fix(openai): Always set gen_ai.response.streaming for Responses

Author: alexander-alderman-webb

sentry_sdk/integrations/openai.py

@@ -237,10 +237,6 @@ def _set_responses_api_input_data(
     if model is not None:
         span.set_data(SPANDATA.GEN_AI_REQUEST_MODEL, model)
 
-    stream = kwargs.get("stream")
-    if stream is not None and _is_given(stream):
-        span.set_data(SPANDATA.GEN_AI_RESPONSE_STREAMING, stream)
-
     max_tokens = kwargs.get("max_output_tokens")
     if max_tokens is not None and _is_given(max_tokens):
         span.set_data(SPANDATA.GEN_AI_REQUEST_MAX_TOKENS, max_tokens)
@@ -1097,12 +1093,14 @@ def _new_responses_create_common(f: "Any", *args: "Any", **kwargs: "Any") -> "An
     span.__enter__()
 
     span.set_data(SPANDATA.GEN_AI_SYSTEM, "openai")
+    is_streaming_response = kwargs.get("stream", False)
+    span.set_data(SPANDATA.GEN_AI_RESPONSE_STREAMING, is_streaming_response)
+
     _set_responses_api_input_data(span, kwargs, integration)
 
     start_time = time.perf_counter()
     response = yield f, args, kwargs
 
-    is_streaming_response = kwargs.get("stream", False)
     if is_streaming_response:
         _set_streaming_responses_api_output_data(
             span, response, kwargs, integration, start_time, finish_span=True

tests/integrations/openai/test_openai.py

@@ -1784,6 +1784,7 @@ def test_ai_client_span_responses_api_no_pii(sentry_init, capture_events):
         "gen_ai.request.top_p": 0.9,
         "gen_ai.request.model": "gpt-4o",
         "gen_ai.response.model": "response-model-id",
+        "gen_ai.response.streaming": False,
         "gen_ai.system": "openai",
         "gen_ai.usage.input_tokens": 20,
         "gen_ai.usage.input_tokens.cached": 5,
@@ -1901,6 +1902,7 @@ def test_ai_client_span_responses_api(
         "gen_ai.request.top_p": 0.9,
         "gen_ai.system": "openai",
         "gen_ai.response.model": "response-model-id",
+        "gen_ai.response.streaming": False,
         "gen_ai.usage.input_tokens": 20,
         "gen_ai.usage.input_tokens.cached": 5,
         "gen_ai.usage.output_tokens": 10,
@@ -2210,6 +2212,7 @@ async def test_ai_client_span_responses_async_api(
         "gen_ai.request.messages": '["How do I check if a Python object is an instance of a class?"]',
         "gen_ai.request.model": "gpt-4o",
         "gen_ai.response.model": "response-model-id",
+        "gen_ai.response.streaming": False,
         "gen_ai.system": "openai",
         "gen_ai.usage.input_tokens": 20,
         "gen_ai.usage.input_tokens.cached": 5,