Files
agent-framework/python/packages/devui/tests/test_conversations.py
T
Victor Dibia c341ee7ed2 Python: DevUI - Internal Refactor, Conversations API support, and per… (#1235)
* Python: DevUI - Internal Refactor, Conversations API support, and performance improvements

Comprehensive refactor of DevUI package including samples relocation,
frontend reorganization, OpenAI Conversations API support, and critical
performance and code quality improvements.

Key Changes:

Architecture & Organization
- Moved DevUI samples to python/samples/getting_started/devui/
- Consolidated with other framework samples for better discoverability
- Added .env.example files and comprehensive README
- Restructured frontend components into feature-based folders (agent, workflow, gallery, layout)
- Created new OpenAI-compliant message renderers (devui should render oai responses types primarily)

New Features
- Added _conversations.py (467 lines) - Full conversation storage abstraction, replaces the /threads endpoint to better match oai conversations api
- Implements OpenAI Conversations API for thread management, Supports in-memory and extensible storage backends

API Simplification
- Use 'model' field as entity_id (agent/workflow name) instead of extra_body
- Use standard OpenAI 'conversation' field for conversation context.

Performance & Quality Improvements
- Improved context management in MessageMapper with bounded memory (~500KB max)
- Implemented hybrid LRU + cleanup approach to prevent unbounded memory growth
- General QOL improvement - Eliminated ~150 lines of dead/duplicate code, Consolidated helper functions into _utils.py, Extracted magic numbers to module-level constants, Optimized conversation item lookups with index-based approach

Testing
- Added test_conversations.py (13 tests)
- Added test_performance_fixes.py (9 tests)
- Updated existing tests for code consolidation
- 53 tests passing

Impact: 76 files changed: +4,106 insertions, -2,373 deletions
All linting and formatting checks passing. No breaking changes - backward compatible.

Migration: Samples moved to python/samples/getting_started/devui/

* readme lint fixes

* initial support for function approval and minor ui fixes
2025-10-08 19:34:30 +00:00

336 lines
11 KiB
Python

# Copyright (c) Microsoft. All rights reserved.
"""Tests for conversation store implementation."""
from typing import cast
import pytest
from openai.types.conversations import InputFileContent, InputImageContent, InputTextContent
from agent_framework_devui._conversations import InMemoryConversationStore
@pytest.mark.asyncio
async def test_create_conversation():
"""Test creating a conversation."""
store = InMemoryConversationStore()
conversation = store.create_conversation(metadata={"agent_id": "test_agent"})
assert conversation.id.startswith("conv_")
assert conversation.object == "conversation"
assert conversation.metadata == {"agent_id": "test_agent"}
@pytest.mark.asyncio
async def test_get_conversation():
"""Test retrieving a conversation."""
store = InMemoryConversationStore()
# Create conversation
created = store.create_conversation(metadata={"agent_id": "test_agent"})
# Retrieve it
retrieved = store.get_conversation(created.id)
assert retrieved is not None
assert retrieved.id == created.id
assert retrieved.metadata == {"agent_id": "test_agent"}
@pytest.mark.asyncio
async def test_get_conversation_not_found():
"""Test retrieving non-existent conversation."""
store = InMemoryConversationStore()
conversation = store.get_conversation("conv_nonexistent")
assert conversation is None
@pytest.mark.asyncio
async def test_update_conversation():
"""Test updating conversation metadata."""
store = InMemoryConversationStore()
# Create conversation
created = store.create_conversation(metadata={"agent_id": "test_agent"})
# Update metadata
updated = store.update_conversation(created.id, metadata={"agent_id": "new_agent", "session_id": "sess_123"})
assert updated.id == created.id
assert updated.metadata == {"agent_id": "new_agent", "session_id": "sess_123"}
@pytest.mark.asyncio
async def test_delete_conversation():
"""Test deleting a conversation."""
store = InMemoryConversationStore()
# Create conversation
created = store.create_conversation(metadata={"agent_id": "test_agent"})
# Delete it
result = store.delete_conversation(created.id)
assert result.id == created.id
assert result.deleted is True
assert result.object == "conversation.deleted"
# Verify it's gone
assert store.get_conversation(created.id) is None
@pytest.mark.asyncio
async def test_get_thread():
"""Test getting underlying AgentThread."""
store = InMemoryConversationStore()
# Create conversation
conversation = store.create_conversation(metadata={"agent_id": "test_agent"})
# Get thread
thread = store.get_thread(conversation.id)
assert thread is not None
# AgentThread should have message_store
assert hasattr(thread, "message_store")
@pytest.mark.asyncio
async def test_get_thread_not_found():
"""Test getting thread for non-existent conversation."""
store = InMemoryConversationStore()
thread = store.get_thread("conv_nonexistent")
assert thread is None
@pytest.mark.asyncio
async def test_list_conversations_by_metadata():
"""Test filtering conversations by metadata."""
store = InMemoryConversationStore()
# Create multiple conversations
_conv1 = store.create_conversation(metadata={"agent_id": "agent1"})
_conv2 = store.create_conversation(metadata={"agent_id": "agent2"})
conv3 = store.create_conversation(metadata={"agent_id": "agent1", "session_id": "sess_1"})
# Filter by agent_id
results = store.list_conversations_by_metadata({"agent_id": "agent1"})
assert len(results) == 2
assert all(cast(dict[str, str], c.metadata).get("agent_id") == "agent1" for c in results if c.metadata)
# Filter by agent_id and session_id
results = store.list_conversations_by_metadata({"agent_id": "agent1", "session_id": "sess_1"})
assert len(results) == 1
assert results[0].id == conv3.id
@pytest.mark.asyncio
async def test_add_items():
"""Test adding items to conversation."""
store = InMemoryConversationStore()
# Create conversation
conversation = store.create_conversation(metadata={"agent_id": "test_agent"})
# Add items
items = [{"role": "user", "content": [{"type": "text", "text": "Hello"}]}]
conv_items = await store.add_items(conversation.id, items=items)
assert len(conv_items) == 1
# Message is a ConversationItem type - check standard OpenAI fields
assert conv_items[0].type == "message"
assert conv_items[0].role == "user"
assert conv_items[0].status == "completed"
assert len(conv_items[0].content) == 1
assert conv_items[0].content[0].type == "text"
text_content = cast(InputTextContent, conv_items[0].content[0])
assert text_content.text == "Hello"
@pytest.mark.asyncio
async def test_list_items():
"""Test listing conversation items."""
store = InMemoryConversationStore()
# Create conversation
conversation = store.create_conversation(metadata={"agent_id": "test_agent"})
# Add items
items = [
{"role": "user", "content": [{"type": "text", "text": "Hello"}]},
{"role": "assistant", "content": [{"type": "text", "text": "Hi there"}]},
]
await store.add_items(conversation.id, items=items)
# List items
retrieved_items, has_more = await store.list_items(conversation.id)
assert len(retrieved_items) >= 2 # At least the items we added
assert has_more is False
@pytest.mark.asyncio
async def test_list_items_pagination():
"""Test pagination when listing items."""
store = InMemoryConversationStore()
# Create conversation
conversation = store.create_conversation(metadata={"agent_id": "test_agent"})
# Add multiple items
items = [{"role": "user", "content": [{"type": "text", "text": f"Message {i}"}]} for i in range(5)]
await store.add_items(conversation.id, items=items)
# List with limit
retrieved_items, has_more = await store.list_items(conversation.id, limit=3)
assert len(retrieved_items) == 3
assert has_more is True
@pytest.mark.asyncio
async def test_list_items_converts_function_calls():
"""Test that list_items properly converts function calls to ResponseFunctionToolCallItem."""
from agent_framework import ChatMessage, ChatMessageStore, Role
store = InMemoryConversationStore()
# Create conversation
conversation = store.create_conversation(metadata={"agent_id": "test_agent"})
# Get the underlying thread and set up message store
thread = store.get_thread(conversation.id)
assert thread is not None
# Initialize message store if not present
if thread.message_store is None:
thread.message_store = ChatMessageStore()
# Simulate messages from agent execution with function calls
messages = [
ChatMessage(role=Role.USER, contents=[{"type": "text", "text": "What's the weather in SF?"}]),
ChatMessage(
role=Role.ASSISTANT,
contents=[
{
"type": "function_call",
"name": "get_weather",
"arguments": '{"city": "San Francisco"}',
"call_id": "call_test123",
}
],
),
ChatMessage(
role=Role.TOOL,
contents=[
{
"type": "function_result",
"call_id": "call_test123",
"output": '{"temperature": 65, "condition": "sunny"}',
}
],
),
ChatMessage(role=Role.ASSISTANT, contents=[{"type": "text", "text": "The weather is sunny, 65°F"}]),
]
# Add messages to thread
await thread.on_new_messages(messages)
# List conversation items
items, has_more = await store.list_items(conversation.id)
# Verify we got the right number and types of items
assert len(items) == 4, f"Expected 4 items, got {len(items)}"
assert has_more is False
# Check item types
assert items[0].type == "message", "First item should be a message"
assert items[0].role == "user"
assert len(items[0].content) == 1
text_content_0 = cast(InputTextContent, items[0].content[0])
assert text_content_0.text == "What's the weather in SF?"
assert items[1].type == "function_call", "Second item should be a function_call"
assert items[1].call_id == "call_test123"
assert items[1].name == "get_weather"
assert items[1].arguments == '{"city": "San Francisco"}'
assert items[1].status == "completed"
assert items[2].type == "function_call_output", "Third item should be a function_call_output"
assert items[2].call_id == "call_test123"
assert items[2].output == '{"temperature": 65, "condition": "sunny"}'
assert items[2].status == "completed"
assert items[3].type == "message", "Fourth item should be a message"
assert items[3].role == "assistant"
assert len(items[3].content) == 1
text_content_3 = cast(InputTextContent, items[3].content[0])
assert text_content_3.text == "The weather is sunny, 65°F"
# CRITICAL: Ensure no empty message items
for item in items:
if item.type == "message":
assert len(item.content) > 0, f"Message item {item.id} has empty content!"
@pytest.mark.asyncio
async def test_list_items_handles_images_and_files():
"""Test that list_items properly converts data content (images/files) to OpenAI types."""
from agent_framework import ChatMessage, ChatMessageStore, Role
store = InMemoryConversationStore()
# Create conversation
conversation = store.create_conversation(metadata={"agent_id": "test_agent"})
# Get the underlying thread
thread = store.get_thread(conversation.id)
assert thread is not None
if thread.message_store is None:
thread.message_store = ChatMessageStore()
# Simulate message with image and file
messages = [
ChatMessage(
role=Role.USER,
contents=[
{"type": "text", "text": "Check this image and PDF"},
{"type": "data", "uri": "data:image/png;base64,iVBORw0KGgo=", "media_type": "image/png"},
{"type": "data", "uri": "data:application/pdf;base64,JVBERi0=", "media_type": "application/pdf"},
],
),
]
await thread.on_new_messages(messages)
# List items
items, has_more = await store.list_items(conversation.id)
assert len(items) == 1
assert items[0].type == "message"
assert items[0].role == "user"
assert len(items[0].content) == 3
# Check content types
assert items[0].content[0].type == "text"
text_content = cast(InputTextContent, items[0].content[0])
assert text_content.text == "Check this image and PDF"
assert items[0].content[1].type == "input_image"
image_content = cast(InputImageContent, items[0].content[1])
assert image_content.image_url == "data:image/png;base64,iVBORw0KGgo="
assert image_content.detail == "auto"
assert items[0].content[2].type == "input_file"
file_content = cast(InputFileContent, items[0].content[2])
assert file_content.file_url == "data:application/pdf;base64,JVBERi0="