Files
temp/robot/helper_tdd_e2e_mock_only_coverage.py
Brent E. Edwards 5e625b22e1 fix(test): convert M1-M6 E2E suites to real subprocess CLI invocations (closes #658)
Replace CliRunner + unittest.mock.patch with subprocess.run for all
21 CLI-facing test functions across the M1-M6 E2E verification helpers.

Application code fixes:
- action.py: _get_lifecycle_service() uses container.plan_lifecycle_service()
- plan.py: _get_lifecycle_service() uses container.plan_lifecycle_service()
- plan.py: three container.resolve(DecisionService) → container.decision_service()

Test infrastructure:
- New robot/helper_e2e_common.py with shared subprocess utilities
  (run_cli, setup_workspace with DB migrations, cleanup_workspace)
- M1-M4, M6 helpers refactored to use run_cli() with real SQLite DB
- M5 unchanged (0 CLI tests, all domain-level)
- TDD detection updated to recognise run_cli() as subprocess invocation
- Remove @tdd_expected_fail from TDD feature + robot tags
- Update 8 Behave step files that mocked container.resolve() to use
  container.decision_service() / container.plan_lifecycle_service()
2026-03-12 20:42:14 +00:00

272 lines
9.0 KiB
Python

"""Helper script for tdd_e2e_mock_only_coverage.robot smoke tests.
Each subcommand uses AST analysis to inspect the M1-M6 E2E verification
helper files and classify each public test function by its mock and CLI
invocation strategy. The helper reports the **real** outcome: it exits 0
and prints the sentinel when the assertion holds (bug is fixed), and exits
1 when the bug is still present. The ``tdd_expected_fail`` tag on the
Robot side handles pass/fail inversion while the bug remains open.
Root cause (Bug #658)
~~~~~~~~~~~~~~~~~~~~~
All 21 CLI-facing tests in ``robot/helper_m*_e2e_verification.py`` use
``unittest.mock.patch`` to replace service factories with ``MagicMock``
objects and invoke the CLI via Typer's in-process ``CliRunner``. Zero
tests invoke the actual ``agents`` CLI binary via ``subprocess.run``.
"""
from __future__ import annotations
import ast
import sys
from collections.abc import Callable
from dataclasses import dataclass, field
from pathlib import Path
_REPO_ROOT = Path(__file__).resolve().parents[1]
_HELPER_FILES = [
_REPO_ROOT / f"robot/helper_m{n}_e2e_verification.py" for n in range(1, 7)
]
# Patch targets that indicate service-layer mocking.
_SERVICE_MOCK_INDICATORS: frozenset[str] = frozenset(
{
"_get_lifecycle_service",
"_get_registry_service",
"_get_namespaced_project_repo",
"_get_resource_link_repo",
"_get_resource_registry_service",
"_get_services",
"_get_service",
"get_container",
"CorrectionService",
}
)
@dataclass
class FunctionAnalysis:
"""Analysis result for a single function in an E2E helper file."""
name: str
suite: str
uses_cli_runner: bool = False
uses_mock_patch: bool = False
uses_subprocess_cli: bool = False
mock_targets: list[str] = field(default_factory=list)
def _analyze_helper(helper_path: Path) -> list[FunctionAnalysis]:
"""Parse a helper file and classify each top-level function."""
source = helper_path.read_text(encoding="utf-8")
tree = ast.parse(source, filename=str(helper_path))
suite = helper_path.stem
results: list[FunctionAnalysis] = []
for node in ast.iter_child_nodes(tree):
if not isinstance(node, ast.FunctionDef):
continue
if node.name.startswith("_"):
continue
fa = FunctionAnalysis(name=node.name, suite=suite)
for child in ast.walk(node):
if isinstance(child, ast.Attribute):
attr_name = child.attr
if attr_name == "invoke" and _is_cli_runner_invoke(child):
fa.uses_cli_runner = True
if attr_name in ("run", "Popen") and _is_subprocess_call(child):
fa.uses_subprocess_cli = True
# Detect ``run_cli(...)`` calls from helper_e2e_common
# which invoke the real CLI via subprocess.run.
if isinstance(child, ast.Call) and _is_run_cli_call(child):
fa.uses_cli_runner = True # CLI-facing function
fa.uses_subprocess_cli = True # via real subprocess
if isinstance(child, ast.Call) and _is_patch_call(child):
fa.uses_mock_patch = True
target = _extract_patch_target(child)
if target:
fa.mock_targets.append(target)
if isinstance(child, ast.Constant) and isinstance(child.value, str):
for indicator in _SERVICE_MOCK_INDICATORS:
if indicator in child.value:
fa.uses_mock_patch = True
results.append(fa)
return results
def _is_cli_runner_invoke(node: ast.Attribute) -> bool:
"""Check if attribute access is ``runner.invoke(...)``."""
return isinstance(node.value, ast.Name) and node.value.id in (
"runner",
"cli_runner",
)
def _is_subprocess_call(node: ast.Attribute) -> bool:
"""Check if attribute access is ``subprocess.run(...)``."""
return isinstance(node.value, ast.Name) and node.value.id == "subprocess"
def _is_run_cli_call(node: ast.Call) -> bool:
"""Check if a Call node is ``run_cli(...)`` from helper_e2e_common.
``run_cli`` is a convenience wrapper around ``subprocess.run`` that
invokes the real ``agents`` CLI binary. Functions calling it are
CLI-facing and exercise subprocess invocation without mocks.
"""
func = node.func
return isinstance(func, ast.Name) and func.id == "run_cli"
def _is_patch_call(node: ast.Call) -> bool:
"""Check if a Call node is ``patch(...)`` or ``mock.patch(...)``."""
func = node.func
if isinstance(func, ast.Attribute) and func.attr == "patch":
return True
return isinstance(func, ast.Name) and func.id == "patch"
def _extract_patch_target(node: ast.Call) -> str | None:
"""Extract the first string argument from a ``patch(...)`` call."""
if node.args and isinstance(node.args[0], ast.Constant):
val = node.args[0].value
if isinstance(val, str):
return val
return None
def _get_all_functions() -> list[FunctionAnalysis]:
"""Analyze all M1-M6 helpers and return flat function list."""
all_functions: list[FunctionAnalysis] = []
for helper_path in _HELPER_FILES:
if not helper_path.is_file():
print(
f"ERROR: missing helper file {helper_path}",
file=sys.stderr,
)
sys.exit(2)
all_functions.extend(_analyze_helper(helper_path))
return all_functions
# ---------------------------------------------------------------------------
# Subcommands
# ---------------------------------------------------------------------------
def check_subprocess_usage() -> None:
"""Assert at least one CLI-facing test uses subprocess.
Exits 0 with sentinel when the assertion holds (bug fixed).
Exits 1 when the bug is still present.
"""
functions = _get_all_functions()
cli_facing = [f for f in functions if f.uses_cli_runner]
if not cli_facing:
print("ERROR: no CLI-facing functions found", file=sys.stderr)
sys.exit(2)
subprocess_users = [f for f in cli_facing if f.uses_subprocess_cli]
if not subprocess_users:
print(
f"Bug #658 present: {len(cli_facing)} CLI-facing E2E test "
f"functions, but NONE use subprocess. All use CliRunner.",
file=sys.stderr,
)
sys.exit(1)
print("tdd-e2e-subprocess-check-ok")
def check_unmocked_services() -> None:
"""Assert at least one CLI-facing test skips service mocking.
Exits 0 with sentinel when the assertion holds (bug fixed).
Exits 1 when the bug is still present.
"""
functions = _get_all_functions()
cli_facing = [f for f in functions if f.uses_cli_runner]
if not cli_facing:
print("ERROR: no CLI-facing functions found", file=sys.stderr)
sys.exit(2)
unmocked = [f for f in cli_facing if not f.uses_mock_patch]
if not unmocked:
print(
f"Bug #658 present: {len(cli_facing)} CLI-facing E2E test "
f"functions, but ALL mock the service layer.",
file=sys.stderr,
)
sys.exit(1)
print("tdd-e2e-unmocked-services-ok")
def check_per_suite_coverage() -> None:
"""Assert no suite has 100% mocked CLI tests.
Exits 0 with sentinel when the assertion holds (bug fixed).
Exits 1 when the bug is still present.
"""
per_suite: dict[str, list[FunctionAnalysis]] = {}
for helper_path in _HELPER_FILES:
if not helper_path.is_file():
print(
f"ERROR: missing helper file {helper_path}",
file=sys.stderr,
)
sys.exit(2)
suite = helper_path.stem
per_suite[suite] = _analyze_helper(helper_path)
all_mocked_suites: list[str] = []
for suite, functions in per_suite.items():
cli_facing = [f for f in functions if f.uses_cli_runner]
if not cli_facing:
continue
unmocked = [f for f in cli_facing if not f.uses_mock_patch]
if not unmocked:
all_mocked_suites.append(
f"{suite} ({len(cli_facing)} CLI tests, all mocked)"
)
if all_mocked_suites:
print(
"Bug #658 present: suites with 100% mocked CLI tests:\n"
+ "\n".join(f" - {s}" for s in all_mocked_suites),
file=sys.stderr,
)
sys.exit(1)
print("tdd-e2e-per-suite-coverage-ok")
# ---------------------------------------------------------------------------
# Dispatcher
# ---------------------------------------------------------------------------
_COMMANDS: dict[str, Callable[[], None]] = {
"check-subprocess-usage": check_subprocess_usage,
"check-unmocked-services": check_unmocked_services,
"check-per-suite-coverage": check_per_suite_coverage,
}
if __name__ == "__main__":
if len(sys.argv) < 2 or sys.argv[1] not in _COMMANDS:
print(
f"Usage: {sys.argv[0]} <{'|'.join(_COMMANDS)}>",
file=sys.stderr,
)
sys.exit(1)
cmd = _COMMANDS[sys.argv[1]]
cmd()