# File generated from our OpenAPI spec by Castiron. See CONTRIBUTING.md for details. from __future__ import annotations import json import logging from types import TracebackType from typing import TYPE_CHECKING, Any, Union, Callable, Iterator, Optional, Awaitable, cast from typing_extensions import AsyncIterator import httpx2 from pydantic import BaseModel from ..._types import Omit, Query, Headers, omit from ..._utils import is_given, path_template, maybe_transform, strip_not_given, async_maybe_transform from ..._compat import cached_property from ..._models import FinalRequestOptions, construct_type_unchecked from ..._resource import SyncAPIResource, AsyncAPIResource from ..._constants import DEFAULT_LIVE_SEND_QUEUE_MAX_BYTES from ...types.live import decision_item_create_event_param from ..._exceptions import OpenAIError, WebSocketConnectionClosedError from ..._send_queue import SendQueue from ..._base_client import _merge_mappings from ..._event_handler import EventHandlerRegistry from ..._websocket_connection import WebSocketConnectionBase, AsyncWebSocketConnectionBase from ...types.live.error_event import ErrorEvent from ...types.live.fork_client_event import ForkClientEvent from ...types.live.fork_server_event import ForkServerEvent from ...types.websocket_reconnection import ReconnectingEvent, ReconnectingOverrides from ...types.live.fork_client_event_param import ForkClientEventParam from ...types.websocket_connection_options import WebSocketConnectionOptions, SyncWebSocketConnectionOptions from ...types.live.fork_session_config_param import ForkSessionConfigParam from ...types.live.session_update_config_param import SessionUpdateConfigParam from ...types.responses.response_input_item_param import ResponseInputItemParam if TYPE_CHECKING: from websockets.sync.client import ClientConnection as WebSocketConnection from websockets.asyncio.client import ClientConnection as AsyncWebSocketConnection from ..._client import OpenAI, AsyncOpenAI __all__ = ["Forks", "AsyncForks"] log: logging.Logger = logging.getLogger(__name__) class Forks(SyncAPIResource): def connect( self, *, session_id: str, extra_query: Query = {}, extra_headers: Headers = {}, websocket_connection_options: SyncWebSocketConnectionOptions = {}, on_reconnecting: Callable[[ReconnectingEvent], ReconnectingOverrides | None] | None = None, max_retries: int = 5, initial_delay: float = 0.5, max_delay: float = 8.0, max_queue_size: int = DEFAULT_LIVE_SEND_QUEUE_MAX_BYTES, ) -> ForksConnectionManager: """Fork a stored Live session over WebSocket. Send session.start with an overrides object (which may be empty), then wait for session.started. The model is inherited. """ return ForksConnectionManager( client=self._client, extra_query=extra_query, extra_headers=extra_headers, websocket_connection_options=websocket_connection_options, on_reconnecting=on_reconnecting, max_retries=max_retries, initial_delay=initial_delay, max_delay=max_delay, max_queue_size=max_queue_size, session_id=session_id, ) class AsyncForks(AsyncAPIResource): def connect( self, *, session_id: str, extra_query: Query = {}, extra_headers: Headers = {}, websocket_connection_options: WebSocketConnectionOptions = {}, on_reconnecting: Callable[[ReconnectingEvent], ReconnectingOverrides | None] | None = None, max_retries: int = 5, initial_delay: float = 0.5, max_delay: float = 8.0, max_queue_size: int = DEFAULT_LIVE_SEND_QUEUE_MAX_BYTES, ) -> AsyncForksConnectionManager: """Fork a stored Live session over WebSocket. Send session.start with an overrides object (which may be empty), then wait for session.started. The model is inherited. """ return AsyncForksConnectionManager( client=self._client, extra_query=extra_query, extra_headers=extra_headers, websocket_connection_options=websocket_connection_options, on_reconnecting=on_reconnecting, max_retries=max_retries, initial_delay=initial_delay, max_delay=max_delay, max_queue_size=max_queue_size, session_id=session_id, ) class AsyncForksConnection(AsyncWebSocketConnectionBase): """Represents a live WebSocket connection to the Forks API""" session: AsyncForksSessionResource decision: AsyncForksDecisionResource response: AsyncForksResponseResource _connection: AsyncWebSocketConnection def __init__( self, connection: AsyncWebSocketConnection, *, make_ws: Callable[[Query, Headers], Awaitable[AsyncWebSocketConnection]] | None = None, on_reconnecting: Callable[[ReconnectingEvent], ReconnectingOverrides | None] | None = None, max_retries: int = 5, initial_delay: float = 0.5, max_delay: float = 8.0, extra_query: Query = {}, extra_headers: Headers = {}, send_queue: SendQueue | None = None, ) -> None: super().__init__( connection, make_ws=make_ws, on_reconnecting=on_reconnecting, max_retries=max_retries, initial_delay=initial_delay, max_delay=max_delay, extra_query=extra_query, extra_headers=extra_headers, send_queue=send_queue if send_queue is not None else SendQueue(max_bytes=DEFAULT_LIVE_SEND_QUEUE_MAX_BYTES), logger=log, ) self._event_handler_registry = EventHandlerRegistry(use_lock=False) self.session = AsyncForksSessionResource(self) self.decision = AsyncForksDecisionResource(self) self.response = AsyncForksResponseResource(self) async def __aiter__(self) -> AsyncIterator[ForkServerEvent]: """ An infinite-iterator that will continue to yield events until the connection is closed. """ from websockets.exceptions import ConnectionClosed, ConnectionClosedOK, ConnectionClosedError while True: try: yield await self.recv() except ConnectionClosed as exc: # Older SDK bases read close codes only from ConnectionClosedError. # Adapt the policy input; keep the original close for EOF and errors. reconnect_exc = ( ConnectionClosedError(exc.rcvd, exc.sent, exc.rcvd_then_sent) if isinstance(exc, ConnectionClosedOK) else exc ) if not await self._reconnect(reconnect_exc): unsent = self._send_queue.drain() if unsent: raise WebSocketConnectionClosedError( "WebSocket connection closed with unsent messages", unsent_messages=unsent, ) from exc if isinstance(exc, ConnectionClosedOK): return raise async def recv(self) -> ForkServerEvent: """ Receive the next message from the connection and parses it into a `ForkServerEvent` object. Canceling this method is safe. There's no risk of losing data. """ event = self.parse_event(await self.recv_bytes()) event_type = ( cast("dict[str, object]", event).get("type") if isinstance(cast(object, event), dict) else getattr(event, "type", None) ) # A successful upgrade can still be followed by an admission error. # Reset the budget only after receiving a non-error application event. if isinstance(event_type, str) and event_type and event_type != "error": self._reconnect_attempt = 0 return event async def recv_bytes(self) -> bytes: """Receive the next message from the connection as raw bytes. Canceling this method is safe. There's no risk of losing data. If you want to parse the message into a `ForkServerEvent` object like `.recv()` does, then you can call `.parse_event(data)`. """ message = await self._connection.recv(decode=False) log.debug("Received WebSocket message: %i bytes", len(message)) if self._reconnect_attempt: # Account for raw application progress without changing frame delivery. try: event_data: object = json.loads(message) except (ValueError, RecursionError): return message event_type = cast("dict[str, object]", event_data).get("type") if isinstance(event_data, dict) else None if isinstance(event_type, str) and event_type and event_type != "error": self._reconnect_attempt = 0 return message async def send(self, event: ForkClientEvent | ForkClientEventParam) -> None: data = ( event.to_json(use_api_names=True, exclude_unset=True) if isinstance(event, BaseModel) else json.dumps(await async_maybe_transform(event, ForkClientEventParam)) ) await AsyncWebSocketConnectionBase.send_raw(self, data) def parse_event(self, data: str | bytes) -> ForkServerEvent: """ Converts a raw `str` or `bytes` message into a `ForkServerEvent` object. This is helpful if you're using `.recv_bytes()`. """ return cast(ForkServerEvent, construct_type_unchecked(value=json.loads(data), type_=cast(Any, ForkServerEvent))) def on( self, event_type: str, handler: Callable[..., Any] | None = None ) -> Union[AsyncForksConnection, Callable[[Callable[..., Any]], Callable[..., Any]]]: """Adds the handler to the end of the handlers list for the given event type. No checks are made to see if the handler has already been added. Multiple calls passing the same combination of event type and handler will result in the handler being added, and called, multiple times. Can be used as a method (returns ``self`` for chaining):: connection.on("session.started", my_handler) Or as a decorator:: @connection.on("session.started") async def my_handler(event): ... """ if handler is not None: self._event_handler_registry.add(event_type, handler) return self def decorator(fn: Callable[..., Any]) -> Callable[..., Any]: self._event_handler_registry.add(event_type, fn) return fn return decorator def off(self, event_type: str, handler: Callable[..., Any]) -> AsyncForksConnection: """Remove a previously registered event handler.""" self._event_handler_registry.remove(event_type, handler) return self def once( self, event_type: str, handler: Callable[..., Any] | None = None ) -> Union[AsyncForksConnection, Callable[[Callable[..., Any]], Callable[..., Any]]]: """Register a one-time event handler. Automatically removed after first invocation. """ if handler is not None: self._event_handler_registry.add(event_type, handler, once=True) return self def decorator(fn: Callable[..., Any]) -> Callable[..., Any]: self._event_handler_registry.add(event_type, fn, once=True) return fn return decorator async def dispatch_events(self) -> None: """Run the event loop, dispatching received events to registered handlers. Blocks until the connection is closed. This is the push-based alternative to iterating with ``async for event in connection``. If an ``"error"`` event arrives and no handler is registered for ``"error"`` or ``"event"``, an ``OpenAIError`` is raised. """ import asyncio async for event in self: event_type = event.type specific = self._event_handler_registry.get_handlers(event_type) generic = self._event_handler_registry.get_handlers("event") if event_type == "error" and not specific and not generic: if isinstance(event, ErrorEvent): raise OpenAIError(f"WebSocket error: {event}") for handler in specific: result = handler(event) if asyncio.iscoroutine(result): await result for handler in generic: result = handler(event) if asyncio.iscoroutine(result): await result class AsyncForksConnectionManager: """ Context manager over a `AsyncForksConnection` that is returned by `live.forks.connect()` This context manager ensures that the connection will be closed when it exits. --- Note that if your application doesn't work well with the context manager approach then you can call the `.enter()` method directly to initiate a connection. **Warning**: You must remember to close the connection with `.close()`. ```py connection = await client.live.forks.connect(...).enter() # ... await connection.close() ``` """ def __init__( self, *, client: AsyncOpenAI, session_id: str, extra_query: Query, extra_headers: Headers, websocket_connection_options: WebSocketConnectionOptions, on_reconnecting: Callable[[ReconnectingEvent], ReconnectingOverrides | None] | None = None, max_retries: int = 5, initial_delay: float = 0.5, max_delay: float = 8.0, max_queue_size: int = DEFAULT_LIVE_SEND_QUEUE_MAX_BYTES, ) -> None: self.__client = client self.__session_id = session_id self.__connection: AsyncForksConnection | None = None self.__extra_query = extra_query self.__extra_headers = extra_headers self.__websocket_connection_options = websocket_connection_options self.__on_reconnecting = on_reconnecting self.__max_retries = max_retries self.__initial_delay = initial_delay self.__max_delay = max_delay self.__send_queue = SendQueue(max_bytes=max_queue_size) self.__event_handler_registry = EventHandlerRegistry(use_lock=False) def send(self, event: ForkClientEvent | ForkClientEventParam) -> None: """Queue a message to be sent when the connection is established. This can be called before entering the context manager. Queued messages are automatically sent once the WebSocket connection opens. """ data = ( event.to_json(use_api_names=True, exclude_unset=True) if isinstance(event, BaseModel) else json.dumps(maybe_transform(event, ForkClientEventParam)) ) self.__send_queue.enqueue(data) def on( self, event_type: str, handler: Callable[..., Any] | None = None ) -> Union[AsyncForksConnectionManager, Callable[[Callable[..., Any]], Callable[..., Any]]]: """Register an event handler before the connection is established. Handlers are transferred to the connection on enter. Supports the same method and decorator forms as ``AsyncForksConnection.on``. """ if handler is not None: self.__event_handler_registry.add(event_type, handler) return self def decorator(fn: Callable[..., Any]) -> Callable[..., Any]: self.__event_handler_registry.add(event_type, fn) return fn return decorator def off(self, event_type: str, handler: Callable[..., Any]) -> AsyncForksConnectionManager: """Remove a previously registered event handler.""" self.__event_handler_registry.remove(event_type, handler) return self def once( self, event_type: str, handler: Callable[..., Any] | None = None ) -> Union[AsyncForksConnectionManager, Callable[[Callable[..., Any]], Callable[..., Any]]]: """Register a one-time event handler before the connection is established.""" if handler is not None: self.__event_handler_registry.add(event_type, handler, once=True) return self def decorator(fn: Callable[..., Any]) -> Callable[..., Any]: self.__event_handler_registry.add(event_type, fn, once=True) return fn return decorator async def __aenter__(self) -> AsyncForksConnection: """ If your application doesn't work well with the context manager approach then you can call this method directly to initiate a connection. **Warning**: You must remember to close the connection with `.close()`. ```py connection = await client.live.forks.connect(...).enter() # ... await connection.close() ``` """ ws = await self._connect_ws(self.__extra_query, self.__extra_headers) try: self.__connection = AsyncForksConnection( ws, make_ws=self._connect_ws if self.__on_reconnecting is not None else None, on_reconnecting=self.__on_reconnecting, max_retries=self.__max_retries, initial_delay=self.__initial_delay, max_delay=self.__max_delay, extra_query=self.__extra_query, extra_headers=self.__extra_headers, send_queue=self.__send_queue, ) self.__event_handler_registry.merge_into(self.__connection._event_handler_registry) await self.__connection._flush_send_queue() except BaseException: self.__connection = None try: await ws.close() except BaseException: # Preserve the exception that prevented successful entry. pass raise return self.__connection enter = __aenter__ async def _connect_ws(self, extra_query: Query, extra_headers: Headers) -> AsyncWebSocketConnection: try: from ...lib._websocket import _WebSocketConnect as connect except ImportError as exc: raise OpenAIError("You need to install `openai[realtime]` to use this method") from exc url = self._prepare_url() try: # Older runtimes do not declare this optional helper. query_params = getattr(self.__client, "_websocket_query_params") # noqa: B009 except AttributeError: query_params = self.__client._websocket_query_params_fallback url = url.copy_with( params=query_params( url, extra_query, ), ) url = url.copy_with(scheme={"http": "ws", "https": "wss"}.get(url.scheme, url.scheme)) options = await self.__client._prepare_options( FinalRequestOptions.construct( method="get", url=str(url), headers=dict(extra_headers), security={"bearer_auth": True}, ) ) url = self.__client._prepare_url(options.url).copy_merge_params( self.__client.qs.stringify(cast(Any, options.params)) ) url = url.copy_with(scheme={"http": "ws", "https": "wss"}.get(url.scheme, url.scheme)) headers = { key.lower(): (key, value) for header_set in ( self.__client.auth_headers, {}, self.__client.default_headers, options.headers if is_given(options.headers) else {}, ) for key, value in header_set.items() } log.debug("Connecting to WebSocket API") if self.__websocket_connection_options: log.debug("Custom WebSocket connection options provided") websocket_options = self.__websocket_connection_options.copy() websocket_options.setdefault("max_size", None) if websocket_options.get("max_size") is None: websocket_options.setdefault("compression", None) return await connect( str(url), user_agent_header=None, additional_headers=_merge_mappings(dict(headers.values()), {}), **websocket_options, ) def _prepare_url(self) -> httpx2.URL: if self.__client.websocket_base_url is not None: base_url = httpx2.URL(self.__client.websocket_base_url) else: scheme = self.__client._base_url.scheme ws_scheme = "ws" if scheme == "http" else "wss" base_url = self.__client._base_url.copy_with(scheme=ws_scheme) path, separator, query = base_url.raw_path.partition(b"?") merge_raw_path = ( path.rstrip(b"/") + path_template("/live/sessions/{session_id}/fork", session_id=self.__session_id).encode("utf-8") + separator + query ) return base_url.copy_with(raw_path=merge_raw_path) async def __aexit__( self, exc_type: type[BaseException] | None, exc: BaseException | None, exc_tb: TracebackType | None ) -> None: if self.__connection is not None: try: await self.__connection.close() except Exception: if exc is None: raise class ForksConnection(WebSocketConnectionBase): """Represents a live WebSocket connection to the Forks API""" session: ForksSessionResource decision: ForksDecisionResource response: ForksResponseResource _connection: WebSocketConnection def __init__( self, connection: WebSocketConnection, *, make_ws: Callable[[Query, Headers], WebSocketConnection] | None = None, on_reconnecting: Callable[[ReconnectingEvent], ReconnectingOverrides | None] | None = None, max_retries: int = 5, initial_delay: float = 0.5, max_delay: float = 8.0, extra_query: Query = {}, extra_headers: Headers = {}, send_queue: SendQueue | None = None, ) -> None: super().__init__( connection, make_ws=make_ws, on_reconnecting=on_reconnecting, max_retries=max_retries, initial_delay=initial_delay, max_delay=max_delay, extra_query=extra_query, extra_headers=extra_headers, send_queue=send_queue if send_queue is not None else SendQueue(max_bytes=DEFAULT_LIVE_SEND_QUEUE_MAX_BYTES), logger=log, ) self._event_handler_registry = EventHandlerRegistry(use_lock=True) self.session = ForksSessionResource(self) self.decision = ForksDecisionResource(self) self.response = ForksResponseResource(self) def __iter__(self) -> Iterator[ForkServerEvent]: """ An infinite-iterator that will continue to yield events until the connection is closed. """ from websockets.exceptions import ConnectionClosed, ConnectionClosedOK, ConnectionClosedError while True: try: yield self.recv() except ConnectionClosed as exc: # Older SDK bases read close codes only from ConnectionClosedError. # Adapt the policy input; keep the original close for EOF and errors. reconnect_exc = ( ConnectionClosedError(exc.rcvd, exc.sent, exc.rcvd_then_sent) if isinstance(exc, ConnectionClosedOK) else exc ) if not self._reconnect(reconnect_exc): unsent = self._send_queue.drain() if unsent: raise WebSocketConnectionClosedError( "WebSocket connection closed with unsent messages", unsent_messages=unsent, ) from exc if isinstance(exc, ConnectionClosedOK): return raise def recv(self) -> ForkServerEvent: """ Receive the next message from the connection and parses it into a `ForkServerEvent` object. Canceling this method is safe. There's no risk of losing data. """ event = self.parse_event(self.recv_bytes()) event_type = ( cast("dict[str, object]", event).get("type") if isinstance(cast(object, event), dict) else getattr(event, "type", None) ) # A successful upgrade can still be followed by an admission error. # Reset the budget only after receiving a non-error application event. if isinstance(event_type, str) and event_type and event_type != "error": self._reconnect_attempt = 0 return event def recv_bytes(self) -> bytes: """Receive the next message from the connection as raw bytes. Canceling this method is safe. There's no risk of losing data. If you want to parse the message into a `ForkServerEvent` object like `.recv()` does, then you can call `.parse_event(data)`. """ from ...lib._websocket import _recv_bytes message = _recv_bytes(self._connection) log.debug("Received WebSocket message: %i bytes", len(message)) if self._reconnect_attempt: # Account for raw application progress without changing frame delivery. try: event_data: object = json.loads(message) except (ValueError, RecursionError): return message event_type = cast("dict[str, object]", event_data).get("type") if isinstance(event_data, dict) else None if isinstance(event_type, str) and event_type and event_type != "error": self._reconnect_attempt = 0 return message def send(self, event: ForkClientEvent | ForkClientEventParam) -> None: data = ( event.to_json(use_api_names=True, exclude_unset=True) if isinstance(event, BaseModel) else json.dumps(maybe_transform(event, ForkClientEventParam)) ) WebSocketConnectionBase.send_raw(self, data) def parse_event(self, data: str | bytes) -> ForkServerEvent: """ Converts a raw `str` or `bytes` message into a `ForkServerEvent` object. This is helpful if you're using `.recv_bytes()`. """ return cast(ForkServerEvent, construct_type_unchecked(value=json.loads(data), type_=cast(Any, ForkServerEvent))) def on( self, event_type: str, handler: Callable[..., Any] | None = None ) -> Union[ForksConnection, Callable[[Callable[..., Any]], Callable[..., Any]]]: """Adds the handler to the end of the handlers list for the given event type. No checks are made to see if the handler has already been added. Multiple calls passing the same combination of event type and handler will result in the handler being added, and called, multiple times. Can be used as a method (returns ``self`` for chaining):: connection.on("session.started", my_handler) Or as a decorator:: @connection.on("session.started") def my_handler(event): ... """ if handler is not None: self._event_handler_registry.add(event_type, handler) return self def decorator(fn: Callable[..., Any]) -> Callable[..., Any]: self._event_handler_registry.add(event_type, fn) return fn return decorator def off(self, event_type: str, handler: Callable[..., Any]) -> ForksConnection: """Remove a previously registered event handler.""" self._event_handler_registry.remove(event_type, handler) return self def once( self, event_type: str, handler: Callable[..., Any] | None = None ) -> Union[ForksConnection, Callable[[Callable[..., Any]], Callable[..., Any]]]: """Register a one-time event handler. Automatically removed after first invocation. """ if handler is not None: self._event_handler_registry.add(event_type, handler, once=True) return self def decorator(fn: Callable[..., Any]) -> Callable[..., Any]: self._event_handler_registry.add(event_type, fn, once=True) return fn return decorator def dispatch_events(self) -> None: """Run the event loop, dispatching received events to registered handlers. Blocks the current thread until the connection is closed. This is the push-based alternative to iterating with ``for event in connection``. If an ``"error"`` event arrives and no handler is registered for ``"error"`` or ``"event"``, an ``OpenAIError`` is raised. """ for event in self: event_type = event.type specific = self._event_handler_registry.get_handlers(event_type) generic = self._event_handler_registry.get_handlers("event") if event_type == "error" and not specific and not generic: if isinstance(event, ErrorEvent): raise OpenAIError(f"WebSocket error: {event}") for handler in specific: handler(event) for handler in generic: handler(event) class ForksConnectionManager: """ Context manager over a `ForksConnection` that is returned by `live.forks.connect()` This context manager ensures that the connection will be closed when it exits. --- Note that if your application doesn't work well with the context manager approach then you can call the `.enter()` method directly to initiate a connection. **Warning**: You must remember to close the connection with `.close()`. ```py connection = client.live.forks.connect(...).enter() # ... connection.close() ``` """ def __init__( self, *, client: OpenAI, session_id: str, extra_query: Query, extra_headers: Headers, websocket_connection_options: SyncWebSocketConnectionOptions, on_reconnecting: Callable[[ReconnectingEvent], ReconnectingOverrides | None] | None = None, max_retries: int = 5, initial_delay: float = 0.5, max_delay: float = 8.0, max_queue_size: int = DEFAULT_LIVE_SEND_QUEUE_MAX_BYTES, ) -> None: self.__client = client self.__session_id = session_id self.__connection: ForksConnection | None = None self.__extra_query = extra_query self.__extra_headers = extra_headers self.__websocket_connection_options = websocket_connection_options self.__on_reconnecting = on_reconnecting self.__max_retries = max_retries self.__initial_delay = initial_delay self.__max_delay = max_delay self.__send_queue = SendQueue(max_bytes=max_queue_size) self.__event_handler_registry = EventHandlerRegistry(use_lock=True) def send(self, event: ForkClientEvent | ForkClientEventParam) -> None: """Queue a message to be sent when the connection is established. This can be called before entering the context manager. Queued messages are automatically sent once the WebSocket connection opens. """ data = ( event.to_json(use_api_names=True, exclude_unset=True) if isinstance(event, BaseModel) else json.dumps(maybe_transform(event, ForkClientEventParam)) ) self.__send_queue.enqueue(data) def on( self, event_type: str, handler: Callable[..., Any] | None = None ) -> Union[ForksConnectionManager, Callable[[Callable[..., Any]], Callable[..., Any]]]: """Register an event handler before the connection is established. Handlers are transferred to the connection on enter. Supports the same method and decorator forms as ``ForksConnection.on``. """ if handler is not None: self.__event_handler_registry.add(event_type, handler) return self def decorator(fn: Callable[..., Any]) -> Callable[..., Any]: self.__event_handler_registry.add(event_type, fn) return fn return decorator def off(self, event_type: str, handler: Callable[..., Any]) -> ForksConnectionManager: """Remove a previously registered event handler.""" self.__event_handler_registry.remove(event_type, handler) return self def once( self, event_type: str, handler: Callable[..., Any] | None = None ) -> Union[ForksConnectionManager, Callable[[Callable[..., Any]], Callable[..., Any]]]: """Register a one-time event handler before the connection is established.""" if handler is not None: self.__event_handler_registry.add(event_type, handler, once=True) return self def decorator(fn: Callable[..., Any]) -> Callable[..., Any]: self.__event_handler_registry.add(event_type, fn, once=True) return fn return decorator def __enter__(self) -> ForksConnection: """ If your application doesn't work well with the context manager approach then you can call this method directly to initiate a connection. **Warning**: You must remember to close the connection with `.close()`. ```py connection = client.live.forks.connect(...).enter() # ... connection.close() ``` """ ws = self._connect_ws(self.__extra_query, self.__extra_headers) try: self.__connection = ForksConnection( ws, make_ws=self._connect_ws if self.__on_reconnecting is not None else None, on_reconnecting=self.__on_reconnecting, max_retries=self.__max_retries, initial_delay=self.__initial_delay, max_delay=self.__max_delay, extra_query=self.__extra_query, extra_headers=self.__extra_headers, send_queue=self.__send_queue, ) self.__event_handler_registry.merge_into(self.__connection._event_handler_registry) self.__connection._flush_send_queue() except BaseException: self.__connection = None try: ws.close() except BaseException: # Preserve the exception that prevented successful entry. pass raise return self.__connection enter = __enter__ def _connect_ws(self, extra_query: Query, extra_headers: Headers) -> WebSocketConnection: try: from websockets.sync.client import connect except ImportError as exc: raise OpenAIError("You need to install `openai[realtime]` to use this method") from exc url = self._prepare_url() try: # Older runtimes do not declare this optional helper. query_params = getattr(self.__client, "_websocket_query_params") # noqa: B009 except AttributeError: query_params = self.__client._websocket_query_params_fallback url = url.copy_with( params=query_params( url, extra_query, ), ) url = url.copy_with(scheme={"http": "ws", "https": "wss"}.get(url.scheme, url.scheme)) options = self.__client._prepare_options( FinalRequestOptions.construct( method="get", url=str(url), headers=dict(extra_headers), security={"bearer_auth": True}, ) ) url = self.__client._prepare_url(options.url).copy_merge_params( self.__client.qs.stringify(cast(Any, options.params)) ) url = url.copy_with(scheme={"http": "ws", "https": "wss"}.get(url.scheme, url.scheme)) headers = { key.lower(): (key, value) for header_set in ( self.__client.auth_headers, {}, self.__client.default_headers, options.headers if is_given(options.headers) else {}, ) for key, value in header_set.items() } log.debug("Connecting to WebSocket API") if self.__websocket_connection_options: log.debug("Custom WebSocket connection options provided") websocket_options = self.__websocket_connection_options.copy() websocket_options.setdefault("max_size", None) if websocket_options.get("max_size") is None: websocket_options.setdefault("compression", None) return connect( str(url), user_agent_header=None, additional_headers=_merge_mappings(dict(headers.values()), {}), **websocket_options, ) def _prepare_url(self) -> httpx2.URL: if self.__client.websocket_base_url is not None: base_url = httpx2.URL(self.__client.websocket_base_url) else: scheme = self.__client._base_url.scheme ws_scheme = "ws" if scheme == "http" else "wss" base_url = self.__client._base_url.copy_with(scheme=ws_scheme) path, separator, query = base_url.raw_path.partition(b"?") merge_raw_path = ( path.rstrip(b"/") + path_template("/live/sessions/{session_id}/fork", session_id=self.__session_id).encode("utf-8") + separator + query ) return base_url.copy_with(raw_path=merge_raw_path) def __exit__( self, exc_type: type[BaseException] | None, exc: BaseException | None, exc_tb: TracebackType | None ) -> None: if self.__connection is not None: try: self.__connection.close() except Exception: if exc is None: raise class BaseForksConnectionResource: def __init__(self, connection: ForksConnection) -> None: self._connection = connection class ForksSessionResource(BaseForksConnectionResource): @cached_property def commentary(self) -> ForksSessionCommentaryResource: return ForksSessionCommentaryResource(self._connection) @cached_property def input_audio(self) -> ForksSessionInputAudioResource: return ForksSessionInputAudioResource(self._connection) @cached_property def instructions(self) -> ForksSessionInstructionsResource: return ForksSessionInstructionsResource(self._connection) @cached_property def output_audio(self) -> ForksSessionOutputAudioResource: return ForksSessionOutputAudioResource(self._connection) @cached_property def thinking(self) -> ForksSessionThinkingResource: return ForksSessionThinkingResource(self._connection) def start(self, *, session: ForkSessionConfigParam, event_id: Optional[str] | Omit = omit) -> None: """ Start a Live session after connecting to a stored session’s fork WebSocket. Send an empty `session` object to use the stored configuration. """ self._connection.send( cast( ForkClientEventParam, strip_not_given({"type": "session.start", "session": session, "event_id": event_id}), ) ) def close(self, *, event_id: Optional[str] | Omit = omit) -> None: """Request that the Live session close. The terminal `session.closed` event contains the close reason and final usage. """ self._connection.send( cast(ForkClientEventParam, strip_not_given({"type": "session.close", "event_id": event_id})) ) def update(self, *, session: SessionUpdateConfigParam, event_id: Optional[str] | Omit = omit) -> None: """ Update the delegation settings or transcript triggers of an active Live session. The server acknowledges accepted changes with `session.updated`. """ self._connection.send( cast( ForkClientEventParam, strip_not_given({"type": "session.update", "session": session, "event_id": event_id}), ) ) class ForksSessionCommentaryResource(BaseForksConnectionResource): def append(self, *, content: str, delegation_id: Optional[str], event_id: Optional[str] | Omit = omit) -> None: """ Provide context the Live model can communicate to the user, optionally for an existing client delegation. Rejected while an audio fixture is active. """ self._connection.send( cast( ForkClientEventParam, strip_not_given( { "type": "session.commentary.append", "content": content, "delegation_id": delegation_id, "event_id": event_id, } ), ) ) class ForksSessionInputAudioResource(BaseForksConnectionResource): def append(self, *, audio: str, event_id: Optional[str] | Omit = omit) -> None: """Send audio to a Live session over its primary WebSocket. WebRTC and SIP sessions send audio over their media transport. """ self._connection.send( cast( ForkClientEventParam, strip_not_given({"type": "session.input_audio.append", "audio": audio, "event_id": event_id}), ) ) def mute(self, *, event_id: Optional[str] | Omit = omit) -> None: """Mute audio input to the Live model without closing the session. The server acknowledges with `session.input_audio.muted`. """ self._connection.send( cast(ForkClientEventParam, strip_not_given({"type": "session.input_audio.mute", "event_id": event_id})) ) def unmute(self, *, event_id: Optional[str] | Omit = omit) -> None: """Resume audio input to a Live model after muting it. The server acknowledges with `session.input_audio.unmuted`. An active audio fixture still pauses inference until playback finishes. """ self._connection.send( cast(ForkClientEventParam, strip_not_given({"type": "session.input_audio.unmute", "event_id": event_id})) ) class ForksSessionInstructionsResource(BaseForksConnectionResource): def append(self, *, content: str, delegation_id: Optional[str], event_id: Optional[str] | Omit = omit) -> None: """Append developer instructions to the Live conversation while it is running, optionally associating them with an existing client delegation. Rejected while an audio fixture is active. These instructions remain in conversation history through compaction and are included in future backend delegations. For external audio, describe content already delivered so the assistant does not repeat it. """ self._connection.send( cast( ForkClientEventParam, strip_not_given( { "type": "session.instructions.append", "content": content, "delegation_id": delegation_id, "event_id": event_id, } ), ) ) class ForksSessionOutputAudioResource(BaseForksConnectionResource): @cached_property def fixture(self) -> ForksSessionOutputAudioFixtureResource: return ForksSessionOutputAudioFixtureResource(self._connection) class ForksSessionThinkingResource(BaseForksConnectionResource): def append(self, *, content: str, delegation_id: Optional[str], event_id: Optional[str] | Omit = omit) -> None: """ Provide silent reasoning or progress context to the Live model, optionally for an existing client delegation. Rejected while an audio fixture is active. """ self._connection.send( cast( ForkClientEventParam, strip_not_given( { "type": "session.thinking.append", "content": content, "delegation_id": delegation_id, "event_id": event_id, } ), ) ) class ForksSessionOutputAudioFixtureResource(BaseForksConnectionResource): def play(self, *, audio: str, event_id: Optional[str] | Omit = omit) -> None: """Play one complete audio fixture through the session output. Send this event over the primary WebSocket or a server sideband. Audio must be raw mono PCM16 little-endian at 24 kHz, regardless of the session audio format, and at most 30 seconds. The entire payload is validated before model speech is interrupted and input is muted. Only one fixture can be active. fixture.playing acknowledges acceptance and fixture.played reports server mixer drain. These notifications are not exact transport-media boundaries. Input arriving during playback is dropped and not replayed. """ self._connection.send( cast( ForkClientEventParam, strip_not_given({"type": "session.output_audio.fixture.play", "audio": audio, "event_id": event_id}), ) ) class ForksDecisionResource(BaseForksConnectionResource): @cached_property def item(self) -> ForksDecisionItemResource: return ForksDecisionItemResource(self._connection) def create(self, *, event_id: str) -> None: """ Evaluate the configured questions using the current conversation and Decision inputs. Reusing an event_id is rejected for the lifetime of the session. """ self._connection.send({"type": "decision.create", "event_id": event_id}) class ForksResponseResource(BaseForksConnectionResource): @cached_property def item(self) -> ForksResponseItemResource: return ForksResponseItemResource(self._connection) def create(self, *, event_id: Optional[str] | Omit = omit) -> None: """ Request a response from the Live session’s Responses backend, or continue a delegated response waiting for tool results. Requires Responses delegation. """ self._connection.send( cast(ForkClientEventParam, strip_not_given({"type": "response.create", "event_id": event_id})) ) class ForksDecisionItemResource(BaseForksConnectionResource): def delete(self, *, item_id: str, event_id: Optional[str] | Omit = omit) -> None: """ Delete an input item from future decisions without starting an evaluation or changing Live voice context. """ self._connection.send( cast( ForkClientEventParam, strip_not_given({"type": "decision.item.delete", "item_id": item_id, "event_id": event_id}), ) ) def create( self, *, item: decision_item_create_event_param.ItemUnion, event_id: Optional[str] | Omit = omit ) -> None: """ Append an input item for future decisions without starting an evaluation or changing Live voice context. """ self._connection.send( cast( ForkClientEventParam, strip_not_given({"type": "decision.item.create", "item": item, "event_id": event_id}), ) ) class ForksResponseItemResource(BaseForksConnectionResource): def create(self, *, item: ResponseInputItemParam, event_id: Optional[str] | Omit = omit) -> None: """Add an input item to the Live session’s Responses backend. Requires Responses delegation; use `response.create` to request a response. """ self._connection.send( cast( ForkClientEventParam, strip_not_given({"type": "response.item.create", "item": item, "event_id": event_id}), ) ) class BaseAsyncForksConnectionResource: def __init__(self, connection: AsyncForksConnection) -> None: self._connection = connection class AsyncForksSessionResource(BaseAsyncForksConnectionResource): @cached_property def commentary(self) -> AsyncForksSessionCommentaryResource: return AsyncForksSessionCommentaryResource(self._connection) @cached_property def input_audio(self) -> AsyncForksSessionInputAudioResource: return AsyncForksSessionInputAudioResource(self._connection) @cached_property def instructions(self) -> AsyncForksSessionInstructionsResource: return AsyncForksSessionInstructionsResource(self._connection) @cached_property def output_audio(self) -> AsyncForksSessionOutputAudioResource: return AsyncForksSessionOutputAudioResource(self._connection) @cached_property def thinking(self) -> AsyncForksSessionThinkingResource: return AsyncForksSessionThinkingResource(self._connection) async def start(self, *, session: ForkSessionConfigParam, event_id: Optional[str] | Omit = omit) -> None: """ Start a Live session after connecting to a stored session’s fork WebSocket. Send an empty `session` object to use the stored configuration. """ await self._connection.send( cast( ForkClientEventParam, strip_not_given({"type": "session.start", "session": session, "event_id": event_id}), ) ) async def close(self, *, event_id: Optional[str] | Omit = omit) -> None: """Request that the Live session close. The terminal `session.closed` event contains the close reason and final usage. """ await self._connection.send( cast(ForkClientEventParam, strip_not_given({"type": "session.close", "event_id": event_id})) ) async def update(self, *, session: SessionUpdateConfigParam, event_id: Optional[str] | Omit = omit) -> None: """ Update the delegation settings or transcript triggers of an active Live session. The server acknowledges accepted changes with `session.updated`. """ await self._connection.send( cast( ForkClientEventParam, strip_not_given({"type": "session.update", "session": session, "event_id": event_id}), ) ) class AsyncForksSessionCommentaryResource(BaseAsyncForksConnectionResource): async def append( self, *, content: str, delegation_id: Optional[str], event_id: Optional[str] | Omit = omit ) -> None: """ Provide context the Live model can communicate to the user, optionally for an existing client delegation. Rejected while an audio fixture is active. """ await self._connection.send( cast( ForkClientEventParam, strip_not_given( { "type": "session.commentary.append", "content": content, "delegation_id": delegation_id, "event_id": event_id, } ), ) ) class AsyncForksSessionInputAudioResource(BaseAsyncForksConnectionResource): async def append(self, *, audio: str, event_id: Optional[str] | Omit = omit) -> None: """Send audio to a Live session over its primary WebSocket. WebRTC and SIP sessions send audio over their media transport. """ await self._connection.send( cast( ForkClientEventParam, strip_not_given({"type": "session.input_audio.append", "audio": audio, "event_id": event_id}), ) ) async def mute(self, *, event_id: Optional[str] | Omit = omit) -> None: """Mute audio input to the Live model without closing the session. The server acknowledges with `session.input_audio.muted`. """ await self._connection.send( cast(ForkClientEventParam, strip_not_given({"type": "session.input_audio.mute", "event_id": event_id})) ) async def unmute(self, *, event_id: Optional[str] | Omit = omit) -> None: """Resume audio input to a Live model after muting it. The server acknowledges with `session.input_audio.unmuted`. An active audio fixture still pauses inference until playback finishes. """ await self._connection.send( cast(ForkClientEventParam, strip_not_given({"type": "session.input_audio.unmute", "event_id": event_id})) ) class AsyncForksSessionInstructionsResource(BaseAsyncForksConnectionResource): async def append( self, *, content: str, delegation_id: Optional[str], event_id: Optional[str] | Omit = omit ) -> None: """Append developer instructions to the Live conversation while it is running, optionally associating them with an existing client delegation. Rejected while an audio fixture is active. These instructions remain in conversation history through compaction and are included in future backend delegations. For external audio, describe content already delivered so the assistant does not repeat it. """ await self._connection.send( cast( ForkClientEventParam, strip_not_given( { "type": "session.instructions.append", "content": content, "delegation_id": delegation_id, "event_id": event_id, } ), ) ) class AsyncForksSessionOutputAudioResource(BaseAsyncForksConnectionResource): @cached_property def fixture(self) -> AsyncForksSessionOutputAudioFixtureResource: return AsyncForksSessionOutputAudioFixtureResource(self._connection) class AsyncForksSessionThinkingResource(BaseAsyncForksConnectionResource): async def append( self, *, content: str, delegation_id: Optional[str], event_id: Optional[str] | Omit = omit ) -> None: """ Provide silent reasoning or progress context to the Live model, optionally for an existing client delegation. Rejected while an audio fixture is active. """ await self._connection.send( cast( ForkClientEventParam, strip_not_given( { "type": "session.thinking.append", "content": content, "delegation_id": delegation_id, "event_id": event_id, } ), ) ) class AsyncForksSessionOutputAudioFixtureResource(BaseAsyncForksConnectionResource): async def play(self, *, audio: str, event_id: Optional[str] | Omit = omit) -> None: """Play one complete audio fixture through the session output. Send this event over the primary WebSocket or a server sideband. Audio must be raw mono PCM16 little-endian at 24 kHz, regardless of the session audio format, and at most 30 seconds. The entire payload is validated before model speech is interrupted and input is muted. Only one fixture can be active. fixture.playing acknowledges acceptance and fixture.played reports server mixer drain. These notifications are not exact transport-media boundaries. Input arriving during playback is dropped and not replayed. """ await self._connection.send( cast( ForkClientEventParam, strip_not_given({"type": "session.output_audio.fixture.play", "audio": audio, "event_id": event_id}), ) ) class AsyncForksDecisionResource(BaseAsyncForksConnectionResource): @cached_property def item(self) -> AsyncForksDecisionItemResource: return AsyncForksDecisionItemResource(self._connection) async def create(self, *, event_id: str) -> None: """ Evaluate the configured questions using the current conversation and Decision inputs. Reusing an event_id is rejected for the lifetime of the session. """ await self._connection.send({"type": "decision.create", "event_id": event_id}) class AsyncForksResponseResource(BaseAsyncForksConnectionResource): @cached_property def item(self) -> AsyncForksResponseItemResource: return AsyncForksResponseItemResource(self._connection) async def create(self, *, event_id: Optional[str] | Omit = omit) -> None: """ Request a response from the Live session’s Responses backend, or continue a delegated response waiting for tool results. Requires Responses delegation. """ await self._connection.send( cast(ForkClientEventParam, strip_not_given({"type": "response.create", "event_id": event_id})) ) class AsyncForksDecisionItemResource(BaseAsyncForksConnectionResource): async def delete(self, *, item_id: str, event_id: Optional[str] | Omit = omit) -> None: """ Delete an input item from future decisions without starting an evaluation or changing Live voice context. """ await self._connection.send( cast( ForkClientEventParam, strip_not_given({"type": "decision.item.delete", "item_id": item_id, "event_id": event_id}), ) ) async def create( self, *, item: decision_item_create_event_param.ItemUnion, event_id: Optional[str] | Omit = omit ) -> None: """ Append an input item for future decisions without starting an evaluation or changing Live voice context. """ await self._connection.send( cast( ForkClientEventParam, strip_not_given({"type": "decision.item.create", "item": item, "event_id": event_id}), ) ) class AsyncForksResponseItemResource(BaseAsyncForksConnectionResource): async def create(self, *, item: ResponseInputItemParam, event_id: Optional[str] | Omit = omit) -> None: """Add an input item to the Live session’s Responses backend. Requires Responses delegation; use `response.create` to request a response. """ await self._connection.send( cast( ForkClientEventParam, strip_not_given({"type": "response.item.create", "item": item, "event_id": event_id}), ) )