letta-client 0.1.276__tar.gz → 0.1.277__tar.gz

{letta_client-0.1.276 → letta_client-0.1.277}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: letta-client
-Version: 0.1.276
+Version: 0.1.277
 Summary: 
 Requires-Python: >=3.8,<4.0
 Classifier: Intended Audience :: Developers

{letta_client-0.1.276 → letta_client-0.1.277}/pyproject.toml

@@ -3,7 +3,7 @@ name = "letta-client"
 
 [tool.poetry]
 name = "letta-client"
-version = "0.1.276"
+version = "0.1.277"
 description = ""
 readme = "README.md"
 authors = []

{letta_client-0.1.276 → letta_client-0.1.277}/src/letta_client/__init__.py

@@ -370,6 +370,7 @@ from .client_side_access_tokens import (
 from .environment import LettaEnvironment
 from .groups import GroupCreateManagerConfig, GroupUpdateManagerConfig
 from .projects import ProjectsListResponse, ProjectsListResponseProjectsItem
+from .runs import LettaStreamingResponse
 from .steps import StepsListRequestFeedback
 from .templates import (
     TemplatesCreateTemplateResponse,
@@ -623,6 +624,7 @@ __all__ = [
     "LettaSerializeSchemasPydanticAgentSchemaToolSchema",
     "LettaStopReason",
     "LettaStreamingRequest",
+    "LettaStreamingResponse",
     "LettaUsageStatistics",
     "LettaUserMessageContentUnion",
     "ListMcpServersResponseValue",

{letta_client-0.1.276 → letta_client-0.1.277}/src/letta_client/core/client_wrapper.py

@@ -24,10 +24,10 @@ class BaseClientWrapper:
 
     def get_headers(self) -> typing.Dict[str, str]:
         headers: typing.Dict[str, str] = {
-            "User-Agent": "letta-client/0.1.276",
+            "User-Agent": "letta-client/0.1.277",
             "X-Fern-Language": "Python",
             "X-Fern-SDK-Name": "letta-client",
-            "X-Fern-SDK-Version": "0.1.276",
+            "X-Fern-SDK-Version": "0.1.277",
             **(self.get_custom_headers() or {}),
         }
         if self._project is not None:

{letta_client-0.1.276 → letta_client-0.1.277}/src/letta_client/runs/__init__.py

@@ -2,6 +2,7 @@
 
 # isort: skip_file
 
+from .types import LettaStreamingResponse
 from . import messages, steps, usage
 
-__all__ = ["messages", "steps", "usage"]
+__all__ = ["LettaStreamingResponse", "messages", "steps", "usage"]

{letta_client-0.1.276 → letta_client-0.1.277}/src/letta_client/runs/client.py

@@ -8,8 +8,12 @@ from ..types.run import Run
 from .messages.client import AsyncMessagesClient, MessagesClient
 from .raw_client import AsyncRawRunsClient, RawRunsClient
 from .steps.client import AsyncStepsClient, StepsClient
+from .types.letta_streaming_response import LettaStreamingResponse
 from .usage.client import AsyncUsageClient, UsageClient
 
+# this is used as the default value for optional parameters
+OMIT = typing.cast(typing.Any, ...)
+
 
 class RunsClient:
     def __init__(self, *, client_wrapper: SyncClientWrapper):
@@ -163,6 +167,65 @@ class RunsClient:
         _response = self._raw_client.delete(run_id, request_options=request_options)
         return _response.data
 
+    def stream(
+        self,
+        run_id: str,
+        *,
+        starting_after: typing.Optional[int] = OMIT,
+        include_pings: typing.Optional[bool] = OMIT,
+        poll_interval: typing.Optional[float] = OMIT,
+        batch_size: typing.Optional[int] = OMIT,
+        request_options: typing.Optional[RequestOptions] = None,
+    ) -> typing.Iterator[LettaStreamingResponse]:
+        """
+        Parameters
+        ----------
+        run_id : str
+
+        starting_after : typing.Optional[int]
+            Sequence id to use as a cursor for pagination. Response will start streaming after this chunk sequence id
+
+        include_pings : typing.Optional[bool]
+            Whether to include periodic keepalive ping messages in the stream to prevent connection timeouts.
+
+        poll_interval : typing.Optional[float]
+            Seconds to wait between polls when no new data.
+
+        batch_size : typing.Optional[int]
+            Number of entries to read per batch.
+
+        request_options : typing.Optional[RequestOptions]
+            Request-specific configuration.
+
+        Yields
+        ------
+        typing.Iterator[LettaStreamingResponse]
+            Successful response
+
+        Examples
+        --------
+        from letta_client import Letta
+
+        client = Letta(
+            project="YOUR_PROJECT",
+            token="YOUR_TOKEN",
+        )
+        response = client.runs.stream(
+            run_id="run_id",
+        )
+        for chunk in response:
+            yield chunk
+        """
+        with self._raw_client.stream(
+            run_id,
+            starting_after=starting_after,
+            include_pings=include_pings,
+            poll_interval=poll_interval,
+            batch_size=batch_size,
+            request_options=request_options,
+        ) as r:
+            yield from r.data
+
 
 class AsyncRunsClient:
     def __init__(self, *, client_wrapper: AsyncClientWrapper):
@@ -347,3 +410,71 @@ class AsyncRunsClient:
         """
         _response = await self._raw_client.delete(run_id, request_options=request_options)
         return _response.data
+
+    async def stream(
+        self,
+        run_id: str,
+        *,
+        starting_after: typing.Optional[int] = OMIT,
+        include_pings: typing.Optional[bool] = OMIT,
+        poll_interval: typing.Optional[float] = OMIT,
+        batch_size: typing.Optional[int] = OMIT,
+        request_options: typing.Optional[RequestOptions] = None,
+    ) -> typing.AsyncIterator[LettaStreamingResponse]:
+        """
+        Parameters
+        ----------
+        run_id : str
+
+        starting_after : typing.Optional[int]
+            Sequence id to use as a cursor for pagination. Response will start streaming after this chunk sequence id
+
+        include_pings : typing.Optional[bool]
+            Whether to include periodic keepalive ping messages in the stream to prevent connection timeouts.
+
+        poll_interval : typing.Optional[float]
+            Seconds to wait between polls when no new data.
+
+        batch_size : typing.Optional[int]
+            Number of entries to read per batch.
+
+        request_options : typing.Optional[RequestOptions]
+            Request-specific configuration.
+
+        Yields
+        ------
+        typing.AsyncIterator[LettaStreamingResponse]
+            Successful response
+
+        Examples
+        --------
+        import asyncio
+
+        from letta_client import AsyncLetta
+
+        client = AsyncLetta(
+            project="YOUR_PROJECT",
+            token="YOUR_TOKEN",
+        )
+
+
+        async def main() -> None:
+            response = await client.runs.stream(
+                run_id="run_id",
+            )
+            async for chunk in response:
+                yield chunk
+
+
+        asyncio.run(main())
+        """
+        async with self._raw_client.stream(
+            run_id,
+            starting_after=starting_after,
+            include_pings=include_pings,
+            poll_interval=poll_interval,
+            batch_size=batch_size,
+            request_options=request_options,
+        ) as r:
+            async for _chunk in r.data:
+                yield _chunk

{letta_client-0.1.276 → letta_client-0.1.277}/src/letta_client/runs/raw_client.py

@@ -1,8 +1,11 @@
 # This file was auto-generated by Fern from our API Definition.
 
+import contextlib
+import json
 import typing
 from json.decoder import JSONDecodeError
 
+import httpx_sse
 from ..core.api_error import ApiError
 from ..core.client_wrapper import AsyncClientWrapper, SyncClientWrapper
 from ..core.http_response import AsyncHttpResponse, HttpResponse
@@ -12,6 +15,10 @@ from ..core.unchecked_base_model import construct_type
 from ..errors.unprocessable_entity_error import UnprocessableEntityError
 from ..types.http_validation_error import HttpValidationError
 from ..types.run import Run
+from .types.letta_streaming_response import LettaStreamingResponse
+
+# this is used as the default value for optional parameters
+OMIT = typing.cast(typing.Any, ...)
 
 
 class RawRunsClient:
@@ -224,6 +231,101 @@ class RawRunsClient:
             raise ApiError(status_code=_response.status_code, headers=dict(_response.headers), body=_response.text)
         raise ApiError(status_code=_response.status_code, headers=dict(_response.headers), body=_response_json)
 
+    @contextlib.contextmanager
+    def stream(
+        self,
+        run_id: str,
+        *,
+        starting_after: typing.Optional[int] = OMIT,
+        include_pings: typing.Optional[bool] = OMIT,
+        poll_interval: typing.Optional[float] = OMIT,
+        batch_size: typing.Optional[int] = OMIT,
+        request_options: typing.Optional[RequestOptions] = None,
+    ) -> typing.Iterator[HttpResponse[typing.Iterator[LettaStreamingResponse]]]:
+        """
+        Parameters
+        ----------
+        run_id : str
+
+        starting_after : typing.Optional[int]
+            Sequence id to use as a cursor for pagination. Response will start streaming after this chunk sequence id
+
+        include_pings : typing.Optional[bool]
+            Whether to include periodic keepalive ping messages in the stream to prevent connection timeouts.
+
+        poll_interval : typing.Optional[float]
+            Seconds to wait between polls when no new data.
+
+        batch_size : typing.Optional[int]
+            Number of entries to read per batch.
+
+        request_options : typing.Optional[RequestOptions]
+            Request-specific configuration.
+
+        Yields
+        ------
+        typing.Iterator[HttpResponse[typing.Iterator[LettaStreamingResponse]]]
+            Successful response
+        """
+        with self._client_wrapper.httpx_client.stream(
+            f"v1/runs/{jsonable_encoder(run_id)}/stream",
+            method="POST",
+            json={
+                "starting_after": starting_after,
+                "include_pings": include_pings,
+                "poll_interval": poll_interval,
+                "batch_size": batch_size,
+            },
+            headers={
+                "content-type": "application/json",
+            },
+            request_options=request_options,
+            omit=OMIT,
+        ) as _response:
+
+            def _stream() -> HttpResponse[typing.Iterator[LettaStreamingResponse]]:
+                try:
+                    if 200 <= _response.status_code < 300:
+
+                        def _iter():
+                            _event_source = httpx_sse.EventSource(_response)
+                            for _sse in _event_source.iter_sse():
+                                if _sse.data == None:
+                                    return
+                                try:
+                                    yield typing.cast(
+                                        LettaStreamingResponse,
+                                        construct_type(
+                                            type_=LettaStreamingResponse,  # type: ignore
+                                            object_=json.loads(_sse.data),
+                                        ),
+                                    )
+                                except Exception:
+                                    pass
+                            return
+
+                        return HttpResponse(response=_response, data=_iter())
+                    _response.read()
+                    if _response.status_code == 422:
+                        raise UnprocessableEntityError(
+                            headers=dict(_response.headers),
+                            body=typing.cast(
+                                HttpValidationError,
+                                construct_type(
+                                    type_=HttpValidationError,  # type: ignore
+                                    object_=_response.json(),
+                                ),
+                            ),
+                        )
+                    _response_json = _response.json()
+                except JSONDecodeError:
+                    raise ApiError(
+                        status_code=_response.status_code, headers=dict(_response.headers), body=_response.text
+                    )
+                raise ApiError(status_code=_response.status_code, headers=dict(_response.headers), body=_response_json)
+
+            yield _stream()
+
 
 class AsyncRawRunsClient:
     def __init__(self, *, client_wrapper: AsyncClientWrapper):
@@ -438,3 +540,98 @@ class AsyncRawRunsClient:
         except JSONDecodeError:
             raise ApiError(status_code=_response.status_code, headers=dict(_response.headers), body=_response.text)
         raise ApiError(status_code=_response.status_code, headers=dict(_response.headers), body=_response_json)
+
+    @contextlib.asynccontextmanager
+    async def stream(
+        self,
+        run_id: str,
+        *,
+        starting_after: typing.Optional[int] = OMIT,
+        include_pings: typing.Optional[bool] = OMIT,
+        poll_interval: typing.Optional[float] = OMIT,
+        batch_size: typing.Optional[int] = OMIT,
+        request_options: typing.Optional[RequestOptions] = None,
+    ) -> typing.AsyncIterator[AsyncHttpResponse[typing.AsyncIterator[LettaStreamingResponse]]]:
+        """
+        Parameters
+        ----------
+        run_id : str
+
+        starting_after : typing.Optional[int]
+            Sequence id to use as a cursor for pagination. Response will start streaming after this chunk sequence id
+
+        include_pings : typing.Optional[bool]
+            Whether to include periodic keepalive ping messages in the stream to prevent connection timeouts.
+
+        poll_interval : typing.Optional[float]
+            Seconds to wait between polls when no new data.
+
+        batch_size : typing.Optional[int]
+            Number of entries to read per batch.
+
+        request_options : typing.Optional[RequestOptions]
+            Request-specific configuration.
+
+        Yields
+        ------
+        typing.AsyncIterator[AsyncHttpResponse[typing.AsyncIterator[LettaStreamingResponse]]]
+            Successful response
+        """
+        async with self._client_wrapper.httpx_client.stream(
+            f"v1/runs/{jsonable_encoder(run_id)}/stream",
+            method="POST",
+            json={
+                "starting_after": starting_after,
+                "include_pings": include_pings,
+                "poll_interval": poll_interval,
+                "batch_size": batch_size,
+            },
+            headers={
+                "content-type": "application/json",
+            },
+            request_options=request_options,
+            omit=OMIT,
+        ) as _response:
+
+            async def _stream() -> AsyncHttpResponse[typing.AsyncIterator[LettaStreamingResponse]]:
+                try:
+                    if 200 <= _response.status_code < 300:
+
+                        async def _iter():
+                            _event_source = httpx_sse.EventSource(_response)
+                            async for _sse in _event_source.aiter_sse():
+                                if _sse.data == None:
+                                    return
+                                try:
+                                    yield typing.cast(
+                                        LettaStreamingResponse,
+                                        construct_type(
+                                            type_=LettaStreamingResponse,  # type: ignore
+                                            object_=json.loads(_sse.data),
+                                        ),
+                                    )
+                                except Exception:
+                                    pass
+                            return
+
+                        return AsyncHttpResponse(response=_response, data=_iter())
+                    await _response.aread()
+                    if _response.status_code == 422:
+                        raise UnprocessableEntityError(
+                            headers=dict(_response.headers),
+                            body=typing.cast(
+                                HttpValidationError,
+                                construct_type(
+                                    type_=HttpValidationError,  # type: ignore
+                                    object_=_response.json(),
+                                ),
+                            ),
+                        )
+                    _response_json = _response.json()
+                except JSONDecodeError:
+                    raise ApiError(
+                        status_code=_response.status_code, headers=dict(_response.headers), body=_response.text
+                    )
+                raise ApiError(status_code=_response.status_code, headers=dict(_response.headers), body=_response_json)
+
+            yield await _stream()

letta_client-0.1.277/src/letta_client/runs/types/__init__.py

@@ -0,0 +1,7 @@
+# This file was auto-generated by Fern from our API Definition.
+
+# isort: skip_file
+
+from .letta_streaming_response import LettaStreamingResponse
+
+__all__ = ["LettaStreamingResponse"]

letta_client-0.1.277/src/letta_client/runs/types/letta_streaming_response.py

@@ -0,0 +1,25 @@
+# This file was auto-generated by Fern from our API Definition.
+
+import typing
+
+from ...types.assistant_message import AssistantMessage
+from ...types.letta_ping import LettaPing
+from ...types.letta_stop_reason import LettaStopReason
+from ...types.letta_usage_statistics import LettaUsageStatistics
+from ...types.reasoning_message import ReasoningMessage
+from ...types.system_message import SystemMessage
+from ...types.tool_call_message import ToolCallMessage
+from ...types.tool_return_message import ToolReturnMessage
+from ...types.user_message import UserMessage
+
+LettaStreamingResponse = typing.Union[
+    SystemMessage,
+    UserMessage,
+    ReasoningMessage,
+    ToolCallMessage,
+    ToolReturnMessage,
+    AssistantMessage,
+    LettaPing,
+    LettaStopReason,
+    LettaUsageStatistics,
+]