Files
Eduard van Valkenburg 1e350ea22f Python: [BREAKING] PR2 — Wire context provider pipeline, remove old types, update all consumers (#3850)
* PR2: Wire context provider pipeline and update all internal consumers

- Replace AgentThread with AgentSession across all packages
- Replace ContextProvider with BaseContextProvider across all packages
- Replace context_provider param with context_providers (Sequence)
- Replace thread= with session= in run() signatures
- Replace get_new_thread() with create_session()
- Add get_session(service_session_id) to agent interface
- DurableAgentThread -> DurableAgentSession
- Remove _notify_thread_of_new_messages from WorkflowAgent
- Wire before_run/after_run context provider pipeline in RawAgent
- Auto-inject InMemoryHistoryProvider when no providers configured

* fix: update all tests for context provider pipeline, fix lazy-loaders, remove old test files

* refactor: update all sample files for context provider pipeline (AgentThread→AgentSession, ContextProvider→BaseContextProvider)

* fix: update remaining ag-ui references (client docstring, getting_started sample)

* fix: make get_session service_session_id keyword-only to avoid confusion with session_id

* refactor: rename _RunContext.thread_messages to session_messages

* refactor: remove _threads.py, _memory.py, and old provider files; migrate devui to use plain message lists

* rename: remove _new_ prefix from test files

* refactor: rewrite SlidingWindowChatMessageStore as SlidingWindowHistoryProvider(InMemoryHistoryProvider)

* fix: read full history from session state directly instead of reaching into provider internals

* fix: update stale .pyi stubs, sample imports, and README references for new provider types

* fix: remove stale message_store, _notify_thread_of_new_messages, and session_id.key references in samples

* refactor: merge context_providers and sessions sample folders into sessions, remove aggregate_context_provider

* refactor: UserInfoMemory stores state in session.state instead of instance attributes

* feat: add Pydantic BaseModel support to session state serialization

Pydantic models stored in session.state are now automatically serialized
via model_dump() and restored via model_validate() during to_dict()/from_dict()
round-trips. Models are auto-registered on first serialization; use
register_state_type() for cold-start deserialization.

Also export register_state_type as a public API.

* fix mem0

* Update sample README links and descriptions for session terminology

- Replace 'thread' with 'session' in sample descriptions across all READMEs
- Update file links for renamed samples (mem0_sessions, redis_sessions, etc.)
- Fix Threads section → Sessions section in main samples/README.md
- Update tools, middleware, workflows, durabletask, azure_functions READMEs
- Update architecture diagrams in concepts/tools/README.md
- Update migration guides (autogen, semantic-kernel)

* Fix broken Redis README link to renamed sample

* Fix Mem0 OSS client search: pass scoping params as direct kwargs

AsyncMemory (OSS) expects user_id/agent_id/run_id as direct kwargs,
while AsyncMemoryClient (Platform) expects them in a filters dict.
Adds tests for both client types.

Port of fix from #3844 to new Mem0ContextProvider.

* Fix rebase issues: restore missing _conversation_state.py and checkpoint decode logic

- Add back _conversation_state.py (encode/decode_chat_messages) lost in rebase
- Fix on_checkpoint_restore to decode cache/conversation with decode_chat_messages
- Fix on_checkpoint_restore to use decode_checkpoint_value for pending requests
- Add tests/workflow/__init__.py for relative import support
- Fix test_agent_executor checkpoint selection (checkpoints[1] not superstep)

* Add STORES_BY_DEFAULT ClassVar to skip redundant InMemoryHistoryProvider injection

Chat clients that store history server-side by default (OpenAI Responses API,
Azure AI Agent) now declare STORES_BY_DEFAULT = True. The agent checks this
during auto-injection and skips InMemoryHistoryProvider unless the user
explicitly sets store=False.

* Fix broken markdown links in azure_ai and redis READMEs

* Fix getting-started samples to use session API instead of removed thread/ContextProvider API

* updates to workflow as agent

* fix group chat import

* Rename Thread→Session throughout, fix service_session_id propagation, remove stale AGUIThread

- Fix: Propagate conversation_id from ChatResponse back to session.service_session_id
  in both streaming and non-streaming paths in _agents.py
- Rename AgentThreadException → AgentSessionException
- Remove stale AGUIThread from ag_ui lazy-loader
- Rename use_service_thread → use_service_session in ag-ui package
- Rename test functions from *_thread_* to *_session_*
- Rename sample files from *_thread* to *_session*
- Update docstrings and comments: thread → session
- Update _mcp.py kwargs filter: add 'session' alongside 'thread'
- Fix ContinuationToken docstring example: thread=thread → session=session
- Fix _clients.py docstring: 'Agent threads' → 'Agent sessions'

* Fix broken markdown links after thread→session file renames

* fix azure ai test
2026-02-12 21:00:32 +00:00

206 lines
7.5 KiB
Python

# Copyright (c) Microsoft. All rights reserved.
"""Tests for sliding window history provider."""
from unittest.mock import patch
from agent_framework._types import Content, Message
from agent_framework_lab_tau2._sliding_window import SlidingWindowHistoryProvider
def _make_state(provider: SlidingWindowHistoryProvider, messages: list[Message] | None = None) -> dict:
"""Helper to create a session state dict with messages pre-loaded."""
state: dict = {}
if messages:
state[provider.source_id] = {"messages": list(messages)}
return state
def test_initialization():
"""Test initializing with parameters."""
provider = SlidingWindowHistoryProvider(
max_tokens=2000,
system_message="You are a helpful assistant",
tool_definitions=[{"name": "test_tool"}],
)
assert provider.max_tokens == 2000
assert provider.system_message == "You are a helpful assistant"
assert provider.tool_definitions == [{"name": "test_tool"}]
assert provider.source_id == "memory"
async def test_get_messages_empty():
"""Test getting messages from empty state."""
provider = SlidingWindowHistoryProvider(max_tokens=1000)
messages = await provider.get_messages(None, state={})
assert messages == []
async def test_get_messages_simple():
"""Test getting messages without truncation."""
provider = SlidingWindowHistoryProvider(max_tokens=10000)
msgs = [
Message(role="user", contents=[Content.from_text(text="What's the weather?")]),
Message(role="assistant", contents=[Content.from_text(text="I can help with that.")]),
]
state = _make_state(provider, msgs)
result = await provider.get_messages(None, state=state)
assert len(result) == 2
assert result[0].text == "What's the weather?"
assert result[1].text == "I can help with that."
async def test_save_and_get_messages():
"""Test saving then getting messages with truncation."""
provider = SlidingWindowHistoryProvider(max_tokens=50)
state: dict = {}
# Save many messages
msgs = [
Message(role="user", contents=[Content.from_text(text=f"Message {i} with some content")]) for i in range(10)
]
await provider.save_messages(None, msgs, state=state)
# get_messages returns truncated
truncated = await provider.get_messages(None, state=state)
# Full history is in session state
all_msgs = state[provider.source_id]["messages"]
assert len(all_msgs) == 10
assert len(truncated) < len(all_msgs)
def test_get_token_count_basic():
"""Test basic token counting."""
provider = SlidingWindowHistoryProvider(max_tokens=1000)
messages = [Message(role="user", contents=[Content.from_text(text="Hello")])]
token_count = provider._get_token_count(messages)
assert token_count > 0
def test_get_token_count_with_system_message():
"""Test token counting includes system message."""
provider = SlidingWindowHistoryProvider(max_tokens=1000, system_message="You are a helpful assistant")
count_empty = provider._get_token_count([])
count_with_msg = provider._get_token_count([Message(role="user", contents=[Content.from_text(text="Hello")])])
assert count_with_msg > count_empty
assert count_empty > 0 # System message contributes tokens
def test_get_token_count_function_call():
"""Test token counting with function calls."""
function_call = Content.from_function_call(call_id="call_123", name="test_function", arguments={"param": "value"})
provider = SlidingWindowHistoryProvider(max_tokens=1000)
token_count = provider._get_token_count([Message(role="assistant", contents=[function_call])])
assert token_count > 0
def test_get_token_count_function_result():
"""Test token counting with function results."""
function_result = Content.from_function_result(call_id="call_123", result={"success": True, "data": "result"})
provider = SlidingWindowHistoryProvider(max_tokens=1000)
token_count = provider._get_token_count([Message(role="tool", contents=[function_result])])
assert token_count > 0
@patch("agent_framework_lab_tau2._sliding_window.logger")
def test_truncate_removes_old_messages(mock_logger):
"""Test that truncation removes old messages when token limit exceeded."""
provider = SlidingWindowHistoryProvider(max_tokens=20)
messages = [
Message(
role="user",
contents=[Content.from_text(text="This is a very long message that should exceed the token limit")],
),
Message(
role="assistant",
contents=[
Content.from_text(text="This is another very long message that should also exceed the token limit")
],
),
Message(role="user", contents=[Content.from_text(text="Short msg")]),
]
result = provider._truncate(list(messages))
assert len(result) < len(messages)
assert mock_logger.warning.called
@patch("agent_framework_lab_tau2._sliding_window.logger")
def test_truncate_removes_leading_tool_messages(mock_logger):
"""Test that truncation removes leading tool messages."""
provider = SlidingWindowHistoryProvider(max_tokens=10000)
tool_message = Message(role="tool", contents=[Content.from_function_result(call_id="call_123", result="result")])
user_message = Message(role="user", contents=[Content.from_text(text="Hello")])
result = provider._truncate([tool_message, user_message])
assert len(result) == 1
assert result[0].role == "user"
mock_logger.warning.assert_called()
def test_estimate_any_object_token_count():
"""Test token counting for various object types."""
provider = SlidingWindowHistoryProvider(max_tokens=1000)
assert provider._estimate_any_object_token_count({"key": "value"}) > 0
assert provider._estimate_any_object_token_count("test string") > 0
# Non-serializable falls back to str()
class Custom:
def __str__(self):
return "Custom instance"
assert provider._estimate_any_object_token_count(Custom()) > 0
async def test_real_world_scenario():
"""Test a realistic conversation scenario."""
provider = SlidingWindowHistoryProvider(max_tokens=30, system_message="You are a helpful assistant")
state: dict = {}
conversation = [
Message(role="user", contents=[Content.from_text(text="Hello, how are you?")]),
Message(
role="assistant",
contents=[Content.from_text(text="I'm doing well, thank you! How can I help you today?")],
),
Message(role="user", contents=[Content.from_text(text="Can you tell me about the weather?")]),
Message(
role="assistant",
contents=[
Content.from_text(
text="I'd be happy to help with weather information, "
"but I don't have access to current weather data."
)
],
),
Message(role="user", contents=[Content.from_text(text="What about telling me a joke instead?")]),
Message(
role="assistant",
contents=[
Content.from_text(text="Sure! Why don't scientists trust atoms? Because they make up everything!")
],
),
]
await provider.save_messages(None, conversation, state=state)
truncated = await provider.get_messages(None, state=state)
all_msgs = state[provider.source_id]["messages"]
assert len(all_msgs) == 6
assert len(truncated) <= 6
token_count = provider._get_token_count(truncated)
assert token_count <= provider.max_tokens * 1.1