feat(openai): Set system instruction attribute for Responses API (#5376)

Set the system instruction attribute on `gen_ai.chat` spans generated when using the Responses API with the `OpenAIIntegration`. Extract instructions from string input, content strings and parts-style content lists.

Author: alexander-alderman-webb

sentry_sdk/integrations/openai.py

@@ -215,6 +215,24 @@ def _get_system_instructions_completions(
     ]
 
 
+def _is_system_instruction_responses(message: "ResponseInputItemParam") -> bool:
+    if not isinstance(message, dict) or not message.get("role") == "system":
+        return False
+
+    return "type" not in message or message["type"] == "message"
+
+
+def _get_system_instructions_responses(
+    messages: "Union[str, ResponseInputParam]",
+) -> "list[ResponseInputItemParam]":
+    if not isinstance(messages, list):
+        return []
+
+    return [
+        message for message in messages if _is_system_instruction_responses(message)
+    ]
+
+
 def _transform_system_instructions(
     system_instructions: "list[ChatCompletionSystemMessageParam]",
 ) -> "list[TextPart]":
@@ -289,17 +307,80 @@ def _set_responses_api_input_data(
     kwargs: "dict[str, Any]",
     integration: "OpenAIIntegration",
 ) -> None:
-    messages: "Optional[Union[ResponseInputParam, list[str]]]" = _get_input_messages(
-        kwargs
-    )
+    explicit_instructions: "Union[Optional[str], Omit]" = kwargs.get("instructions")
+    messages: "Optional[Union[str, ResponseInputParam]]" = kwargs.get("input")
+
+    if not should_send_default_pii() or not integration.include_prompts:
+        set_data_normalized(span, SPANDATA.GEN_AI_OPERATION_NAME, "responses")
+        _commmon_set_input_data(span, kwargs)
+        return
 
     if (
-        messages is not None
-        and len(messages) > 0
-        and should_send_default_pii()
-        and integration.include_prompts
+        messages is None
+        and explicit_instructions is not None
+        and _is_given(explicit_instructions)
     ):
-        normalized_messages = normalize_message_roles(messages)  # type: ignore
+        set_data_normalized(
+            span,
+            SPANDATA.GEN_AI_SYSTEM_INSTRUCTIONS,
+            [
+                {
+                    "type": "text",
+                    "content": explicit_instructions,
+                }
+            ],
+            unpack=False,
+        )
+
+        set_data_normalized(span, SPANDATA.GEN_AI_OPERATION_NAME, "responses")
+        _commmon_set_input_data(span, kwargs)
+        return
+
+    if messages is None:
+        set_data_normalized(span, SPANDATA.GEN_AI_OPERATION_NAME, "responses")
+        _commmon_set_input_data(span, kwargs)
+        return
+
+    instructions_text_parts: "list[TextPart]" = []
+    if explicit_instructions is not None and _is_given(explicit_instructions):
+        instructions_text_parts.append(
+            {
+                "type": "text",
+                "content": explicit_instructions,
+            }
+        )
+
+    system_instructions = _get_system_instructions_responses(messages)
+    # Deliberate use of function accepting completions API type because
+    # of shared structure FOR THIS PURPOSE ONLY.
+    instructions_text_parts += _transform_system_instructions(system_instructions)
+
+    if len(instructions_text_parts) > 0:
+        set_data_normalized(
+            span,
+            SPANDATA.GEN_AI_SYSTEM_INSTRUCTIONS,
+            instructions_text_parts,
+            unpack=False,
+        )
+
+    if isinstance(messages, str):
+        normalized_messages = normalize_message_roles([messages])  # type: ignore
+        scope = sentry_sdk.get_current_scope()
+        messages_data = truncate_and_annotate_messages(normalized_messages, span, scope)
+        if messages_data is not None:
+            set_data_normalized(
+                span, SPANDATA.GEN_AI_REQUEST_MESSAGES, messages_data, unpack=False
+            )
+
+        set_data_normalized(span, SPANDATA.GEN_AI_OPERATION_NAME, "responses")
+        _commmon_set_input_data(span, kwargs)
+        return
+
+    non_system_messages = [
+        message for message in messages if not _is_system_instruction_responses(message)
+    ]
+    if len(non_system_messages) > 0:
+        normalized_messages = normalize_message_roles(non_system_messages)
         scope = sentry_sdk.get_current_scope()
         messages_data = truncate_and_annotate_messages(normalized_messages, span, scope)
         if messages_data is not None: