feat(api): support Responses WebSocket warmup and local parsing

Castiron-Internal-PR: openai/openai-python-internal#387
Castiron-Source-SHA: 011966e4e1c8ef5879fda79b8bbde5d79794b760
Castiron-Public-Base-SHA: 68b962c

Author: markstuart-oai

.castiron.stats.yml

@@ -1,7 +1,7 @@
 schema_version: 1
-generation_id: e511c488-31b9-4833-baf5-cfb49bbba7ac
-openapi_spec_hash: d57efec7c76a156447c9ab739b446dc1
-openapi_transformed_spec_hash: b34c897c8cec867ec4ac8e22727c5975
+generation_id: 66f75554-d35a-43fb-a349-c03a542189d2
+openapi_spec_hash: b006a5de3d55cecef746265fa2d4d7c9
+openapi_transformed_spec_hash: d5d0c68e913851f8de7b3e6d2cd9a03b
 config_hash: 8da38e8d8f706786c2d8586691168746
-codegen_sha: e46410d5a9751f068b9afce57ecf25abda75de09
-codegen_hash: 33536449fa4898b45941d5fa4b5a958f88e2701a4b85fae21c372b75614198cd
+codegen_sha: 36e52c095cd7ec03ea1e65043d0fdd2ec06adca3
+codegen_hash: 18c5a7e1999ee5ccbb262fe37132309a27d93ca84f08b834602e5e58f7345e67

api_reference/openapi.transformed.yml

@@ -68571,6 +68571,12 @@ components:
               `stream_id` controls routing; `previous_response_id` controls
               conversation lineage, so a new lane can fork from a response created
               on another lane.
+          generate:
+            type: boolean
+            description: |
+              Whether to generate output for this response. Defaults to true when omitted.
+              Set to false to prepare request state without model output; the response
+              ID can be used as previous_response_id on a later turn.
         required:
         - type
       - $ref: '#/components/schemas/CreateResponse'
@@ -111494,6 +111500,12 @@ components:
               `stream_id` controls routing; `previous_response_id` controls
               conversation lineage, so a new lane can fork from a response created
               on another lane.
+          generate:
+            type: boolean
+            description: |
+              Whether to generate output for this response. Defaults to true when omitted.
+              Set to false to prepare request state without model output; the response
+              ID can be used as previous_response_id on a later turn.
         required:
         - type
       - $ref: '#/components/schemas/BetaCreateResponse'

scripts/castiron/generated-files.json

@@ -379,7 +379,7 @@
     },
     "src/openai/resources/beta/responses/responses.py": {
       "mode": "100644",
-      "sha256": "043ff25a6a7f078bb0b11f6368ee136d7f3bb371c65360c8eb608afaf86de023"
+      "sha256": "505f230e5616690d534818b4b23db7fea448fa7c9ae2901f8b46bfcc1283b629"
     },
     "src/openai/resources/beta/threads/__init__.py": {
       "mode": "100644",
@@ -559,19 +559,19 @@
     },
     "src/openai/resources/live/forks.py": {
       "mode": "100644",
-      "sha256": "eda42fc4ed92675d23eb62905f154a5b115567c158dd4b0fe5cf3d0824e2bd9d"
+      "sha256": "7fe78c144680df54628773683f9dd5b06c5241d0218b4f4c745890ddae225c92"
     },
     "src/openai/resources/live/live.py": {
       "mode": "100644",
-      "sha256": "a42b9133f64a4421eafbb1d0d10a9cddcefdc872270518f7bdf61ce192265985"
+      "sha256": "86be3e074d011a750fa52a4d216a7e87ee5c8e82f6ccec00102c03a539199882"
     },
     "src/openai/resources/live/sessions.py": {
       "mode": "100644",
       "sha256": "9ba2d59d9c7498c96ac2e7e48660c91468e0eeb5df881e77f123ee37985c3dfd"
     },
     "src/openai/resources/live/sideband.py": {
       "mode": "100644",
-      "sha256": "7ed377e577ce3b229692bd112059914700741cbff042a7a92cc2a581ec2ac1f8"
+      "sha256": "d19149c042afcd8c8efca3d7794092acc1c2f76a4bcb4f6f620547314a78e59b"
     },
     "src/openai/resources/models.py": {
       "mode": "100644",
@@ -599,7 +599,7 @@
     },
     "src/openai/resources/realtime/realtime.py": {
       "mode": "100644",
-      "sha256": "563d955afa8c0b023f4fcb353565db0b4348394f36e6a02659db6c2885286149"
+      "sha256": "d45ef5f2d1e25e6f8761592a9fb5739d7a344c6e64a4e7a4006418c16f05a540"
     },
     "src/openai/resources/realtime/translations/__init__.py": {
       "mode": "100644",
@@ -611,15 +611,15 @@
     },
     "src/openai/resources/realtime/translations/translations.py": {
       "mode": "100644",
-      "sha256": "8f45807c74c477a1d13df9979c6274a089c380004b943fcb6d9342b1a7ac3ef1"
+      "sha256": "4a6f072cea78d5afab3fe1c995b939ff8c79f449665346c729abb5c516c53f21"
     },
     "src/openai/resources/responses/__init__.py": {
       "mode": "100644",
       "sha256": "70034ef169af2cd1991aaaed26fb565d2a2d5a5b6b14309a72ab1f483f0aba15"
     },
     "src/openai/resources/responses/api.md": {
       "mode": "100644",
-      "sha256": "e7815b1278aa2780bb885ac5a249fe00859dff67d3899b708d25d2968e734785"
+      "sha256": "b635f37005fb7126cc69a7b69e3f14a9df135d9523271cc4407b1ffcb9bd5c8c"
     },
     "src/openai/resources/responses/input_items.py": {
       "mode": "100644",
@@ -631,7 +631,7 @@
     },
     "src/openai/resources/responses/responses.py": {
       "mode": "100644",
-      "sha256": "6e536e48d77d9ed583bd0c86a4f189b664e8636ae46a52a1c80fac28ffc3b4ec"
+      "sha256": "5fa683c532adcec0b0ecfc9e0108075923e3243eab319a79325a880b707acaee"
     },
     "src/openai/resources/safety/__init__.py": {
       "mode": "100644",
@@ -3087,11 +3087,11 @@
     },
     "src/openai/types/beta/beta_responses_client_event.py": {
       "mode": "100644",
-      "sha256": "939052a456df5a661ed334033a9151c18ed63dfba6c520fd92a7572d50efc091"
+      "sha256": "977d2d22d8468005d9dcf27c5e9c7e5c6430bb9dacd5258369022f1513431d3d"
     },
     "src/openai/types/beta/beta_responses_client_event_param.py": {
       "mode": "100644",
-      "sha256": "ba2e2d06061761345308086b4b1d2ce39f86db631c46db825c5bbc7f62f63ccb"
+      "sha256": "0ad4be42df7202197e66ed97ae6fbfbdcad2aeb4e92328064630497d18fb457c"
     },
     "src/openai/types/beta/beta_responses_server_event.py": {
       "mode": "100644",
@@ -6735,11 +6735,11 @@
     },
     "src/openai/types/responses/responses_client_event.py": {
       "mode": "100644",
-      "sha256": "77d64183c9c69abfd4300c8094fbd3bc9c3a89743b9efa83a5264bd3dc165961"
+      "sha256": "3993eec234392a96646408fb959860a86b2711ac53dc7e8ae1338c43ca6383b4"
     },
     "src/openai/types/responses/responses_client_event_param.py": {
       "mode": "100644",
-      "sha256": "95444697e02d702e9f568da381903d251b3d5c649de9c5d26662a1782431647b"
+      "sha256": "20cbab38d02cfd2506ad2104d09e2b3bfb0d2b0d8a55adec13882f3a15c9db2d"
     },
     "src/openai/types/responses/responses_server_event.py": {
       "mode": "100644",

src/openai/lib/_response_resources.py

@@ -40,6 +40,7 @@ def parse(
     resource: Responses,
     *,
     text_format: type[TextFormatT] | Omit = omit,
+    parse_as: type[TextFormatT] | None | Omit = omit,
     access_programs: response_create_params.AccessPrograms | Omit = omit,
     background: Optional[bool] | Omit = omit,
     context_management: Optional[Iterable[response_create_params.ContextManagement]] | Omit = omit,
@@ -94,7 +95,7 @@ def parse(
     def parser(raw_response: Response) -> ParsedResponse[TextFormatT]:
         return parse_response(
             input_tools=tools,
-            text_format=text_format,
+            text_format=text_format if not is_given(parse_as) else omit if parse_as is None else parse_as,
             response=raw_response,
         )
 
@@ -156,6 +157,7 @@ async def async_parse(
     resource: AsyncResponses,
     *,
     text_format: type[TextFormatT] | Omit = omit,
+    parse_as: type[TextFormatT] | None | Omit = omit,
     access_programs: response_create_params.AccessPrograms | Omit = omit,
     background: Optional[bool] | Omit = omit,
     context_management: Optional[Iterable[response_create_params.ContextManagement]] | Omit = omit,
@@ -210,7 +212,7 @@ async def async_parse(
     def parser(raw_response: Response) -> ParsedResponse[TextFormatT]:
         return parse_response(
             input_tools=tools,
-            text_format=text_format,
+            text_format=text_format if not is_given(parse_as) else omit if parse_as is None else parse_as,
             response=raw_response,
         )
 

src/openai/lib/responses_tool_runner.py

@@ -8,7 +8,7 @@
 import hashlib
 from copy import copy, deepcopy
 from time import monotonic
-from typing import TypeVar, Callable, Awaitable, cast
+from typing import Any, List, Union, TypeVar, Callable, Awaitable, cast, get_args
 from inspect import isawaitable, iscoroutine
 from functools import partial
 from contextvars import copy_context
@@ -24,7 +24,7 @@
 from .._types import Headers, omit
 from .._utils import is_given, transform
 from .browser import BrowserToolset, AsyncBrowserToolset, _legacy_state_url
-from .._compat import model_copy
+from .._compat import PYDANTIC_V1, model_copy
 from .._models import validate_type, construct_type_unchecked
 from .computer import (
     ComputerToolset,
@@ -55,6 +55,7 @@
 from ..types.responses.response_output_message import ResponseOutputMessage
 from ..types.responses.response_input_item_param import McpApprovalResponse
 from ..types.responses.response_function_call_output_item_list_param import ResponseFunctionCallOutputItemListParam
+from ..types.responses.response_configuration_update_item_param_param import ResponseConfigurationUpdateItemParamParam
 
 __all__ = [
     "RegisteredFunction",
@@ -78,6 +79,21 @@
 ToolOutput: TypeAlias = FunctionCallOutput | ComputerCallOutput | McpApprovalResponse
 _ModelT = TypeVar("_ModelT", bound=pydantic.BaseModel)
 
+# The runner rejects configuration_update histories. Pydantic v1 cannot even
+# construct that item's TypedDict validator because "schema" shadows BaseModel.
+# Omit only this unsupported branch from replay validation, not public inputs.
+_history_input_type: Any = ResponseInputParam
+if PYDANTIC_V1:
+    _history_input_type = List[
+        Union[
+            tuple(
+                item
+                for item in get_args(get_args(ResponseInputParam)[0])
+                if item is not ResponseConfigurationUpdateItemParamParam
+            )
+        ]
+    ]
+
 
 def _ledger_tools(tools: list[ToolParam]) -> list[ToolParam]:
     result: list[ToolParam] = []
@@ -507,7 +523,7 @@ def _history_calls(history: ResponseInputParam, tools: list[ToolParam], *, budge
     # the same action parser as dispatch, without changing the saved input.
     try:
         validate_type(
-            type_=ResponseInputParam,
+            type_=_history_input_type,
             value=[item for item in history if item.get("type") not in ("computer_call", "computer_call_output")],
         )
     except pydantic.ValidationError:

src/openai/resources/beta/responses/responses.py

@@ -5276,7 +5276,11 @@ async def __aexit__(
         self, exc_type: type[BaseException] | None, exc: BaseException | None, exc_tb: TracebackType | None
     ) -> None:
         if self.__connection is not None:
-            await self.__connection.close()
+            try:
+                await self.__connection.close()
+            except BaseException:
+                if exc is None:
+                    raise
 
 
 class ResponsesConnection(WebSocketConnectionBase):
@@ -5690,7 +5694,11 @@ def __exit__(
         self, exc_type: type[BaseException] | None, exc: BaseException | None, exc_tb: TracebackType | None
     ) -> None:
         if self.__connection is not None:
-            self.__connection.close()
+            try:
+                self.__connection.close()
+            except BaseException:
+                if exc is None:
+                    raise
 
 
 class BaseResponsesConnectionResource:
@@ -5707,6 +5715,7 @@ def create(
         context_management: Optional[Iterable[beta_responses_client_event_param.ResponseCreateContextManagement]]
         | Omit = omit,
         conversation: Optional[beta_responses_client_event_param.ResponseCreateConversation] | Omit = omit,
+        generate: bool | Omit = omit,
         include: Optional[List[BetaResponseIncludable]] | Omit = omit,
         input: Union[str, BetaResponseInputParam] | Omit = omit,
         instructions: Optional[str] | Omit = omit,
@@ -5873,6 +5882,7 @@ def create(
                         "background": background,
                         "context_management": context_management,
                         "conversation": conversation,
+                        "generate": generate,
                         "include": include,
                         "input": input,
                         "instructions": instructions,
@@ -5965,6 +5975,7 @@ async def create(
         context_management: Optional[Iterable[beta_responses_client_event_param.ResponseCreateContextManagement]]
         | Omit = omit,
         conversation: Optional[beta_responses_client_event_param.ResponseCreateConversation] | Omit = omit,
+        generate: bool | Omit = omit,
         include: Optional[List[BetaResponseIncludable]] | Omit = omit,
         input: Union[str, BetaResponseInputParam] | Omit = omit,
         instructions: Optional[str] | Omit = omit,
@@ -6131,6 +6142,7 @@ async def create(
                         "background": background,
                         "context_management": context_management,
                         "conversation": conversation,
+                        "generate": generate,
                         "include": include,
                         "input": input,
                         "instructions": instructions,

src/openai/resources/live/forks.py

@@ -531,7 +531,11 @@ async def __aexit__(
         self, exc_type: type[BaseException] | None, exc: BaseException | None, exc_tb: TracebackType | None
     ) -> None:
         if self.__connection is not None:
-            await self.__connection.close()
+            try:
+                await self.__connection.close()
+            except BaseException:
+                if exc is None:
+                    raise
 
 
 class ForksConnection(WebSocketConnectionBase):
@@ -953,7 +957,11 @@ def __exit__(
         self, exc_type: type[BaseException] | None, exc: BaseException | None, exc_tb: TracebackType | None
     ) -> None:
         if self.__connection is not None:
-            self.__connection.close()
+            try:
+                self.__connection.close()
+            except BaseException:
+                if exc is None:
+                    raise
 
 
 class BaseForksConnectionResource:

src/openai/resources/live/live.py

@@ -744,7 +744,11 @@ async def __aexit__(
         self, exc_type: type[BaseException] | None, exc: BaseException | None, exc_tb: TracebackType | None
     ) -> None:
         if self.__connection is not None:
-            await self.__connection.close()
+            try:
+                await self.__connection.close()
+            except BaseException:
+                if exc is None:
+                    raise
 
 
 class LiveConnection(WebSocketConnectionBase):
@@ -1158,7 +1162,11 @@ def __exit__(
         self, exc_type: type[BaseException] | None, exc: BaseException | None, exc_tb: TracebackType | None
     ) -> None:
         if self.__connection is not None:
-            self.__connection.close()
+            try:
+                self.__connection.close()
+            except BaseException:
+                if exc is None:
+                    raise
 
 
 class BaseLiveConnectionResource:

src/openai/resources/live/sideband.py

@@ -543,7 +543,11 @@ async def __aexit__(
         self, exc_type: type[BaseException] | None, exc: BaseException | None, exc_tb: TracebackType | None
     ) -> None:
         if self.__connection is not None:
-            await self.__connection.close()
+            try:
+                await self.__connection.close()
+            except BaseException:
+                if exc is None:
+                    raise
 
 
 class SidebandConnection(WebSocketConnectionBase):
@@ -974,7 +978,11 @@ def __exit__(
         self, exc_type: type[BaseException] | None, exc: BaseException | None, exc_tb: TracebackType | None
     ) -> None:
         if self.__connection is not None:
-            self.__connection.close()
+            try:
+                self.__connection.close()
+            except BaseException:
+                if exc is None:
+                    raise
 
 
 class BaseSidebandConnectionResource:

src/openai/resources/realtime/realtime.py

@@ -750,7 +750,11 @@ async def __aexit__(
         self, exc_type: type[BaseException] | None, exc: BaseException | None, exc_tb: TracebackType | None
     ) -> None:
         if self.__connection is not None:
-            await self.__connection.close()
+            try:
+                await self.__connection.close()
+            except BaseException:
+                if exc is None:
+                    raise
 
 
 class RealtimeConnection(WebSocketConnectionBase):
@@ -1199,7 +1203,11 @@ def __exit__(
         self, exc_type: type[BaseException] | None, exc: BaseException | None, exc_tb: TracebackType | None
     ) -> None:
         if self.__connection is not None:
-            self.__connection.close()
+            try:
+                self.__connection.close()
+            except BaseException:
+                if exc is None:
+                    raise
 
 
 class BaseRealtimeConnectionResource: