test: updated the testcases to the latest agent framework. Dev v4 to main

[Dhruvkumar-Microsoft]

Purpose

This pull request updates the backend and test suite to support breaking changes in the agent-framework ecosystem, especially the move to rc4 versions and the removal of the HostedCodeInterpreterTool in favor of server-side code interpreter handling. It also updates dependencies to specific versions and adjusts test mocks and logic to match the new agent framework APIs.

Key changes include:

Dependency and agent framework updates:

• Updated requirements.txt to pin all dependencies to specific versions, upgrade core Azure and OpenAI libraries, and switch to agent-framework-azure-ai, agent-framework-core, and agent-framework-orchestrations release candidates. Added new dependencies required by the updated agent framework and removed semantic-kernel.

Test suite and mocking adjustments for agent framework v1.0.0rc4:

• Overhauled agent framework mocks in conftest.py to include new classes (Agent, Message, etc.), submodules, and attributes required by the updated agent framework, and added mocks for the new agent_framework_orchestrations modules. [1] [2]
• Updated test imports and mocks to use Agent and Message instead of ChatAgent and ChatMessage, reflecting the new agent framework API. Adjusted test assertions and patching accordingly. [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11]
• Modified test logic for code interpreter tool collection: removed local instantiation of HostedCodeInterpreterTool and updated assertions to expect server-side handling, per the new framework behavior. [1] [2] [3]

Other test and mock improvements:

• Added missing mocks for azure.identity.aio and common.config.app_config to avoid import errors in tests.
• Updated test logic to use agent.client instead of agent.chat_client to align with new agent interface. [1] [2]

These changes ensure compatibility with the latest agent framework APIs and maintain test reliability as the backend migrates to the new orchestration and agent management model.

• Dependency and agent framework updates: src/backend/requirements.txtL1-R39
• Agent framework mocks and test suite adjustments: [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13]
• Code interpreter handling and test logic updates: [1] [2] [3]
• Additional test mock improvements: [1] [2] [3]

Does this introduce a breaking change?

• Yes
• No

How to Test

• Get the code

git clone [repo-address]
cd [repo-name]
git checkout [branch-name]
npm install

• Test the code

What to Check

Verify that the following are valid

• ...

Other Information

[Copilot]

Pull request overview

Updates the backend dependency pins and refactors the v4 backend test suite/mocks to align with breaking API changes in the agent-framework ecosystem (notably workflow.run(..., stream=True), Agent/Message, and new agent_framework_orchestrations import paths).

Changes:

• Pinned/updated backend Python dependencies and migrated agent-framework packages to rc/beta versions.
• Updated orchestration tests to reflect event dispatch via event.type strings and streaming via workflow.run(..., stream=True).
• Refreshed test mocks (including agent_framework_orchestrations) and adjusted Foundry agent tests for server-side Code Interpreter handling.

Reviewed changes

Copilot reviewed 7 out of 7 changed files in this pull request and generated 3 comments.

Show a summary per file

File	Description
src/tests/backend/v4/orchestration/test_orchestration_manager.py	Updates orchestration mocks/tests for workflow.run(..., stream=True) and event.type dispatch.
src/tests/backend/v4/orchestration/test_human_approval_manager.py	Adds agent_framework_orchestrations mocks and updates agent framework symbols.
src/tests/backend/v4/magentic_agents/test_foundry_agent.py	Migrates tests from ChatAgent/ChatMessage to Agent/Message and removes Hosted Code Interpreter tool expectations.
src/tests/backend/v4/magentic_agents/common/test_lifecycle.py	Updates lifecycle tests for agent.client (replacing agent.chat_client).
src/tests/backend/v4/callbacks/test_response_handlers.py	Updates response handler tests to use Message for isinstance checks.
src/tests/backend/conftest.py	Expands global agent framework/orchestrations module stubs for import compatibility.
src/backend/requirements.txt	Pins and updates backend dependencies to targeted versions (including agent-framework rc/beta).

Comments suppressed due to low confidence (1)

src/tests/backend/v4/orchestration/test_orchestration_manager.py:781

• test_run_orchestration_event_processing_error no longer exercises the intended error-handling path: the only event emitted is MockMagenticAgentDeltaEvent() which has no .type, so the production code treats it as an unknown event type and never calls streaming_agent_response_callback (meaning the injected side_effect is unused). Update the test to emit an event object with type="output", a non-null executor_id, and data that is an AgentResponseUpdate instance so the streaming callback is invoked and its exception is caught/logged.

    async def test_run_orchestration_event_processing_error(self):
        """Test handling of errors during event processing."""
        # Set up workflow with events that cause processing errors
        mock_workflow = Mock()
        mock_events = [MockMagenticAgentDeltaEvent()]
        mock_workflow.run = AsyncGeneratorMock(mock_events)
        mock_workflow.executors = {}
        
        # Make streaming callback raise exception
        streaming_agent_response_callback.side_effect = Exception("Callback error")
        

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

You can also share your feedback on Copilot code review. Take the survey.