langchain_oci bugfix: include ToolMessages in chat history (#8)

* include ToolMessages in chat history

* fix linting / format issues

* use self.tool_call and self.tool_result

Author: streamnsight

libs/oci/langchain_oci/__init__.py

@@ -1,14 +1,16 @@
 # Copyright (c) 2025 Oracle and/or its affiliates.
 # Licensed under the Universal Permissive License v 1.0 as shown at https://oss.oracle.com/licenses/upl/
 
-from langchain_oci.chat_models.oci_generative_ai import ChatOCIGenAI
 from langchain_oci.chat_models.oci_data_science import (
     ChatOCIModelDeployment,
     ChatOCIModelDeploymentTGI,
-    ChatOCIModelDeploymentVLLM
+    ChatOCIModelDeploymentVLLM,
+)
+from langchain_oci.chat_models.oci_generative_ai import ChatOCIGenAI
+from langchain_oci.embeddings.oci_data_science_model_deployment_endpoint import (
+    OCIModelDeploymentEndpointEmbeddings,
 )
 from langchain_oci.embeddings.oci_generative_ai import OCIGenAIEmbeddings
-from langchain_oci.embeddings.oci_data_science_model_deployment_endpoint import OCIModelDeploymentEndpointEmbeddings
 from langchain_oci.llms.oci_data_science_model_deployment_endpoint import (
     BaseOCIModelDeployment,
     OCIModelDeploymentLLM,

libs/oci/langchain_oci/chat_models/__init__.py

@@ -4,8 +4,13 @@
 from langchain_oci.chat_models.oci_data_science import (
     ChatOCIModelDeployment,
     ChatOCIModelDeploymentTGI,
-    ChatOCIModelDeploymentVLLM
+    ChatOCIModelDeploymentVLLM,
 )
 from langchain_oci.chat_models.oci_generative_ai import ChatOCIGenAI
 
-__all__ = ["ChatOCIGenAI", "ChatOCIModelDeployment", "ChatOCIModelDeploymentTGI", "ChatOCIModelDeploymentVLLM"]
+__all__ = [
+    "ChatOCIGenAI",
+    "ChatOCIModelDeployment",
+    "ChatOCIModelDeploymentTGI",
+    "ChatOCIModelDeploymentVLLM",
+]

libs/oci/langchain_oci/chat_models/oci_generative_ai.py

@@ -362,6 +362,19 @@ def messages_to_oci_params(
                         message=msg_content, tool_calls=tool_calls
                     )
                 )
+            elif isinstance(msg, ToolMessage):
+                oci_chat_history.append(
+                    self.oci_chat_message[self.get_role(msg)](
+                        tool_results=[
+                            self.oci_tool_result(
+                                call=self.oci_tool_call(
+                                    name=msg.name, parameters={}
+                                ),
+                                outputs=[{"output": msg.content}],
+                            )
+                        ],
+                    )
+                )
 
         # Process current turn messages in reverse order until a HumanMessage
         current_turn = []

libs/oci/langchain_oci/embeddings/__init__.py

@@ -1,7 +1,9 @@
 # Copyright (c) 2025 Oracle and/or its affiliates.
 # Licensed under the Universal Permissive License v 1.0 as shown at https://oss.oracle.com/licenses/upl/
 
-from langchain_oci.embeddings.oci_data_science_model_deployment_endpoint import OCIModelDeploymentEndpointEmbeddings
+from langchain_oci.embeddings.oci_data_science_model_deployment_endpoint import (
+    OCIModelDeploymentEndpointEmbeddings,
+)
 from langchain_oci.embeddings.oci_generative_ai import OCIGenAIEmbeddings
 
 __all__ = ["OCIModelDeploymentEndpointEmbeddings", "OCIGenAIEmbeddings"]

libs/oci/langchain_oci/embeddings/oci_data_science_model_deployment_endpoint.py

@@ -1,13 +1,13 @@
 # Copyright (c) 2025 Oracle and/or its affiliates.
 # Licensed under the Universal Permissive License v 1.0 as shown at https://oss.oracle.com/licenses/upl/
 
+from typing import Any, Callable, Dict, List, Mapping, Optional
+
+import requests
 from langchain_core.embeddings import Embeddings
 from langchain_core.language_models.llms import create_base_retry_decorator
 from langchain_core.utils import get_from_dict_or_env
 from pydantic import BaseModel, Field, model_validator
-import requests
-from typing import Any, Callable, Dict, List, Mapping, Optional
-
 
 DEFAULT_HEADER = {
     "Content-Type": "application/json",
@@ -39,7 +39,7 @@ class OCIModelDeploymentEndpointEmbeddings(BaseModel, Embeddings):
             embeddings = OCIModelDeploymentEndpointEmbeddings(
                 endpoint="https://modeldeployment.us-ashburn-1.oci.customer-oci.com/<md_ocid>/predict",
             )
-    """ # noqa: E501
+    """  # noqa: E501
 
     auth: dict = Field(default_factory=dict, exclude=True)
     """ADS auth dictionary for OCI authentication:

libs/oci/langchain_oci/embeddings/oci_generative_ai.py

@@ -143,13 +143,13 @@ def make_security_token_signer(oci_config):  # type: ignore[no-untyped-def]
                     oci_config=client_kwargs["config"]
                 )
             elif values["auth_type"] == OCIAuthType(3).name:
-                client_kwargs[
-                    "signer"
-                ] = oci.auth.signers.InstancePrincipalsSecurityTokenSigner()
+                client_kwargs["signer"] = (
+                    oci.auth.signers.InstancePrincipalsSecurityTokenSigner()
+                )
             elif values["auth_type"] == OCIAuthType(4).name:
-                client_kwargs[
-                    "signer"
-                ] = oci.auth.signers.get_resource_principals_signer()
+                client_kwargs["signer"] = (
+                    oci.auth.signers.get_resource_principals_signer()
+                )
             else:
                 raise ValueError("Please provide valid value to auth_type")
 

libs/oci/langchain_oci/llms/oci_data_science_model_deployment_endpoint.py

@@ -3,10 +3,10 @@
 
 """LLM for OCI data science model deployment endpoint."""
 
-from contextlib import asynccontextmanager
 import json
 import logging
 import traceback
+from contextlib import asynccontextmanager
 from typing import (
     Any,
     AsyncGenerator,
@@ -793,6 +793,7 @@ class OCIModelDeploymentTGI(OCIModelDeploymentLLM):
             )
 
     """
+
     max_tokens: int = 256
     """Denotes the number of tokens to predict per generation."""
 
@@ -943,6 +944,7 @@ class OCIModelDeploymentVLLM(OCIModelDeploymentLLM):
             )
 
     """
+
     max_tokens: int = 256
     """Denotes the number of tokens to predict per generation."""
 

libs/oci/langchain_oci/llms/oci_generative_ai.py

@@ -22,12 +22,10 @@
 class Provider(ABC):
     @property
     @abstractmethod
-    def stop_sequence_key(self) -> str:
-        ...
+    def stop_sequence_key(self) -> str: ...
 
     @abstractmethod
-    def completion_response_to_text(self, response: Any) -> str:
-        ...
+    def completion_response_to_text(self, response: Any) -> str: ...
 
 
 class CohereProvider(Provider):
@@ -159,13 +157,13 @@ def make_security_token_signer(oci_config):  # type: ignore[no-untyped-def]
                     oci_config=client_kwargs["config"]
                 )
             elif values["auth_type"] == OCIAuthType(3).name:
-                client_kwargs[
-                    "signer"
-                ] = oci.auth.signers.InstancePrincipalsSecurityTokenSigner()
+                client_kwargs["signer"] = (
+                    oci.auth.signers.InstancePrincipalsSecurityTokenSigner()
+                )
             elif values["auth_type"] == OCIAuthType(4).name:
-                client_kwargs[
-                    "signer"
-                ] = oci.auth.signers.get_resource_principals_signer()
+                client_kwargs["signer"] = (
+                    oci.auth.signers.get_resource_principals_signer()
+                )
             else:
                 raise ValueError(
                     "Please provide valid value to auth_type, "

libs/oci/tests/unit_tests/embeddings/test_oci_model_deployment_endpoint.py

@@ -1,8 +1,9 @@
 """Test OCI Data Science Model Deployment Endpoint."""
 
-import responses
 import pytest
+import responses
 from pytest_mock import MockerFixture
+
 from langchain_oci.embeddings import OCIModelDeploymentEndpointEmbeddings
 
 
@@ -17,11 +18,7 @@ def test_embedding_call(mocker: MockerFixture) -> None:
         responses.POST,
         endpoint,
         json={
-            "data": [
-                {
-                    "embedding": expected_output
-                }
-            ],
+            "data": [{"embedding": expected_output}],
         },
         status=200,
     )