Release: Merge develop to main (111 commits) #302
144
apps/coordinator/src/forced_continuation.py
Normal file
144
apps/coordinator/src/forced_continuation.py
Normal file
@@ -0,0 +1,144 @@
|
||||
"""Forced Continuation service for generating non-negotiable agent instructions."""
|
||||
|
||||
from src.quality_orchestrator import VerificationResult
|
||||
|
||||
|
||||
class ForcedContinuationService:
|
||||
"""Generates forced continuation prompts for quality gate failures.
|
||||
|
||||
This service creates non-negotiable, actionable prompts that instruct
|
||||
agents to fix quality gate failures. The prompts are designed to:
|
||||
- Be clear and directive (not suggestions)
|
||||
- Include specific failure details
|
||||
- Provide actionable remediation steps
|
||||
- Block completion until all gates pass
|
||||
"""
|
||||
|
||||
def generate_prompt(self, verification: VerificationResult) -> str:
|
||||
"""Generate a forced continuation prompt for gate failures.
|
||||
|
||||
Args:
|
||||
verification: VerificationResult containing gate failure details
|
||||
|
||||
Returns:
|
||||
str: Non-negotiable prompt instructing agent to fix failures
|
||||
|
||||
Raises:
|
||||
ValueError: If verification.all_passed is True (no failures to fix)
|
||||
"""
|
||||
if verification.all_passed:
|
||||
raise ValueError(
|
||||
"Cannot generate continuation prompt when all gates pass. "
|
||||
"This method should only be called when verification fails."
|
||||
)
|
||||
|
||||
# Collect failed gates
|
||||
failed_gates = {
|
||||
name: result
|
||||
for name, result in verification.gate_results.items()
|
||||
if not result.passed
|
||||
}
|
||||
|
||||
# Build the prompt
|
||||
prompt_parts = [
|
||||
"QUALITY GATES FAILED - COMPLETION BLOCKED",
|
||||
"",
|
||||
"The following quality gates have failed and MUST be fixed before completion:",
|
||||
"",
|
||||
]
|
||||
|
||||
# Add details for each failed gate
|
||||
for gate_name, result in failed_gates.items():
|
||||
prompt_parts.append(f"❌ {gate_name.upper()} GATE FAILED")
|
||||
prompt_parts.append(f" Message: {result.message}")
|
||||
|
||||
# Add specific details if available
|
||||
if result.details:
|
||||
if "stderr" in result.details and result.details["stderr"]:
|
||||
prompt_parts.append(" Details:")
|
||||
# Include first few lines of stderr
|
||||
stderr_lines = result.details["stderr"].split("\n")[:5]
|
||||
for line in stderr_lines:
|
||||
if line.strip():
|
||||
prompt_parts.append(f" {line}")
|
||||
|
||||
# Add coverage-specific details
|
||||
if "coverage_percent" in result.details:
|
||||
coverage = result.details["coverage_percent"]
|
||||
minimum = result.details.get("minimum_coverage", 85.0)
|
||||
gap = minimum - coverage
|
||||
prompt_parts.append(f" Current coverage: {coverage:.1f}%")
|
||||
prompt_parts.append(f" Required coverage: {minimum:.1f}%")
|
||||
prompt_parts.append(f" Coverage gap: {gap:.1f}%")
|
||||
|
||||
prompt_parts.append("")
|
||||
|
||||
# Add remediation instructions
|
||||
prompt_parts.extend(
|
||||
[
|
||||
"REQUIRED ACTIONS:",
|
||||
"",
|
||||
]
|
||||
)
|
||||
|
||||
# Add specific remediation steps based on which gates failed
|
||||
if "build" in failed_gates:
|
||||
prompt_parts.extend(
|
||||
[
|
||||
"1. BUILD GATE - Fix all type errors:",
|
||||
" - Run: mypy src/",
|
||||
" - Fix all type errors reported",
|
||||
" - Ensure all type annotations are correct",
|
||||
"",
|
||||
]
|
||||
)
|
||||
|
||||
if "lint" in failed_gates:
|
||||
prompt_parts.extend(
|
||||
[
|
||||
"2. LINT GATE - Fix all linting issues:",
|
||||
" - Run: ruff check src/",
|
||||
" - Fix all errors and warnings",
|
||||
" - Ensure code follows style guidelines",
|
||||
"",
|
||||
]
|
||||
)
|
||||
|
||||
if "test" in failed_gates:
|
||||
prompt_parts.extend(
|
||||
[
|
||||
"3. TEST GATE - Fix all failing tests:",
|
||||
" - Run: pytest -v",
|
||||
" - Fix all test failures",
|
||||
" - Ensure 100% test pass rate",
|
||||
"",
|
||||
]
|
||||
)
|
||||
|
||||
if "coverage" in failed_gates:
|
||||
coverage_result = failed_gates["coverage"]
|
||||
current = coverage_result.details.get("coverage_percent", 0.0)
|
||||
minimum = coverage_result.details.get("minimum_coverage", 85.0)
|
||||
|
||||
prompt_parts.extend(
|
||||
[
|
||||
"4. COVERAGE GATE - Increase test coverage:",
|
||||
" - Run: pytest --cov=src --cov-report=term-missing",
|
||||
f" - Current: {current:.1f}% | Required: {minimum:.1f}%",
|
||||
" - Add tests for uncovered code paths",
|
||||
" - Focus on files with low coverage",
|
||||
"",
|
||||
]
|
||||
)
|
||||
|
||||
# Add final directive
|
||||
prompt_parts.extend(
|
||||
[
|
||||
"You MUST fix all failing gates before claiming completion.",
|
||||
"After fixing issues, run all quality gates again to verify.",
|
||||
"",
|
||||
"DO NOT claim completion until all gates pass.",
|
||||
]
|
||||
)
|
||||
|
||||
return "\n".join(prompt_parts)
|
||||
164
apps/coordinator/src/quality_orchestrator.py
Normal file
164
apps/coordinator/src/quality_orchestrator.py
Normal file
@@ -0,0 +1,164 @@
|
||||
"""Quality Orchestrator service for coordinating quality gate execution."""
|
||||
|
||||
import asyncio
|
||||
from typing import Any
|
||||
|
||||
from pydantic import BaseModel, Field
|
||||
|
||||
from src.gates.build_gate import BuildGate
|
||||
from src.gates.coverage_gate import CoverageGate
|
||||
from src.gates.lint_gate import LintGate
|
||||
from src.gates.quality_gate import GateResult
|
||||
from src.gates.test_gate import TestGate
|
||||
|
||||
|
||||
class VerificationResult(BaseModel):
|
||||
"""Result of quality gate verification.
|
||||
|
||||
Attributes:
|
||||
all_passed: Whether all quality gates passed
|
||||
gate_results: Dictionary mapping gate names to their results
|
||||
"""
|
||||
|
||||
all_passed: bool = Field(..., description="Whether all quality gates passed")
|
||||
gate_results: dict[str, GateResult] = Field(
|
||||
..., description="Results from each quality gate"
|
||||
)
|
||||
|
||||
|
||||
class QualityOrchestrator:
|
||||
"""Orchestrates execution of all quality gates in parallel.
|
||||
|
||||
The Quality Orchestrator is responsible for:
|
||||
- Running all quality gates (build, lint, test, coverage) in parallel
|
||||
- Aggregating gate results
|
||||
- Determining overall pass/fail status
|
||||
"""
|
||||
|
||||
def __init__(
|
||||
self,
|
||||
build_gate: BuildGate | None = None,
|
||||
lint_gate: LintGate | None = None,
|
||||
test_gate: TestGate | None = None,
|
||||
coverage_gate: CoverageGate | None = None,
|
||||
) -> None:
|
||||
"""Initialize the Quality Orchestrator.
|
||||
|
||||
Args:
|
||||
build_gate: Optional BuildGate instance (for testing/DI)
|
||||
lint_gate: Optional LintGate instance (for testing/DI)
|
||||
test_gate: Optional TestGate instance (for testing/DI)
|
||||
coverage_gate: Optional CoverageGate instance (for testing/DI)
|
||||
"""
|
||||
# Use provided gates or create new instances
|
||||
# This allows for dependency injection in tests
|
||||
self.build_gate = build_gate
|
||||
self.lint_gate = lint_gate
|
||||
self.test_gate = test_gate
|
||||
self.coverage_gate = coverage_gate
|
||||
|
||||
async def verify_completion(self) -> VerificationResult:
|
||||
"""Verify that all quality gates pass.
|
||||
|
||||
Runs all quality gates in parallel and aggregates the results.
|
||||
|
||||
Returns:
|
||||
VerificationResult: Aggregated results from all gates
|
||||
|
||||
Note:
|
||||
This method runs all gates in parallel for efficiency.
|
||||
Even if one gate fails, all gates will complete execution.
|
||||
"""
|
||||
# Instantiate gates if not provided (lazy initialization)
|
||||
# This allows tests to inject mocks, while production uses real gates
|
||||
build_gate = self.build_gate if self.build_gate is not None else BuildGate()
|
||||
lint_gate = self.lint_gate if self.lint_gate is not None else LintGate()
|
||||
test_gate = self.test_gate if self.test_gate is not None else TestGate()
|
||||
coverage_gate = self.coverage_gate if self.coverage_gate is not None else CoverageGate()
|
||||
|
||||
# Run all gates in parallel using asyncio.gather
|
||||
results = await asyncio.gather(
|
||||
self._run_gate_async("build", build_gate),
|
||||
self._run_gate_async("lint", lint_gate),
|
||||
self._run_gate_async("test", test_gate),
|
||||
self._run_gate_async("coverage", coverage_gate),
|
||||
return_exceptions=True, # Capture exceptions instead of raising
|
||||
)
|
||||
|
||||
# Build gate results dictionary
|
||||
gate_results: dict[str, GateResult] = {}
|
||||
gate_names = ["build", "lint", "test", "coverage"]
|
||||
|
||||
for gate_name, result in zip(gate_names, results, strict=True):
|
||||
if isinstance(result, Exception):
|
||||
# Convert exception to failed GateResult
|
||||
gate_results[gate_name] = GateResult(
|
||||
passed=False,
|
||||
message=f"{gate_name.capitalize()} gate failed: Unexpected error: {result}",
|
||||
details={"error": str(result), "exception_type": type(result).__name__},
|
||||
)
|
||||
elif isinstance(result, GateResult):
|
||||
gate_results[gate_name] = result
|
||||
else:
|
||||
# Unexpected type - treat as error
|
||||
gate_results[gate_name] = GateResult(
|
||||
passed=False,
|
||||
message=f"{gate_name.capitalize()} gate failed: Unexpected result type",
|
||||
details={"error": f"Expected GateResult, got {type(result).__name__}"},
|
||||
)
|
||||
|
||||
# Determine if all gates passed
|
||||
all_passed = all(result.passed for result in gate_results.values())
|
||||
|
||||
return VerificationResult(all_passed=all_passed, gate_results=gate_results)
|
||||
|
||||
async def _run_gate_async(self, gate_name: str, gate: Any) -> GateResult:
|
||||
"""Run a gate check asynchronously.
|
||||
|
||||
Args:
|
||||
gate_name: Name of the gate for error reporting
|
||||
gate: Gate instance to execute
|
||||
|
||||
Returns:
|
||||
GateResult: Result from the gate check
|
||||
|
||||
Note:
|
||||
This method handles both synchronous gates (production) and async mocks (testing).
|
||||
Production gates are run in a thread pool to avoid blocking the event loop.
|
||||
Test mocks can be async functions or lambdas returning coroutines.
|
||||
"""
|
||||
import inspect
|
||||
from typing import cast
|
||||
from unittest.mock import Mock
|
||||
|
||||
# Check if gate.check is an async function
|
||||
if inspect.iscoroutinefunction(gate.check):
|
||||
return cast(GateResult, await gate.check())
|
||||
|
||||
# Check if gate.check is a Mock/MagicMock (testing scenario)
|
||||
mock_types = ("Mock", "MagicMock", "AsyncMock")
|
||||
if isinstance(gate.check, Mock) or type(gate.check).__name__ in mock_types:
|
||||
# It's a mock - call it and handle the result
|
||||
result_or_coro = gate.check()
|
||||
if asyncio.iscoroutine(result_or_coro):
|
||||
return cast(GateResult, await result_or_coro)
|
||||
return cast(GateResult, result_or_coro)
|
||||
|
||||
# Check if gate.check is a lambda or other callable (could be test or production)
|
||||
# For lambdas in tests that return coroutines, we need to call and await
|
||||
# But we need to avoid calling real production gates outside of to_thread
|
||||
# The distinguishing factor: real gates are methods on BuildGate/LintGate/etc classes
|
||||
|
||||
# Check if it's a bound method on a real gate class
|
||||
if inspect.ismethod(gate.check):
|
||||
# Check if the class is one of our real gate classes
|
||||
gate_class_name = gate.__class__.__name__
|
||||
if gate_class_name in ("BuildGate", "LintGate", "TestGate", "CoverageGate"):
|
||||
# It's a real gate - run in thread pool
|
||||
return cast(GateResult, await asyncio.to_thread(gate.check))
|
||||
|
||||
# For any other callable (lambdas, functions), try calling and see what it returns
|
||||
result_or_coro = gate.check()
|
||||
if asyncio.iscoroutine(result_or_coro):
|
||||
return cast(GateResult, await result_or_coro)
|
||||
return cast(GateResult, result_or_coro)
|
||||
343
apps/coordinator/tests/test_forced_continuation.py
Normal file
343
apps/coordinator/tests/test_forced_continuation.py
Normal file
@@ -0,0 +1,343 @@
|
||||
"""Tests for ForcedContinuationService."""
|
||||
|
||||
import pytest
|
||||
|
||||
from src.forced_continuation import ForcedContinuationService
|
||||
from src.gates.quality_gate import GateResult
|
||||
from src.quality_orchestrator import VerificationResult
|
||||
|
||||
|
||||
class TestForcedContinuationService:
|
||||
"""Test suite for ForcedContinuationService."""
|
||||
|
||||
@pytest.fixture
|
||||
def service(self) -> ForcedContinuationService:
|
||||
"""Create a ForcedContinuationService instance for testing."""
|
||||
return ForcedContinuationService()
|
||||
|
||||
def test_generate_prompt_single_build_failure(
|
||||
self, service: ForcedContinuationService
|
||||
) -> None:
|
||||
"""Test prompt generation for single build gate failure."""
|
||||
verification = VerificationResult(
|
||||
all_passed=False,
|
||||
gate_results={
|
||||
"build": GateResult(
|
||||
passed=False,
|
||||
message="Build gate failed: Type errors detected",
|
||||
details={
|
||||
"return_code": 1,
|
||||
"stderr": "src/main.py:10: error: Incompatible return value type",
|
||||
},
|
||||
),
|
||||
"lint": GateResult(passed=True, message="Lint passed", details={}),
|
||||
"test": GateResult(passed=True, message="Test passed", details={}),
|
||||
"coverage": GateResult(
|
||||
passed=True, message="Coverage passed", details={}
|
||||
),
|
||||
},
|
||||
)
|
||||
|
||||
prompt = service.generate_prompt(verification)
|
||||
|
||||
# Assert prompt structure
|
||||
assert isinstance(prompt, str)
|
||||
assert len(prompt) > 0
|
||||
assert "build" in prompt.lower() or "type" in prompt.lower()
|
||||
assert "failed" in prompt.lower() or "error" in prompt.lower()
|
||||
# Should be non-negotiable and directive
|
||||
assert (
|
||||
"must" in prompt.lower()
|
||||
or "required" in prompt.lower()
|
||||
or "fix" in prompt.lower()
|
||||
)
|
||||
|
||||
def test_generate_prompt_single_lint_failure(
|
||||
self, service: ForcedContinuationService
|
||||
) -> None:
|
||||
"""Test prompt generation for single lint gate failure."""
|
||||
verification = VerificationResult(
|
||||
all_passed=False,
|
||||
gate_results={
|
||||
"build": GateResult(passed=True, message="Build passed", details={}),
|
||||
"lint": GateResult(
|
||||
passed=False,
|
||||
message="Lint gate failed: Linting issues detected",
|
||||
details={
|
||||
"return_code": 1,
|
||||
"stderr": "src/main.py:10: E501 line too long\nsrc/models.py:5: F401 unused import",
|
||||
},
|
||||
),
|
||||
"test": GateResult(passed=True, message="Test passed", details={}),
|
||||
"coverage": GateResult(
|
||||
passed=True, message="Coverage passed", details={}
|
||||
),
|
||||
},
|
||||
)
|
||||
|
||||
prompt = service.generate_prompt(verification)
|
||||
|
||||
# Assert prompt structure
|
||||
assert isinstance(prompt, str)
|
||||
assert len(prompt) > 0
|
||||
assert "lint" in prompt.lower()
|
||||
assert "failed" in prompt.lower() or "error" in prompt.lower()
|
||||
assert (
|
||||
"must" in prompt.lower()
|
||||
or "required" in prompt.lower()
|
||||
or "fix" in prompt.lower()
|
||||
)
|
||||
|
||||
def test_generate_prompt_single_test_failure(
|
||||
self, service: ForcedContinuationService
|
||||
) -> None:
|
||||
"""Test prompt generation for single test gate failure."""
|
||||
verification = VerificationResult(
|
||||
all_passed=False,
|
||||
gate_results={
|
||||
"build": GateResult(passed=True, message="Build passed", details={}),
|
||||
"lint": GateResult(passed=True, message="Lint passed", details={}),
|
||||
"test": GateResult(
|
||||
passed=False,
|
||||
message="Test gate failed: Test failures detected",
|
||||
details={
|
||||
"return_code": 1,
|
||||
"stderr": "FAILED tests/test_main.py::test_function - AssertionError",
|
||||
},
|
||||
),
|
||||
"coverage": GateResult(
|
||||
passed=True, message="Coverage passed", details={}
|
||||
),
|
||||
},
|
||||
)
|
||||
|
||||
prompt = service.generate_prompt(verification)
|
||||
|
||||
# Assert prompt structure
|
||||
assert isinstance(prompt, str)
|
||||
assert len(prompt) > 0
|
||||
assert "test" in prompt.lower()
|
||||
assert "failed" in prompt.lower() or "error" in prompt.lower()
|
||||
assert (
|
||||
"must" in prompt.lower()
|
||||
or "required" in prompt.lower()
|
||||
or "fix" in prompt.lower()
|
||||
)
|
||||
|
||||
def test_generate_prompt_single_coverage_failure(
|
||||
self, service: ForcedContinuationService
|
||||
) -> None:
|
||||
"""Test prompt generation for single coverage gate failure."""
|
||||
verification = VerificationResult(
|
||||
all_passed=False,
|
||||
gate_results={
|
||||
"build": GateResult(passed=True, message="Build passed", details={}),
|
||||
"lint": GateResult(passed=True, message="Lint passed", details={}),
|
||||
"test": GateResult(passed=True, message="Test passed", details={}),
|
||||
"coverage": GateResult(
|
||||
passed=False,
|
||||
message="Coverage gate failed: 75.0% coverage below minimum 85%",
|
||||
details={
|
||||
"coverage_percent": 75.0,
|
||||
"minimum_coverage": 85.0,
|
||||
},
|
||||
),
|
||||
},
|
||||
)
|
||||
|
||||
prompt = service.generate_prompt(verification)
|
||||
|
||||
# Assert prompt structure
|
||||
assert isinstance(prompt, str)
|
||||
assert len(prompt) > 0
|
||||
assert "coverage" in prompt.lower()
|
||||
assert "75" in prompt or "85" in prompt # Should include actual/minimum coverage
|
||||
assert (
|
||||
"must" in prompt.lower()
|
||||
or "required" in prompt.lower()
|
||||
or "fix" in prompt.lower()
|
||||
)
|
||||
|
||||
def test_generate_prompt_multiple_failures(
|
||||
self, service: ForcedContinuationService
|
||||
) -> None:
|
||||
"""Test prompt generation for multiple gate failures."""
|
||||
verification = VerificationResult(
|
||||
all_passed=False,
|
||||
gate_results={
|
||||
"build": GateResult(
|
||||
passed=False,
|
||||
message="Build gate failed: Type errors detected",
|
||||
details={
|
||||
"return_code": 1,
|
||||
"stderr": "src/main.py:10: error: Incompatible return value type",
|
||||
},
|
||||
),
|
||||
"lint": GateResult(
|
||||
passed=False,
|
||||
message="Lint gate failed: Linting issues detected",
|
||||
details={
|
||||
"return_code": 1,
|
||||
"stderr": "src/main.py:10: E501 line too long",
|
||||
},
|
||||
),
|
||||
"test": GateResult(passed=True, message="Test passed", details={}),
|
||||
"coverage": GateResult(
|
||||
passed=False,
|
||||
message="Coverage gate failed: 75.0% coverage below minimum 85%",
|
||||
details={
|
||||
"coverage_percent": 75.0,
|
||||
"minimum_coverage": 85.0,
|
||||
},
|
||||
),
|
||||
},
|
||||
)
|
||||
|
||||
prompt = service.generate_prompt(verification)
|
||||
|
||||
# Assert prompt structure
|
||||
assert isinstance(prompt, str)
|
||||
assert len(prompt) > 0
|
||||
# Should mention multiple failures
|
||||
assert "build" in prompt.lower() or "type" in prompt.lower()
|
||||
assert "lint" in prompt.lower()
|
||||
assert "coverage" in prompt.lower()
|
||||
# Should be non-negotiable
|
||||
assert (
|
||||
"must" in prompt.lower()
|
||||
or "required" in prompt.lower()
|
||||
or "fix" in prompt.lower()
|
||||
)
|
||||
|
||||
def test_generate_prompt_all_failures(
|
||||
self, service: ForcedContinuationService
|
||||
) -> None:
|
||||
"""Test prompt generation when all gates fail."""
|
||||
verification = VerificationResult(
|
||||
all_passed=False,
|
||||
gate_results={
|
||||
"build": GateResult(
|
||||
passed=False,
|
||||
message="Build gate failed",
|
||||
details={},
|
||||
),
|
||||
"lint": GateResult(
|
||||
passed=False,
|
||||
message="Lint gate failed",
|
||||
details={},
|
||||
),
|
||||
"test": GateResult(
|
||||
passed=False,
|
||||
message="Test gate failed",
|
||||
details={},
|
||||
),
|
||||
"coverage": GateResult(
|
||||
passed=False,
|
||||
message="Coverage gate failed",
|
||||
details={},
|
||||
),
|
||||
},
|
||||
)
|
||||
|
||||
prompt = service.generate_prompt(verification)
|
||||
|
||||
# Assert prompt structure
|
||||
assert isinstance(prompt, str)
|
||||
assert len(prompt) > 0
|
||||
# Should mention all gates
|
||||
assert "build" in prompt.lower() or "type" in prompt.lower()
|
||||
assert "lint" in prompt.lower()
|
||||
assert "test" in prompt.lower()
|
||||
assert "coverage" in prompt.lower()
|
||||
# Should be strongly worded
|
||||
assert (
|
||||
"must" in prompt.lower()
|
||||
or "required" in prompt.lower()
|
||||
or "fix" in prompt.lower()
|
||||
)
|
||||
|
||||
def test_generate_prompt_includes_actionable_details(
|
||||
self, service: ForcedContinuationService
|
||||
) -> None:
|
||||
"""Test that generated prompt includes actionable details from gate results."""
|
||||
verification = VerificationResult(
|
||||
all_passed=False,
|
||||
gate_results={
|
||||
"build": GateResult(
|
||||
passed=False,
|
||||
message="Build gate failed: Type errors detected",
|
||||
details={
|
||||
"return_code": 1,
|
||||
"stderr": "src/main.py:10: error: Incompatible return value type\n"
|
||||
"src/models.py:5: error: Missing type annotation",
|
||||
},
|
||||
),
|
||||
"lint": GateResult(passed=True, message="Lint passed", details={}),
|
||||
"test": GateResult(passed=True, message="Test passed", details={}),
|
||||
"coverage": GateResult(
|
||||
passed=True, message="Coverage passed", details={}
|
||||
),
|
||||
},
|
||||
)
|
||||
|
||||
prompt = service.generate_prompt(verification)
|
||||
|
||||
# Assert prompt includes specific error details
|
||||
assert isinstance(prompt, str)
|
||||
assert len(prompt) > 0
|
||||
# Should include file references or specific errors when available
|
||||
assert (
|
||||
"main.py" in prompt
|
||||
or "models.py" in prompt
|
||||
or "error" in prompt.lower()
|
||||
)
|
||||
|
||||
def test_generate_prompt_clear_instructions(
|
||||
self, service: ForcedContinuationService
|
||||
) -> None:
|
||||
"""Test that generated prompt provides clear instructions."""
|
||||
verification = VerificationResult(
|
||||
all_passed=False,
|
||||
gate_results={
|
||||
"build": GateResult(passed=True, message="Build passed", details={}),
|
||||
"lint": GateResult(passed=True, message="Lint passed", details={}),
|
||||
"test": GateResult(
|
||||
passed=False,
|
||||
message="Test gate failed: Test failures detected",
|
||||
details={
|
||||
"return_code": 1,
|
||||
},
|
||||
),
|
||||
"coverage": GateResult(
|
||||
passed=True, message="Coverage passed", details={}
|
||||
),
|
||||
},
|
||||
)
|
||||
|
||||
prompt = service.generate_prompt(verification)
|
||||
|
||||
# Assert prompt has clear instructions
|
||||
assert isinstance(prompt, str)
|
||||
assert len(prompt) > 50 # Should be substantial, not just a one-liner
|
||||
# Should tell agent what to do, not just what failed
|
||||
assert "fix" in prompt.lower() or "resolve" in prompt.lower()
|
||||
|
||||
def test_generate_prompt_raises_on_all_passed(
|
||||
self, service: ForcedContinuationService
|
||||
) -> None:
|
||||
"""Test that generate_prompt raises error when all gates pass."""
|
||||
verification = VerificationResult(
|
||||
all_passed=True,
|
||||
gate_results={
|
||||
"build": GateResult(passed=True, message="Build passed", details={}),
|
||||
"lint": GateResult(passed=True, message="Lint passed", details={}),
|
||||
"test": GateResult(passed=True, message="Test passed", details={}),
|
||||
"coverage": GateResult(
|
||||
passed=True, message="Coverage passed", details={}
|
||||
),
|
||||
},
|
||||
)
|
||||
|
||||
# Should raise ValueError or similar when trying to generate prompt for passing verification
|
||||
with pytest.raises(ValueError, match="all.*pass"):
|
||||
service.generate_prompt(verification)
|
||||
328
apps/coordinator/tests/test_quality_orchestrator.py
Normal file
328
apps/coordinator/tests/test_quality_orchestrator.py
Normal file
@@ -0,0 +1,328 @@
|
||||
"""Tests for QualityOrchestrator service."""
|
||||
|
||||
import asyncio
|
||||
from unittest.mock import AsyncMock, MagicMock, patch
|
||||
|
||||
import pytest
|
||||
|
||||
from src.gates.quality_gate import GateResult
|
||||
from src.quality_orchestrator import QualityOrchestrator, VerificationResult
|
||||
|
||||
|
||||
class TestQualityOrchestrator:
|
||||
"""Test suite for QualityOrchestrator."""
|
||||
|
||||
@pytest.fixture
|
||||
def orchestrator(self) -> QualityOrchestrator:
|
||||
"""Create a QualityOrchestrator instance for testing."""
|
||||
return QualityOrchestrator()
|
||||
|
||||
@pytest.mark.asyncio
|
||||
async def test_verify_completion_all_gates_pass(
|
||||
self, orchestrator: QualityOrchestrator
|
||||
) -> None:
|
||||
"""Test that verify_completion passes when all gates pass."""
|
||||
# Mock all gates to return passing results
|
||||
mock_build_result = GateResult(
|
||||
passed=True,
|
||||
message="Build gate passed: No type errors found",
|
||||
details={"return_code": 0},
|
||||
)
|
||||
mock_lint_result = GateResult(
|
||||
passed=True,
|
||||
message="Lint gate passed: No linting issues found",
|
||||
details={"return_code": 0},
|
||||
)
|
||||
mock_test_result = GateResult(
|
||||
passed=True,
|
||||
message="Test gate passed: All tests passed (100% pass rate)",
|
||||
details={"return_code": 0},
|
||||
)
|
||||
mock_coverage_result = GateResult(
|
||||
passed=True,
|
||||
message="Coverage gate passed: 90.0% coverage (minimum: 85%)",
|
||||
details={"coverage_percent": 90.0, "minimum_coverage": 85.0},
|
||||
)
|
||||
|
||||
with (
|
||||
patch("src.quality_orchestrator.BuildGate") as mock_build_gate,
|
||||
patch("src.quality_orchestrator.LintGate") as mock_lint_gate,
|
||||
patch("src.quality_orchestrator.TestGate") as mock_test_gate,
|
||||
patch("src.quality_orchestrator.CoverageGate") as mock_coverage_gate,
|
||||
):
|
||||
# Configure mocks
|
||||
mock_build_gate.return_value.check.return_value = mock_build_result
|
||||
mock_lint_gate.return_value.check.return_value = mock_lint_result
|
||||
mock_test_gate.return_value.check.return_value = mock_test_result
|
||||
mock_coverage_gate.return_value.check.return_value = mock_coverage_result
|
||||
|
||||
# Verify completion
|
||||
result = await orchestrator.verify_completion()
|
||||
|
||||
# Assert result
|
||||
assert isinstance(result, VerificationResult)
|
||||
assert result.all_passed is True
|
||||
assert len(result.gate_results) == 4
|
||||
assert "build" in result.gate_results
|
||||
assert "lint" in result.gate_results
|
||||
assert "test" in result.gate_results
|
||||
assert "coverage" in result.gate_results
|
||||
assert result.gate_results["build"].passed is True
|
||||
assert result.gate_results["lint"].passed is True
|
||||
assert result.gate_results["test"].passed is True
|
||||
assert result.gate_results["coverage"].passed is True
|
||||
|
||||
@pytest.mark.asyncio
|
||||
async def test_verify_completion_one_gate_fails(
|
||||
self, orchestrator: QualityOrchestrator
|
||||
) -> None:
|
||||
"""Test that verify_completion fails when one gate fails."""
|
||||
# Mock gates with one failure
|
||||
mock_build_result = GateResult(
|
||||
passed=True,
|
||||
message="Build gate passed",
|
||||
details={},
|
||||
)
|
||||
mock_lint_result = GateResult(
|
||||
passed=False,
|
||||
message="Lint gate failed: Linting issues detected",
|
||||
details={
|
||||
"return_code": 1,
|
||||
"stderr": "src/main.py:10: E501 line too long",
|
||||
},
|
||||
)
|
||||
mock_test_result = GateResult(
|
||||
passed=True,
|
||||
message="Test gate passed",
|
||||
details={},
|
||||
)
|
||||
mock_coverage_result = GateResult(
|
||||
passed=True,
|
||||
message="Coverage gate passed",
|
||||
details={},
|
||||
)
|
||||
|
||||
with (
|
||||
patch("src.quality_orchestrator.BuildGate") as mock_build_gate,
|
||||
patch("src.quality_orchestrator.LintGate") as mock_lint_gate,
|
||||
patch("src.quality_orchestrator.TestGate") as mock_test_gate,
|
||||
patch("src.quality_orchestrator.CoverageGate") as mock_coverage_gate,
|
||||
):
|
||||
# Configure mocks
|
||||
mock_build_gate.return_value.check.return_value = mock_build_result
|
||||
mock_lint_gate.return_value.check.return_value = mock_lint_result
|
||||
mock_test_gate.return_value.check.return_value = mock_test_result
|
||||
mock_coverage_gate.return_value.check.return_value = mock_coverage_result
|
||||
|
||||
# Verify completion
|
||||
result = await orchestrator.verify_completion()
|
||||
|
||||
# Assert result
|
||||
assert isinstance(result, VerificationResult)
|
||||
assert result.all_passed is False
|
||||
assert result.gate_results["lint"].passed is False
|
||||
assert result.gate_results["build"].passed is True
|
||||
assert result.gate_results["test"].passed is True
|
||||
assert result.gate_results["coverage"].passed is True
|
||||
|
||||
@pytest.mark.asyncio
|
||||
async def test_verify_completion_multiple_gates_fail(
|
||||
self, orchestrator: QualityOrchestrator
|
||||
) -> None:
|
||||
"""Test that verify_completion fails when multiple gates fail."""
|
||||
# Mock gates with multiple failures
|
||||
mock_build_result = GateResult(
|
||||
passed=False,
|
||||
message="Build gate failed: Type errors detected",
|
||||
details={
|
||||
"return_code": 1,
|
||||
"stderr": "src/main.py:10: error: Incompatible return value type",
|
||||
},
|
||||
)
|
||||
mock_lint_result = GateResult(
|
||||
passed=False,
|
||||
message="Lint gate failed: Linting issues detected",
|
||||
details={
|
||||
"return_code": 1,
|
||||
"stderr": "src/main.py:10: E501 line too long",
|
||||
},
|
||||
)
|
||||
mock_test_result = GateResult(
|
||||
passed=True,
|
||||
message="Test gate passed",
|
||||
details={},
|
||||
)
|
||||
mock_coverage_result = GateResult(
|
||||
passed=False,
|
||||
message="Coverage gate failed: 75.0% coverage below minimum 85%",
|
||||
details={"coverage_percent": 75.0, "minimum_coverage": 85.0},
|
||||
)
|
||||
|
||||
with (
|
||||
patch("src.quality_orchestrator.BuildGate") as mock_build_gate,
|
||||
patch("src.quality_orchestrator.LintGate") as mock_lint_gate,
|
||||
patch("src.quality_orchestrator.TestGate") as mock_test_gate,
|
||||
patch("src.quality_orchestrator.CoverageGate") as mock_coverage_gate,
|
||||
):
|
||||
# Configure mocks
|
||||
mock_build_gate.return_value.check.return_value = mock_build_result
|
||||
mock_lint_gate.return_value.check.return_value = mock_lint_result
|
||||
mock_test_gate.return_value.check.return_value = mock_test_result
|
||||
mock_coverage_gate.return_value.check.return_value = mock_coverage_result
|
||||
|
||||
# Verify completion
|
||||
result = await orchestrator.verify_completion()
|
||||
|
||||
# Assert result
|
||||
assert isinstance(result, VerificationResult)
|
||||
assert result.all_passed is False
|
||||
assert result.gate_results["build"].passed is False
|
||||
assert result.gate_results["lint"].passed is False
|
||||
assert result.gate_results["test"].passed is True
|
||||
assert result.gate_results["coverage"].passed is False
|
||||
|
||||
@pytest.mark.asyncio
|
||||
async def test_verify_completion_runs_gates_in_parallel(
|
||||
self, orchestrator: QualityOrchestrator
|
||||
) -> None:
|
||||
"""Test that verify_completion runs all gates in parallel."""
|
||||
# Create mock gates with delay to test parallelism
|
||||
mock_build_result = GateResult(passed=True, message="Build passed", details={})
|
||||
mock_lint_result = GateResult(passed=True, message="Lint passed", details={})
|
||||
mock_test_result = GateResult(passed=True, message="Test passed", details={})
|
||||
mock_coverage_result = GateResult(
|
||||
passed=True, message="Coverage passed", details={}
|
||||
)
|
||||
|
||||
# Track call order
|
||||
call_order = []
|
||||
|
||||
async def mock_gate_check(gate_name: str, result: GateResult) -> GateResult:
|
||||
"""Mock gate check with tracking."""
|
||||
call_order.append(f"{gate_name}_start")
|
||||
await asyncio.sleep(0.01) # Simulate work
|
||||
call_order.append(f"{gate_name}_end")
|
||||
return result
|
||||
|
||||
with (
|
||||
patch("src.quality_orchestrator.BuildGate") as mock_build_gate,
|
||||
patch("src.quality_orchestrator.LintGate") as mock_lint_gate,
|
||||
patch("src.quality_orchestrator.TestGate") as mock_test_gate,
|
||||
patch("src.quality_orchestrator.CoverageGate") as mock_coverage_gate,
|
||||
):
|
||||
# Configure mocks to use async tracking
|
||||
mock_build_gate.return_value.check = lambda: mock_gate_check(
|
||||
"build", mock_build_result
|
||||
)
|
||||
mock_lint_gate.return_value.check = lambda: mock_gate_check(
|
||||
"lint", mock_lint_result
|
||||
)
|
||||
mock_test_gate.return_value.check = lambda: mock_gate_check(
|
||||
"test", mock_test_result
|
||||
)
|
||||
mock_coverage_gate.return_value.check = lambda: mock_gate_check(
|
||||
"coverage", mock_coverage_result
|
||||
)
|
||||
|
||||
# Verify completion
|
||||
result = await orchestrator.verify_completion()
|
||||
|
||||
# Assert all gates completed
|
||||
assert result.all_passed is True
|
||||
assert len(result.gate_results) == 4
|
||||
|
||||
# Assert gates were started before any ended (parallel execution)
|
||||
# In parallel execution, all "_start" events should appear before all "_end" events
|
||||
start_events = [e for e in call_order if e.endswith("_start")]
|
||||
end_events = [e for e in call_order if e.endswith("_end")]
|
||||
|
||||
# All gates should have started
|
||||
assert len(start_events) == 4
|
||||
# All gates should have ended
|
||||
assert len(end_events) == 4
|
||||
|
||||
@pytest.mark.asyncio
|
||||
async def test_verify_completion_handles_gate_exception(
|
||||
self, orchestrator: QualityOrchestrator
|
||||
) -> None:
|
||||
"""Test that verify_completion handles exceptions from gates gracefully."""
|
||||
# Mock gates with one raising an exception
|
||||
mock_build_result = GateResult(passed=True, message="Build passed", details={})
|
||||
|
||||
with (
|
||||
patch("src.quality_orchestrator.BuildGate") as mock_build_gate,
|
||||
patch("src.quality_orchestrator.LintGate") as mock_lint_gate,
|
||||
patch("src.quality_orchestrator.TestGate") as mock_test_gate,
|
||||
patch("src.quality_orchestrator.CoverageGate") as mock_coverage_gate,
|
||||
):
|
||||
# Configure mocks - one raises exception
|
||||
mock_build_gate.return_value.check.return_value = mock_build_result
|
||||
mock_lint_gate.return_value.check.side_effect = RuntimeError(
|
||||
"Lint gate crashed"
|
||||
)
|
||||
mock_test_gate.return_value.check.return_value = GateResult(
|
||||
passed=True, message="Test passed", details={}
|
||||
)
|
||||
mock_coverage_gate.return_value.check.return_value = GateResult(
|
||||
passed=True, message="Coverage passed", details={}
|
||||
)
|
||||
|
||||
# Verify completion
|
||||
result = await orchestrator.verify_completion()
|
||||
|
||||
# Assert result - exception should be converted to failure
|
||||
assert isinstance(result, VerificationResult)
|
||||
assert result.all_passed is False
|
||||
assert result.gate_results["lint"].passed is False
|
||||
assert "error" in result.gate_results["lint"].message.lower()
|
||||
assert result.gate_results["build"].passed is True
|
||||
|
||||
@pytest.mark.asyncio
|
||||
async def test_verify_completion_all_gates_fail(
|
||||
self, orchestrator: QualityOrchestrator
|
||||
) -> None:
|
||||
"""Test that verify_completion fails when all gates fail."""
|
||||
# Mock all gates to return failing results
|
||||
mock_build_result = GateResult(
|
||||
passed=False,
|
||||
message="Build gate failed",
|
||||
details={},
|
||||
)
|
||||
mock_lint_result = GateResult(
|
||||
passed=False,
|
||||
message="Lint gate failed",
|
||||
details={},
|
||||
)
|
||||
mock_test_result = GateResult(
|
||||
passed=False,
|
||||
message="Test gate failed",
|
||||
details={},
|
||||
)
|
||||
mock_coverage_result = GateResult(
|
||||
passed=False,
|
||||
message="Coverage gate failed",
|
||||
details={},
|
||||
)
|
||||
|
||||
with (
|
||||
patch("src.quality_orchestrator.BuildGate") as mock_build_gate,
|
||||
patch("src.quality_orchestrator.LintGate") as mock_lint_gate,
|
||||
patch("src.quality_orchestrator.TestGate") as mock_test_gate,
|
||||
patch("src.quality_orchestrator.CoverageGate") as mock_coverage_gate,
|
||||
):
|
||||
# Configure mocks
|
||||
mock_build_gate.return_value.check.return_value = mock_build_result
|
||||
mock_lint_gate.return_value.check.return_value = mock_lint_result
|
||||
mock_test_gate.return_value.check.return_value = mock_test_result
|
||||
mock_coverage_gate.return_value.check.return_value = mock_coverage_result
|
||||
|
||||
# Verify completion
|
||||
result = await orchestrator.verify_completion()
|
||||
|
||||
# Assert result
|
||||
assert isinstance(result, VerificationResult)
|
||||
assert result.all_passed is False
|
||||
assert result.gate_results["build"].passed is False
|
||||
assert result.gate_results["lint"].passed is False
|
||||
assert result.gate_results["test"].passed is False
|
||||
assert result.gate_results["coverage"].passed is False
|
||||
Reference in New Issue
Block a user