Files
temp/features/steps/tdd_e2e_mock_only_coverage_steps.py
hurui200320 1878998b7a refactor(testing): rename tdd_bug/tdd_bug_N tags to tdd_issue/tdd_issue_N
Rename the TDD tag system from tdd_bug/tdd_bug_<N> to tdd_issue/tdd_issue_<N>
across the entire codebase. The tdd_expected_fail tag is unchanged.

The TDD expected-failure workflow is not limited to bug fixes — it applies
equally to any issue type (features, tasks, refactors). The _bug suffix was
misleading and narrowed the perceived scope. The new _issue suffix accurately
reflects that the TDD tagging system applies to any Forgejo issue.

Changes span 92 files:
- features/environment.py: validate_tdd_tags(), should_invert_result(), and
  apply_tdd_inversion() updated — regex, variables, error messages
- robot/tdd_expected_fail_listener.py: _validate_tdd_tags(), _should_invert_result(),
  start_test(), end_test() updated consistently
- 33 Behave .feature files: all @tdd_bug/@tdd_bug_<N> tags renamed
- 29 Robot .robot files: all tdd_bug/tdd_bug_<N> tags renamed
- 3 Robot fixture files renamed (tdd_bug_alone, tdd_missing_tdd_bug,
  tdd_expected_fail_missing_bug_n) with content and references updated
- Tag validation tests and helpers updated (function names, command dispatch
  keys, output strings, fixture references)
- CONTRIBUTING.md: section renamed from 'TDD Bug Test Tags' to
  'TDD Issue Test Tags', all tag references and examples updated
- noxfile.py: comment references updated
- Step definition files, mock helpers, and benchmark files: docstring
  references updated

ISSUES CLOSED: #965
2026-03-27 05:58:35 +00:00

275 lines
10 KiB
Python

"""Step definitions for TDD Issue #658 — E2E mock-only coverage.
These steps use AST analysis to inspect the M1-M6 E2E verification helper
files (``robot/helper_m*_e2e_verification.py``) and classify each public
test function by its mock and CLI invocation strategy.
A function is classified as "CLI-facing" if it uses ``CliRunner`` (from
``typer.testing``) to invoke CLI commands. The test asserts that at least
one such CLI-facing function invokes the real ``agents`` CLI via
``subprocess.run`` (or equivalent) without mocking the service layer.
With ``@tdd_expected_fail`` the assertions are inverted: the test passes
CI while the bug is present (all CLI-facing tests use mocks) and fails
once the bug is fixed (signalling that the tag should be removed).
Root cause
~~~~~~~~~~
All 21 CLI-facing tests in the M1-M6 E2E helpers use
``unittest.mock.patch`` to replace service factory functions with
``MagicMock`` objects and invoke the CLI via Typer's in-process
``CliRunner``. Zero tests invoke the actual ``agents`` CLI binary via
``subprocess.run``. This prevents the E2E suites from detecting DI
wiring bugs, database issues, or process-level failures.
"""
from __future__ import annotations
import ast
from dataclasses import dataclass, field
from pathlib import Path
from typing import Any
from behave import given, then, when
_REPO_ROOT = Path(__file__).resolve().parents[2]
_HELPER_GLOBS = [
"robot/helper_m1_e2e_verification.py",
"robot/helper_m2_e2e_verification.py",
"robot/helper_m3_e2e_verification.py",
"robot/helper_m4_e2e_verification.py",
"robot/helper_m5_e2e_verification.py",
"robot/helper_m6_e2e_verification.py",
]
# 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 # e.g. "helper_m1_e2e_verification"
results: list[FunctionAnalysis] = []
for node in ast.iter_child_nodes(tree):
if not isinstance(node, ast.FunctionDef):
continue
# Skip private/dunder helpers and the dispatcher
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
# Detect CliRunner.invoke or runner.invoke
if attr_name == "invoke" and _is_cli_runner_invoke(child):
fa.uses_cli_runner = True
# Detect subprocess.run / subprocess.Popen for CLI
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):
# Detect patch target strings that reference service factories
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 an attribute access looks like ``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 an attribute access looks like ``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
# ---------------------------------------------------------------------------
# Given steps
# ---------------------------------------------------------------------------
@given("the M1-M6 E2E verification helper files exist for e2e-mock-audit")
def step_helpers_exist(context: Any) -> None:
"""Verify that all 6 helper files exist on disk."""
missing: list[str] = []
for rel_path in _HELPER_GLOBS:
full = _REPO_ROOT / rel_path
if not full.is_file():
missing.append(rel_path)
assert not missing, f"Missing E2E helper files: {', '.join(missing)}"
context.e2e_mock_audit_helpers = [_REPO_ROOT / p for p in _HELPER_GLOBS]
# ---------------------------------------------------------------------------
# When steps
# ---------------------------------------------------------------------------
@when("I analyze mock and subprocess usage in M1-M6 helpers for e2e-mock-audit")
def step_analyze_all(context: Any) -> None:
"""Analyze all 6 helpers and store flat list of function analyses."""
all_functions: list[FunctionAnalysis] = []
for helper_path in context.e2e_mock_audit_helpers:
all_functions.extend(_analyze_helper(helper_path))
context.e2e_mock_audit_functions = all_functions
@when("I analyze mock and subprocess usage per suite for e2e-mock-audit")
def step_analyze_per_suite(context: Any) -> None:
"""Analyze all 6 helpers and store per-suite results."""
per_suite: dict[str, list[FunctionAnalysis]] = {}
for helper_path in context.e2e_mock_audit_helpers:
suite = helper_path.stem
per_suite[suite] = _analyze_helper(helper_path)
context.e2e_mock_audit_per_suite = per_suite
# ---------------------------------------------------------------------------
# Then steps
# ---------------------------------------------------------------------------
@then(
"at least one CLI-facing test should use subprocess instead of "
"CliRunner for e2e-mock-audit"
)
def step_at_least_one_subprocess(context: Any) -> None:
"""Assert that at least one CLI-facing function uses subprocess."""
functions: list[FunctionAnalysis] = context.e2e_mock_audit_functions
cli_facing = [f for f in functions if f.uses_cli_runner]
assert cli_facing, "No CLI-facing functions found at all"
subprocess_users = [f for f in cli_facing if f.uses_subprocess_cli]
assert subprocess_users, (
f"Bug #658 confirmed: {len(cli_facing)} CLI-facing E2E test "
f"functions found across M1-M6, but NONE invoke the real CLI "
f"via subprocess. All use Typer's in-process CliRunner.\n"
f"CLI-facing functions: "
f"{', '.join(f'{f.suite}::{f.name}' for f in cli_facing)}"
)
@then(
"at least one CLI-facing test should not mock the service factory "
"for e2e-mock-audit"
)
def step_at_least_one_unmocked(context: Any) -> None:
"""Assert that at least one CLI-facing function skips service mocking."""
functions: list[FunctionAnalysis] = context.e2e_mock_audit_functions
cli_facing = [f for f in functions if f.uses_cli_runner]
assert cli_facing, "No CLI-facing functions found at all"
unmocked = [f for f in cli_facing if not f.uses_mock_patch]
assert unmocked, (
f"Bug #658 confirmed: {len(cli_facing)} CLI-facing E2E test "
f"functions found across M1-M6, but ALL mock the service layer "
f"via unittest.mock.patch. None exercise the real DI container.\n"
f"Mocked functions: "
f"{', '.join(f'{f.suite}::{f.name}' for f in cli_facing)}"
)
@then(
"every suite with CLI tests should have at least one unmocked test "
"for e2e-mock-audit"
)
def step_every_suite_has_unmocked(context: Any) -> None:
"""Assert that no suite has 100% mocked CLI tests."""
per_suite: dict[str, list[FunctionAnalysis]] = context.e2e_mock_audit_per_suite
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 # Suite has no CLI tests — skip (e.g. M5)
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)"
)
assert not all_mocked_suites, (
"Bug #658 confirmed: the following suites have 100% mocked CLI "
"tests with no real code path coverage:\n"
+ "\n".join(f" - {s}" for s in all_mocked_suites)
)