-
Notifications
You must be signed in to change notification settings - Fork 3.3k
Expand file tree
/
Copy pathtest_lowlevel_input_validation.py
More file actions
311 lines (253 loc) · 11.1 KB
/
test_lowlevel_input_validation.py
File metadata and controls
311 lines (253 loc) · 11.1 KB
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
"""Test input schema validation for lowlevel server."""
import logging
from collections.abc import Awaitable, Callable
from typing import Any
import anyio
import pytest
from mcp.client.session import ClientSession
from mcp.server import Server
from mcp.server.lowlevel import NotificationOptions
from mcp.server.models import InitializationOptions
from mcp.server.session import ServerSession
from mcp.shared.message import SessionMessage
from mcp.shared.session import RequestResponder
from mcp.types import CallToolResult, ClientResult, ServerNotification, ServerRequest, TextContent, Tool
async def run_tool_test(
tools: list[Tool],
call_tool_handler: Callable[[str, dict[str, Any]], Awaitable[list[TextContent]]],
test_callback: Callable[[ClientSession], Awaitable[CallToolResult]],
) -> CallToolResult | None:
"""Helper to run a tool test with minimal boilerplate.
Args:
tools: List of tools to register
call_tool_handler: Handler function for tool calls
test_callback: Async function that performs the test using the client session
Returns:
The result of the tool call
"""
server = Server("test")
result = None
@server.list_tools()
async def list_tools():
return tools
@server.call_tool()
async def call_tool(name: str, arguments: dict[str, Any]) -> list[TextContent]:
return await call_tool_handler(name, arguments)
server_to_client_send, server_to_client_receive = anyio.create_memory_object_stream[SessionMessage](10)
client_to_server_send, client_to_server_receive = anyio.create_memory_object_stream[SessionMessage](10)
# Message handler for client
async def message_handler(
message: RequestResponder[ServerRequest, ClientResult] | ServerNotification | Exception,
) -> None:
if isinstance(message, Exception):
raise message
# Server task
async def run_server():
async with ServerSession(
client_to_server_receive,
server_to_client_send,
InitializationOptions(
server_name="test-server",
server_version="1.0.0",
capabilities=server.get_capabilities(
notification_options=NotificationOptions(),
experimental_capabilities={},
),
),
) as server_session:
async with anyio.create_task_group() as tg:
async def handle_messages():
async for message in server_session.incoming_messages:
await server._handle_message(message, server_session, {}, False)
tg.start_soon(handle_messages)
await anyio.sleep_forever()
# Run the test
async with anyio.create_task_group() as tg:
tg.start_soon(run_server)
async with ClientSession(
server_to_client_receive,
client_to_server_send,
message_handler=message_handler,
) as client_session:
# Initialize the session
await client_session.initialize()
# Run the test callback
result = await test_callback(client_session)
# Cancel the server task
tg.cancel_scope.cancel()
return result
def create_add_tool() -> Tool:
"""Create a standard 'add' tool for testing."""
return Tool(
name="add",
description="Add two numbers",
inputSchema={
"type": "object",
"properties": {
"a": {"type": "number"},
"b": {"type": "number"},
},
"required": ["a", "b"],
"additionalProperties": False,
},
)
@pytest.mark.anyio
async def test_valid_tool_call():
"""Test that valid arguments pass validation."""
async def call_tool_handler(name: str, arguments: dict[str, Any]) -> list[TextContent]:
if name == "add":
result = arguments["a"] + arguments["b"]
return [TextContent(type="text", text=f"Result: {result}")]
else:
raise ValueError(f"Unknown tool: {name}")
async def test_callback(client_session: ClientSession) -> CallToolResult:
return await client_session.call_tool("add", {"a": 5, "b": 3})
result = await run_tool_test([create_add_tool()], call_tool_handler, test_callback)
# Verify results
assert result is not None
assert not result.isError
assert len(result.content) == 1
assert result.content[0].type == "text"
assert isinstance(result.content[0], TextContent)
assert result.content[0].text == "Result: 8"
@pytest.mark.anyio
async def test_invalid_tool_call_missing_required():
"""Test that missing required arguments fail validation."""
async def call_tool_handler(name: str, arguments: dict[str, Any]) -> list[TextContent]:
# This should not be reached due to validation
raise RuntimeError("Should not reach here")
async def test_callback(client_session: ClientSession) -> CallToolResult:
return await client_session.call_tool("add", {"a": 5}) # missing 'b'
result = await run_tool_test([create_add_tool()], call_tool_handler, test_callback)
# Verify results
assert result is not None
assert result.isError
assert len(result.content) == 1
assert result.content[0].type == "text"
assert isinstance(result.content[0], TextContent)
assert "Input validation error" in result.content[0].text
assert "'b' is a required property" in result.content[0].text
@pytest.mark.anyio
async def test_invalid_tool_call_wrong_type():
"""Test that wrong argument types fail validation."""
async def call_tool_handler(name: str, arguments: dict[str, Any]) -> list[TextContent]:
# This should not be reached due to validation
raise RuntimeError("Should not reach here")
async def test_callback(client_session: ClientSession) -> CallToolResult:
return await client_session.call_tool("add", {"a": "five", "b": 3}) # 'a' should be number
result = await run_tool_test([create_add_tool()], call_tool_handler, test_callback)
# Verify results
assert result is not None
assert result.isError
assert len(result.content) == 1
assert result.content[0].type == "text"
assert isinstance(result.content[0], TextContent)
assert "Input validation error" in result.content[0].text
assert "'five' is not of type 'number'" in result.content[0].text
@pytest.mark.anyio
async def test_cache_refresh_on_missing_tool():
"""Test that tool cache is refreshed when tool is not found."""
tools = [
Tool(
name="multiply",
description="Multiply two numbers",
inputSchema={
"type": "object",
"properties": {
"x": {"type": "number"},
"y": {"type": "number"},
},
"required": ["x", "y"],
},
)
]
async def call_tool_handler(name: str, arguments: dict[str, Any]) -> list[TextContent]:
if name == "multiply":
result = arguments["x"] * arguments["y"]
return [TextContent(type="text", text=f"Result: {result}")]
else:
raise ValueError(f"Unknown tool: {name}")
async def test_callback(client_session: ClientSession) -> CallToolResult:
# Call tool without first listing tools (cache should be empty)
# The cache should be refreshed automatically
return await client_session.call_tool("multiply", {"x": 10, "y": 20})
result = await run_tool_test(tools, call_tool_handler, test_callback)
# Verify results - should work because cache will be refreshed
assert result is not None
assert not result.isError
assert len(result.content) == 1
assert result.content[0].type == "text"
assert isinstance(result.content[0], TextContent)
assert result.content[0].text == "Result: 200"
@pytest.mark.anyio
async def test_enum_constraint_validation():
"""Test that enum constraints are validated."""
tools = [
Tool(
name="greet",
description="Greet someone",
inputSchema={
"type": "object",
"properties": {
"name": {"type": "string"},
"title": {"type": "string", "enum": ["Mr", "Ms", "Dr"]},
},
"required": ["name"],
},
)
]
async def call_tool_handler(name: str, arguments: dict[str, Any]) -> list[TextContent]:
# This should not be reached due to validation failure
raise RuntimeError("Should not reach here")
async def test_callback(client_session: ClientSession) -> CallToolResult:
return await client_session.call_tool("greet", {"name": "Smith", "title": "Prof"}) # Invalid title
result = await run_tool_test(tools, call_tool_handler, test_callback)
# Verify results
assert result is not None
assert result.isError
assert len(result.content) == 1
assert result.content[0].type == "text"
assert isinstance(result.content[0], TextContent)
assert "Input validation error" in result.content[0].text
assert "'Prof' is not one of" in result.content[0].text
@pytest.mark.anyio
async def test_tool_not_in_list_logs_warning(caplog: pytest.LogCaptureFixture):
"""Test that calling a tool not in list_tools logs a warning and skips validation."""
tools = [
Tool(
name="add",
description="Add two numbers",
inputSchema={
"type": "object",
"properties": {
"a": {"type": "number"},
"b": {"type": "number"},
},
"required": ["a", "b"],
},
)
]
async def call_tool_handler(name: str, arguments: dict[str, Any]) -> list[TextContent]:
# This should be reached since validation is skipped for unknown tools
if name == "unknown_tool":
# Even with invalid arguments, this should execute since validation is skipped
return [TextContent(type="text", text="Unknown tool executed without validation")]
else:
raise ValueError(f"Unknown tool: {name}")
async def test_callback(client_session: ClientSession) -> CallToolResult:
# Call a tool that's not in the list with invalid arguments
# This should trigger the warning about validation not being performed
return await client_session.call_tool("unknown_tool", {"invalid": "args"})
with caplog.at_level(logging.WARNING):
result = await run_tool_test(tools, call_tool_handler, test_callback)
# Verify results - should succeed because validation is skipped for unknown tools
assert result is not None
assert not result.isError
assert len(result.content) == 1
assert result.content[0].type == "text"
assert isinstance(result.content[0], TextContent)
assert result.content[0].text == "Unknown tool executed without validation"
# Verify warning was logged
assert any(
"Tool 'unknown_tool' not listed, no validation will be performed" in record.message for record in caplog.records
)