Python: [Breaking] removed pydantic from types and workflows (#917)

* removed pydantic from types

* fix test

* fix test

* fix tests

* fix assistants client

* Remove Pydantic usage from workflow code.

* updated pydantic removal

* updated lock and test fixes

* fix mypy

* updated build system

* updated chat client parsing

* fix broken test

---------

Co-authored-by: Evan Mattson <evan.mattson@microsoft.com>

python/samples/getting_started/agents/custom/custom_chat_client.py

@@ -113,7 +113,7 @@ class EchoingChatClient(BaseChatClient):
                     contents=[TextContent(text=char)],
                     role=Role.ASSISTANT,
                     response_id=f"echo-stream-resp-{random.randint(1000, 9999)}",
-                    ai_model_id="echo-model-v1",
+                    model_id="echo-model-v1",
                 )
                 await asyncio.sleep(0.05)
 

python/samples/getting_started/agents/openai/openai_responses_client_with_structured_output.py

@@ -32,7 +32,7 @@ async def non_streaming_example() -> None:
 
     # Access the structured output directly from the response value
     if result.value:
-        structured_data = result.value
+        structured_data: OutputStruct = result.value  # type: ignore
         print("Structured Output Agent (from result.value):")
         print(f"City: {structured_data.city}")
         print(f"Description: {structured_data.description}")
@@ -62,7 +62,7 @@ async def streaming_example() -> None:
 
     # Access the structured output directly from the response value
     if result.value:
-        structured_data = result.value
+        structured_data: OutputStruct = result.value  # type: ignore
         print("Structured Output (from streaming with AgentRunResponse.from_agent_response_generator):")
         print(f"City: {structured_data.city}")
         print(f"Description: {structured_data.description}")

python/samples/getting_started/context_providers/simple_context_provider.py

@@ -53,7 +53,7 @@ class UserInfoMemory(ContextProvider):
                 )
 
                 # Update user info with extracted data
-                if result.value:
+                if result.value and isinstance(result.value, UserInfo):
                     if self.user_info.name is None and result.value.name:
                         self.user_info.name = result.value.name
                     if self.user_info.age is None and result.value.age:

python/samples/getting_started/workflow/_start-here/step1_executors_and_edges.py

@@ -2,8 +2,6 @@
 
 import asyncio
 
-from typing_extensions import Never
-
 from agent_framework import (
     Executor,
     WorkflowBuilder,
@@ -11,6 +9,7 @@ from agent_framework import (
     executor,
     handler,
 )
+from typing_extensions import Never
 
 """
 Step 1: Foundational patterns: Executors and edges

python/samples/getting_started/workflow/agents/workflow_as_agent_human_in_the_loop.py

@@ -2,8 +2,10 @@
 
 import asyncio
 import sys
+from collections.abc import Mapping
 from dataclasses import dataclass
 from pathlib import Path
+from typing import Any, cast
 
 # Ensure local getting_started package can be imported when running as a script.
 _SAMPLES_ROOT = Path(__file__).resolve().parents[3]
@@ -138,15 +140,33 @@ async def main() -> None:
     # Handle the human review if required.
     if human_review_function_call:
         # Parse the human review request arguments.
-        if isinstance(human_review_function_call.arguments, str):
-            request = WorkflowAgent.RequestInfoFunctionArgs.model_validate_json(human_review_function_call.arguments)
+        human_request_args = human_review_function_call.arguments
+        if isinstance(human_request_args, str):
+            request: WorkflowAgent.RequestInfoFunctionArgs = WorkflowAgent.RequestInfoFunctionArgs.from_json(
+                human_request_args
+            )
+        elif isinstance(human_request_args, Mapping):
+            request = WorkflowAgent.RequestInfoFunctionArgs.from_dict(dict(human_request_args))
         else:
-            request = WorkflowAgent.RequestInfoFunctionArgs.model_validate(human_review_function_call.arguments)
+            raise TypeError("Unexpected argument type for human review function call.")
+
+        request_payload_obj: Any = request.data
+        if not isinstance(request_payload_obj, Mapping):
+            raise ValueError("Human review request payload must be a mapping.")
+        request_payload = cast(Mapping[str, Any], request_payload_obj)
+
+        agent_request_obj = request_payload.get("agent_request")
+        if not isinstance(agent_request_obj, Mapping):
+            raise ValueError("Human review request must include agent_request mapping data.")
+        agent_request_data = cast(Mapping[str, Any], agent_request_obj)
+
+        request_id_obj = agent_request_data.get("request_id")
+        if not isinstance(request_id_obj, str):
+            raise ValueError("Human review request_id must be a string.")
+        request_id_value = request_id_obj
 
         # Mock a human response approval for demonstration purposes.
-        human_response = ReviewResponse(
-            request_id=request.data["agent_request"]["request_id"], feedback="Approved", approved=True
-        )
+        human_response = ReviewResponse(request_id=request_id_value, feedback="Approved", approved=True)
 
         # Create the function call result object to send back to the agent.
         human_review_function_result = FunctionResultContent(