Python: (samples): adopt AzureOpenAIResponsesClient, reorganize orchestration examples, and fix workflow/orchestration bugs (#3873)

* adopt AzureOpenAIResponsesClient, reorganize orchestration examples, and fix workflow/orchestration bugs

* Updates

* add comment

python/samples/getting_started/workflows/parallelism/fan_out_fan_in_edges.py

@@ -1,19 +1,21 @@
 # Copyright (c) Microsoft. All rights reserved.
 
 import asyncio
+import os
 from dataclasses import dataclass
 
 from agent_framework import (
     AgentExecutor,  # Wraps a ChatAgent as an Executor for use in workflows
     AgentExecutorRequest,  # The message bundle sent to an AgentExecutor
     AgentExecutorResponse,  # The structured result returned by an AgentExecutor
+    AgentResponseUpdate,
     Executor,  # Base class for custom Python executors
     Message,  # Chat message structure
     WorkflowBuilder,  # Fluent builder for wiring the workflow graph
     WorkflowContext,  # Per run context and event bus
     handler,  # Decorator to mark an Executor method as invokable
 )
-from agent_framework.azure import AzureOpenAIChatClient
+from agent_framework.azure import AzureOpenAIResponsesClient
 from azure.identity import AzureCliCredential  # Uses your az CLI login for credentials
 from typing_extensions import Never
 
@@ -29,8 +31,9 @@ Show how to construct a parallel branch pattern in workflows. Demonstrate:
 - Fan in by collecting a list of AgentExecutorResponse objects and reducing them to a single result.
 
 Prerequisites:
+- AZURE_AI_PROJECT_ENDPOINT must be your Azure AI Foundry Agent Service (V2) project endpoint.
 - Familiarity with WorkflowBuilder, executors, edges, events, and streaming runs.
-- Azure OpenAI access configured for AzureOpenAIChatClient. Log in with Azure CLI and set any required environment variables.
+- Azure OpenAI access configured for AzureOpenAIResponsesClient. Log in with Azure CLI and set any required environment variables.
 - Comfort reading AgentExecutorResponse.agent_response.text for assistant output aggregation.
 """
 
@@ -87,13 +90,31 @@ class AggregateInsights(Executor):
         await ctx.yield_output(consolidated)
 
 
+def render_live_streams(buffers: dict[str, str], order: list[str], completed: set[str]) -> None:
+    """Render concurrent agent streams in separate sections."""
+    # Clear terminal and move cursor to top-left for a live dashboard effect.
+    print("\033[2J\033[H", end="")
+    print("=== Expert Streams (Live) ===")
+    print("Concurrent agent updates are shown below as they stream.\n")
+    for agent_id in order:
+        state = "completed" if agent_id in completed else "streaming"
+        print(f"[{agent_id}] ({state})")
+        print(buffers.get(agent_id, ""))
+        print("-" * 80)
+    print("", end="", flush=True)
+
+
 async def main() -> None:
     # 1) Create executor and agent instances
     dispatcher = DispatchToExperts(id="dispatcher")
     aggregator = AggregateInsights(id="aggregator")
 
     researcher = AgentExecutor(
-        AzureOpenAIChatClient(credential=AzureCliCredential()).as_agent(
+        AzureOpenAIResponsesClient(
+            project_endpoint=os.environ["AZURE_AI_PROJECT_ENDPOINT"],
+            deployment_name=os.environ["AZURE_AI_MODEL_DEPLOYMENT_NAME"],
+            credential=AzureCliCredential(),
+        ).as_agent(
             instructions=(
                 "You're an expert market and product researcher. Given a prompt, provide concise, factual insights,"
                 " opportunities, and risks."
@@ -102,7 +123,11 @@ async def main() -> None:
         )
     )
     marketer = AgentExecutor(
-        AzureOpenAIChatClient(credential=AzureCliCredential()).as_agent(
+        AzureOpenAIResponsesClient(
+            project_endpoint=os.environ["AZURE_AI_PROJECT_ENDPOINT"],
+            deployment_name=os.environ["AZURE_AI_MODEL_DEPLOYMENT_NAME"],
+            credential=AzureCliCredential(),
+        ).as_agent(
             instructions=(
                 "You're a creative marketing strategist. Craft compelling value propositions and target messaging"
                 " aligned to the prompt."
@@ -111,7 +136,11 @@ async def main() -> None:
         )
     )
     legal = AgentExecutor(
-        AzureOpenAIChatClient(credential=AzureCliCredential()).as_agent(
+        AzureOpenAIResponsesClient(
+            project_endpoint=os.environ["AZURE_AI_PROJECT_ENDPOINT"],
+            deployment_name=os.environ["AZURE_AI_MODEL_DEPLOYMENT_NAME"],
+            credential=AzureCliCredential(),
+        ).as_agent(
             instructions=(
                 "You're a cautious legal/compliance reviewer. Highlight constraints, disclaimers, and policy concerns"
                 " based on the prompt."
@@ -128,18 +157,32 @@ async def main() -> None:
         .build()
     )
 
-    # 3) Run with a single prompt and print progress plus the final consolidated output
+    # 3) Run with a single prompt and render live expert streams plus final consolidated output.
+    expert_order = ["researcher", "marketer", "legal"]
+    expert_buffers: dict[str, str] = {expert_id: "" for expert_id in expert_order}
+    completed_experts: set[str] = set()
+    final_output: str | None = None
+
     async for event in workflow.run(
         "We are launching a new budget-friendly electric bike for urban commuters.", stream=True
     ):
-        if event.type == "executor_invoked":
-            # Show when executors are invoked and completed for lightweight observability.
-            print(f"{event.executor_id} invoked")
-        elif event.type == "executor_completed":
-            print(f"{event.executor_id} completed")
+        if event.type == "executor_completed" and event.executor_id in expert_buffers:
+            completed_experts.add(event.executor_id)
+            render_live_streams(expert_buffers, expert_order, completed_experts)
         elif event.type == "output":
-            print("===== Final Aggregated Output =====")
-            print(event.data)
+            if isinstance(event.data, AgentResponseUpdate):
+                executor_id = event.executor_id or ""
+                if executor_id in expert_buffers:
+                    expert_buffers[executor_id] += event.data.text
+                    render_live_streams(expert_buffers, expert_order, completed_experts)
+                continue
+
+            if event.executor_id == "aggregator":
+                final_output = str(event.data)
+
+    if final_output:
+        print("\n=== Final Consolidated Output ===\n")
+        print(final_output)
 
 
 if __name__ == "__main__":