scope)
         query = _get_query(scope)
         if query:
+            attributes["url.query"] = query
             full_url = f"{full_url}?{query}"
 
         attributes["url.full"] = full_url

sentry_sdk/integrations/wsgi.py

@@ -48,6 +48,15 @@ def __call__(self, status, response_headers, exc_info=None):  # type: ignore
 
 DEFAULT_TRANSACTION_NAME = "generic WSGI request"
 
+ENVIRON_TO_ATTRIBUTE = {
+    "PATH_INFO": "url.path",
+    "QUERY_STRING": "url.query",
+    "REQUEST_METHOD": "http.request.method",
+    "SERVER_NAME": "server.address",
+    "SERVER_PORT": "server.port",
+    "wsgi.url_scheme": "url.scheme",
+}
+
 
 def wsgi_decoding_dance(s, charset="utf-8", errors="replace"):
     # type: (str, str, str) -> str
@@ -120,7 +129,9 @@ def __call__(self, environ, start_response):
                                 name=DEFAULT_TRANSACTION_NAME,
                                 source=TRANSACTION_SOURCE_ROUTE,
                                 origin=self.span_origin,
-                                custom_sampling_context={"wsgi_environ": environ},
+                                attributes=_prepopulate_attributes(
+                                    environ, self.use_x_forwarded_for
+                                ),
                             )
                             if should_trace
                             else nullcontext()
@@ -309,3 +320,29 @@ def event_processor(event, hint):
         return event
 
     return event_processor
+
+
+def _prepopulate_attributes(wsgi_environ, use_x_forwarded_for=False):
+    """Extract span attributes from the WSGI environment."""
+    attributes = {}
+
+    for property, attr in ENVIRON_TO_ATTRIBUTE.items():
+        if wsgi_environ.get(property) is not None:
+            attributes[attr] = wsgi_environ[property]
+
+    if wsgi_environ.get("SERVER_PROTOCOL") is not None:
+        try:
+            proto, version = wsgi_environ["SERVER_PROTOCOL"].split("/")
+            attributes["network.protocol.name"] = proto
+            attributes["network.protocol.version"] = version
+        except Exception:
+            attributes["network.protocol.name"] = wsgi_environ["SERVER_PROTOCOL"]
+
+    try:
+        url = get_request_url(wsgi_environ, use_x_forwarded_for)
+        query = wsgi_environ.get("QUERY_STRING")
+        attributes["url.full"] = f"{url}?{query}"
+    except Exception:
+        pass
+
+    return attributes

tests/integrations/asgi/test_asgi.py

@@ -728,6 +728,7 @@ async def test_asgi_scope_in_traces_sampler(sentry_init, asgi3_app):
     def dummy_traces_sampler(sampling_context):
         assert sampling_context["url.path"] == "/test"
         assert sampling_context["url.scheme"] == "http"
+        assert sampling_context["url.query"] == "hello=there"
         assert sampling_context["url.full"] == "/test?hello=there"
         assert sampling_context["http.request.method"] == "GET"
         assert sampling_context["network.protocol.version"] == "1.1"

tests/integrations/wsgi/test_wsgi.py

@@ -334,25 +334,21 @@ def app(environ, start_response):
         start_response("200 OK", [])
         return ["Go get the ball! Good dog!"]
 
-    traces_sampler = mock.Mock(return_value=True)
+    def traces_sampler(sampling_context):
+        assert sampling_context["http.request.method"] == "GET"
+        assert sampling_context["url.path"] == "/dogs/are/great/"
+        assert sampling_context["url.query"] == "cats=too"
+        assert sampling_context["url.scheme"] == "http"
+        assert (
+            sampling_context["url.full"] == "http://localhost/dogs/are/great/?cats=too"
+        )
+        return True
+
     sentry_init(send_default_pii=True, traces_sampler=traces_sampler)
     app = SentryWsgiMiddleware(app)
     client = Client(app)
 
-    client.get("/dogs/are/great/")
-
-    traces_sampler.assert_any_call(
-        DictionaryContaining(
-            {
-                "wsgi_environ": DictionaryContaining(
-                    {
-                        "PATH_INFO": "/dogs/are/great/",
-                        "REQUEST_METHOD": "GET",
-                    },
-                ),
-            }
-        )
-    )
+    client.get("/dogs/are/great/?cats=too")
 
 
 def test_session_mode_defaults_to_request_mode_in_wsgi_handler(