Files
manual_slop/tests/mock_concurrent_mma.py
T
ed 913aa48ca9 fix(mock_concurrent_mma): route sprints on prompt content not session_id
The prior session_id-based routing (added in 635ca552) had two bugs:
1. call_n literal matching (== 2, == 3) is fragile to test ordering:
   the file-based counter persists across tests in the same session,
   so call_n != 2 for the 1st sprint if a prior test ran.
2. session_id='mock-sprint-A' means 'this is a follow-up call after
   the 1st sprint returned mock-sprint-A', so the response should be
   sprint-B (2nd track tickets), not sprint-A. The prior code routed
   this to sprint-A, which means track-b's worker has stream id
   'ticket-A-1' (not 'ticket-B-1') and the test's 'ticket-B-1' poll
   never finds it.

Fix: route on prompt content. The production's conductor_tech_lead
passes the track_brief (containing 'Track A Goal' or 'Track B Goal')
in the user_message. The prompt is NOT empty in --resume mode (the
gemini_cli_adapter passes the prompt as the first turn of the resumed
session).

The prompt-based routing is the original pre-635ca552 design and
works correctly for any number of tracks (A, B, C) without depending
on call ordering.

Verified: 3 consecutive test runs PASS (7.81s, 8.90s, 7.95s) after
the fix. The 'Worker from Track B never appeared' flakiness is gone.
2026-06-27 14:20:33 -04:00

141 lines
4.4 KiB
Python

import sys
import json
import os
# Persistent call counter (file-based so the mock survives across subprocess
# invocations). The mock gemini CLI is a short-lived subprocess invoked once
# per send() call; the session_id set by the adapter (--resume) tells the
# mock which response to return. Path is relative to the repo root (the test
# fixture sets subprocess cwd to tests/artifacts/live_gui_workspace_<ts>/ but
# the mock is invoked from the project root by its absolute path).
_CALL_COUNT_FILE = os.path.join(
os.path.dirname(os.path.abspath(__file__)),
"..", "artifacts", ".mock_concurrent_mma_call_count",
)
_CALL_COUNT_FILE = os.path.abspath(_CALL_COUNT_FILE)
def _next_call_count() -> int:
"""Atomically increment and return the per-test mock call count."""
try:
n = 0
if os.path.exists(_CALL_COUNT_FILE):
with open(_CALL_COUNT_FILE, "r", encoding="utf-8") as f:
n = int((f.read() or "0").strip() or "0")
n += 1
os.makedirs(os.path.dirname(_CALL_COUNT_FILE), exist_ok=True)
with open(_CALL_COUNT_FILE, "w", encoding="utf-8") as f:
f.write(str(n))
return n
except Exception:
return 0
def main() -> None:
# Read prompt from stdin
try:
prompt = sys.stdin.read()
except Exception:
prompt = ""
# Detect the session we're "resuming" via --resume arg (set by the
# gemini_cli_adapter on subsequent calls).
session_id = ""
argv = sys.argv[1:]
if "--resume" in argv:
i = argv.index("--resume")
if i + 1 < len(argv):
session_id = argv[i + 1]
call_n = _next_call_count()
# 1. Epic Initialization
if 'PATH: Epic Initialization' in prompt:
mock_response = [
{"id": "track-a", "goal": "Track A Goal", "title": "Track A"},
{"id": "track-b", "goal": "Track B Goal", "title": "Track B"}
]
print(json.dumps({
"type": "message",
"role": "assistant",
"content": json.dumps(mock_response)
}), flush=True)
print(json.dumps({
"type": "result",
"status": "success",
"stats": {"total_tokens": 100, "input_tokens": 50, "output_tokens": 50},
"session_id": "mock-epic"
}), flush=True)
return
# 2. Sprint Planning (different tickets for different tracks)
# Route on prompt content (the production passes the track_brief which
# contains "Track A" or "Track B"). The prior session_id-based routing was
# fragile because:
# 1. The call_n counter is shared across tests in the same session, so
# call_n != 2 for the 1st sprint if a prior test ran.
# 2. session_id="mock-sprint-A" means "this is a follow-up call after
# the 1st sprint returned mock-sprint-A", so the response should be
# sprint-B (2nd track), not sprint-A.
if 'generate the implementation tickets' in prompt:
if "Track A" in prompt: track_label = "A"
elif "Track B" in prompt: track_label = "B"
elif "Track C" in prompt: track_label = "C"
else: track_label = "A"
_emit_sprint_ticket(track_label)
return
# 3. Worker Execution
if 'You are assigned to Ticket' in prompt or session_id.startswith("mock-worker-"):
import re
match = re.search(r'Ticket (ticket-[A-Ba-b]-1)', prompt, re.IGNORECASE)
if match:
tid = match.group(1)
elif session_id.startswith("mock-worker-"):
tid = session_id[len("mock-worker-"):]
else:
tid = "unknown"
print(json.dumps({
"type": "message",
"role": "assistant",
"content": f"Working on {tid}. Done."
}), flush=True)
print(json.dumps({
"type": "result",
"status": "success",
"stats": {"total_tokens": 50, "input_tokens": 25, "output_tokens": 25},
"session_id": f"mock-worker-{tid}"
}), flush=True)
return
# Default
print(json.dumps({
"type": "message",
"role": "assistant",
"content": f"Mock response. Received prompt: {prompt[:100]}..."
}), flush=True)
print(json.dumps({
"type": "result",
"status": "success",
"stats": {"total_tokens": 10, "input_tokens": 5, "output_tokens": 5},
"session_id": "mock-default"
}), flush=True)
def _emit_sprint_ticket(track_label: str) -> None:
mock_response = [
{"id": f"ticket-{track_label}-1", "description": f"Ticket {track_label} 1", "status": "todo", "assigned_to": "worker", "depends_on": []}
]
print(json.dumps({
"type": "message",
"role": "assistant",
"content": json.dumps(mock_response)
}), flush=True)
print(json.dumps({
"type": "result",
"status": "success",
"stats": {"total_tokens": 100, "input_tokens": 50, "output_tokens": 50},
"session_id": f"mock-sprint-{track_label}"
}), flush=True)
if __name__ == "__main__":
main()