60e0ffc959
Upgrade checks / Notify on failure (push) Has been cancelled
Upgrade checks / Close issue on success (push) Has been cancelled
Schema Crash Test / Real-world schema crash test (232K schemas) (push) Has been cancelled
Run static analysis / static_analysis (push) Has been cancelled
Tests / Tests: Python 3.10 on ubuntu-latest (push) Has been cancelled
Tests / Tests: Python 3.13 on ubuntu-latest (push) Has been cancelled
Tests / Tests: Python 3.10 on windows-latest (push) Has been cancelled
Tests / Tests with lowest-direct dependencies (push) Has been cancelled
Tests / MCP conformance tests (push) Has been cancelled
Tests / Integration tests (push) Has been cancelled
Tests / Package install smoke (push) Has been cancelled
Upgrade checks / Static analysis (push) Has been cancelled
Upgrade checks / Tests: Python 3.10 on ubuntu-latest (push) Has been cancelled
Upgrade checks / Tests: Python 3.13 on ubuntu-latest (push) Has been cancelled
Upgrade checks / Tests: Python 3.10 on windows-latest (push) Has been cancelled
Upgrade checks / Integration tests (push) Has been cancelled
Update MCPServerConfig Schema / update-config-schema (push) Has been cancelled
Update SDK Documentation / update-sdk-docs (push) Has been cancelled
88 lines
3.0 KiB
Python
88 lines
3.0 KiB
Python
"""
|
|
Tests for SEP-1686 ttl parameter handling.
|
|
|
|
Per the spec, servers MUST return ttl in all tasks/get responses,
|
|
and results should be retained for ttl milliseconds after completion.
|
|
"""
|
|
|
|
import asyncio
|
|
|
|
import pytest
|
|
|
|
from fastmcp import FastMCP
|
|
from fastmcp.client import Client
|
|
|
|
|
|
@pytest.fixture
|
|
async def keepalive_server():
|
|
"""Create a server for testing ttl behavior."""
|
|
mcp = FastMCP("keepalive-test")
|
|
|
|
@mcp.tool(task=True)
|
|
async def quick_task(value: int) -> int:
|
|
return value * 2
|
|
|
|
@mcp.tool(task=True)
|
|
async def slow_task() -> str:
|
|
await asyncio.sleep(1)
|
|
return "done"
|
|
|
|
return mcp
|
|
|
|
|
|
async def test_keepalive_returned_in_submitted_state(keepalive_server: FastMCP):
|
|
"""ttl is returned in tasks/get even when task is submitted/working."""
|
|
async with Client(keepalive_server) as client:
|
|
# Submit task with explicit ttl
|
|
task = await client.call_tool(
|
|
"slow_task",
|
|
{},
|
|
task=True,
|
|
ttl=30000, # 30 seconds (client-requested)
|
|
)
|
|
|
|
# Check status immediately - should be submitted or working
|
|
status = await task.status()
|
|
assert status.status in ["working"]
|
|
|
|
# ttl should be present per spec (MUST return in all responses)
|
|
# TODO: Docket uses a global execution_ttl for all tasks, not per-task TTLs.
|
|
# The spec allows servers to override client-requested TTL (line 431).
|
|
# FastMCP returns the server's actual global TTL (60000ms default from Docket).
|
|
# If Docket gains per-task TTL support, update this to verify client-requested TTL is respected.
|
|
assert status.ttl == 60000 # Server's global TTL, not client-requested 30000
|
|
|
|
|
|
async def test_keepalive_returned_in_completed_state(keepalive_server: FastMCP):
|
|
"""ttl is returned in tasks/get after task completes."""
|
|
async with Client(keepalive_server) as client:
|
|
# Submit and complete task
|
|
task = await client.call_tool(
|
|
"quick_task",
|
|
{"value": 5},
|
|
task=True,
|
|
ttl=45000, # Client-requested TTL
|
|
)
|
|
await task.wait(timeout=2.0)
|
|
|
|
# Check status - should be completed
|
|
status = await task.status()
|
|
assert status.status == "completed"
|
|
|
|
# TODO: Docket uses global execution_ttl, not per-task TTLs.
|
|
# Server returns its global TTL (60000ms), not the client-requested 45000ms.
|
|
# This is spec-compliant - servers MAY override requested TTL (spec line 431).
|
|
assert status.ttl == 60000 # Server's global TTL, not client-requested 45000
|
|
|
|
|
|
async def test_default_keepalive_when_not_specified(keepalive_server: FastMCP):
|
|
"""Default ttl is used when client doesn't specify."""
|
|
async with Client(keepalive_server) as client:
|
|
# Submit without explicit ttl
|
|
task = await client.call_tool("quick_task", {"value": 3}, task=True)
|
|
await task.wait(timeout=2.0)
|
|
|
|
status = await task.status()
|
|
# Should have default ttl (60000ms = 60 seconds)
|
|
assert status.ttl == 60000
|