mirror of
https://github.com/Niko1221/Strata.git
synced 2026-11-07 06:15:58 -10:00
Ron Forrester
dd667967b9
The function name was everything between `<function=` and `>`, untrimmed. A model that writes a space after the `=` (seen from Qwen3.8-Flash-Next IQ2_XS in Zed's agent panel) got: - inside the thinking: the name " capture" matched no declared tool, so the #804 rescue kept the call as reasoning text; the reply ended with finish_reason "stop" and no tool_calls, and the agent stopped mid-task; - after the thinking: a call named " capture", which the client does not know, its arguments typed without the tool's schema. The name is now stripped where the call is parsed (parse_tool_call), where the #804 rescue checks it against the declared tools, and where a finished call looks up its schema (_finish_call). A name with surrounding whitespace is never meaningful, so no text becomes a call that was not one: the #804/#1058 rules still decide whether a call inside the thinking is an act. serve/test_tool_name_space.py: both places, streamed and not, fed in 3-character pieces; fails 4/4 before, passes after. test_reasoning_rescue, test_reasoning_tools and test_frontend unchanged and passing.
47 lines
1.8 KiB
Python
47 lines
1.8 KiB
Python
"""serve/test_tool_name_space.py - a tool call written `<function= NAME>` (a space after the `=`) is the call NAME.
|
|
|
|
Seen live (Qwen3.8-Flash-Next IQ2_XS, Zed's agent panel): the model wrote `<function= capture>` inside its thinking;
|
|
the name was read as " capture", matched no declared tool, stayed reasoning text (#804), and the turn ended with no
|
|
call, so the agent stopped. After the thinking the same call came out named " capture", and its arguments were
|
|
typed without the tool's schema.
|
|
|
|
python -m unittest serve.test_tool_name_space -v
|
|
"""
|
|
import sys
|
|
import unittest
|
|
from pathlib import Path
|
|
|
|
sys.path.insert(0, str(Path(__file__).resolve().parents[1]))
|
|
|
|
from serve.frontend import OutputParser # noqa: E402
|
|
|
|
# "42" for a string parameter stays a string only when the call finds its tool's schema
|
|
SCHEMA = [{"name": "capture", "parameters": {"properties": {"content": {"type": "string"}}}}]
|
|
CALL = "<tool_call>\n<function= capture>\n<parameter=content>\n42\n</parameter>\n</function>\n</tool_call>"
|
|
|
|
|
|
def calls(text, stream_tools):
|
|
p = OutputParser(thinking=True, tools=SCHEMA, stream_tools=stream_tools)
|
|
evs = []
|
|
for i in range(0, len(text), 3): # fed in small pieces, as a stream arrives
|
|
evs += p.feed(text[i:i + 3])
|
|
evs += p.finish("stop")
|
|
return [(e.call.name, e.call.arguments) for e in evs if e.kind == "tool_call"]
|
|
|
|
|
|
class SpacedName(unittest.TestCase):
|
|
def check(self, text):
|
|
for stream_tools in (False, True):
|
|
with self.subTest(stream_tools=stream_tools):
|
|
self.assertEqual(calls(text, stream_tools), [("capture", {"content": "42"})])
|
|
|
|
def test_inside_the_thinking(self):
|
|
self.check("Let me save this.\n" + CALL + "\n")
|
|
|
|
def test_after_the_thinking(self):
|
|
self.check("Saving it.</think>\n\n" + CALL)
|
|
|
|
|
|
if __name__ == "__main__":
|
|
unittest.main()
|