cleveragents-core/benchmarks/cli_extensions_bench.py

"""ASV benchmarks for CLI extensions parsing overhead.

Measures the performance of:
- plan use with automation profile and invariant flags
- plan use with actor override flags and validation
- plan status / list rendering with extended fields
- action show rendering with optional actors/invariants/inputs_schema
"""

from __future__ import annotations

import importlib
import sys
from datetime import datetime
from pathlib import Path
from unittest.mock import MagicMock, patch

# Ensure the local *source* tree is importable even when ASV has an
# older build of the package installed.
_SRC = str(Path(__file__).resolve().parents[1] / "src")
if _SRC not in sys.path:
    sys.path.insert(0, _SRC)

import cleveragents  # noqa: E402

importlib.reload(cleveragents)

from typer.testing import CliRunner  # noqa: E402

from cleveragents.cli.commands.action import app as action_app  # noqa: E402
from cleveragents.cli.commands.plan import app as plan_app  # noqa: E402
from cleveragents.domain.models.core.action import (  # noqa: E402
    Action,
    ActionState,
)
from cleveragents.domain.models.core.plan import (  # noqa: E402
    AutomationProfileProvenance,
    AutomationProfileRef,
    InvariantSource,
    NamespacedName,
    Plan,
    PlanIdentity,
    PlanInvariant,
    PlanPhase,
    PlanTimestamps,
    ProcessingState,
    ProjectLink,
)

_runner = CliRunner()
_PLAN_ULID = "01KHDE6WWS2171PWW3GJEBXZ8S"


def _mock_action(name: str = "local/bench-action") -> Action:
    return Action(
        namespaced_name=NamespacedName.parse(name),
        description="Benchmark action",
        long_description=None,
        definition_of_done="Benchmarks pass",
        strategy_actor="openai/gpt-4",
        execution_actor="openai/gpt-4",
        estimation_actor="openai/gpt-4",
        invariant_actor="anthropic/claude-3",
        reusable=True,
        read_only=False,
        invariants=["No regressions", "Keep backward compat"],
        inputs_schema={
            "type": "object",
            "properties": {"coverage": {"type": "integer"}},
        },
        state=ActionState.AVAILABLE,
        created_by=None,
        created_at=datetime.now(),
        updated_at=datetime.now(),
    )


def _mock_plan(
    name: str = "local/bench-plan",
    *,
    with_profile: bool = False,
    with_invariants: bool = False,
) -> Plan:
    now = datetime.now()
    profile = None
    if with_profile:
        profile = AutomationProfileRef(
            profile_name="trusted",
            provenance=AutomationProfileProvenance.PLAN,
        )
    invariants: list[PlanInvariant] = []
    if with_invariants:
        invariants = [
            PlanInvariant(text="No warnings", source=InvariantSource.PLAN),
            PlanInvariant(text="Keep compat", source=InvariantSource.ACTION),
        ]
    return Plan(
        identity=PlanIdentity(plan_id=_PLAN_ULID),
        namespaced_name=NamespacedName.parse(name),
        description="Benchmark plan",
        definition_of_done="Benchmarks pass",
        action_name="local/bench-action",
        phase=PlanPhase.STRATEGIZE,
        processing_state=ProcessingState.QUEUED,
        project_links=[ProjectLink(project_name="my-project")],
        arguments={},
        arguments_order=[],
        automation_profile=profile,
        invariants=invariants,
        strategy_actor="openai/gpt-4",
        execution_actor="openai/gpt-4",
        reusable=True,
        read_only=False,
        created_by=None,
        timestamps=PlanTimestamps(created_at=now, updated_at=now),
    )


class PlanUseWithProfileSuite:
    """Benchmark plan use with --automation-profile."""

    def setup(self) -> None:
        self._mock_service = MagicMock()
        self._mock_service.get_action_by_name.return_value = _mock_action()
        self._mock_service.use_action.return_value = _mock_plan(with_profile=True)
        self._patcher = patch(
            "cleveragents.cli.commands.plan._get_lifecycle_service",
            return_value=self._mock_service,
        )
        self._patcher.start()

    def teardown(self) -> None:
        self._patcher.stop()

    def time_plan_use_with_profile(self) -> None:
        """Benchmark plan use with automation profile."""
        _runner.invoke(
            plan_app,
            [
                "use",
                "local/bench-action",
                "--automation-profile",
                "trusted",
            ],
        )


class PlanUseWithInvariantsSuite:
    """Benchmark plan use with --invariant flags."""

    def setup(self) -> None:
        self._mock_service = MagicMock()
        self._mock_service.get_action_by_name.return_value = _mock_action()
        self._mock_service.use_action.return_value = _mock_plan(with_invariants=True)
        self._patcher = patch(
            "cleveragents.cli.commands.plan._get_lifecycle_service",
            return_value=self._mock_service,
        )
        self._patcher.start()

    def teardown(self) -> None:
        self._patcher.stop()

    def time_plan_use_with_invariants(self) -> None:
        """Benchmark plan use with invariant flags."""
        _runner.invoke(
            plan_app,
            [
                "use",
                "local/bench-action",
                "--invariant",
                "No warnings",
                "--invariant",
                "Keep compat",
            ],
        )


class PlanUseActorValidationSuite:
    """Benchmark plan use with actor override flags + validation."""

    def setup(self) -> None:
        self._mock_service = MagicMock()
        self._mock_service.get_action_by_name.return_value = _mock_action()
        self._mock_service.use_action.return_value = _mock_plan()
        self._patcher = patch(
            "cleveragents.cli.commands.plan._get_lifecycle_service",
            return_value=self._mock_service,
        )
        self._patcher.start()

    def teardown(self) -> None:
        self._patcher.stop()

    def time_plan_use_with_actor_overrides(self) -> None:
        """Benchmark plan use with all actor overrides."""
        _runner.invoke(
            plan_app,
            [
                "use",
                "local/bench-action",
                "--strategy-actor",
                "openai/gpt-4",
                "--execution-actor",
                "anthropic/claude-3",
                "--estimation-actor",
                "openai/gpt-4",
                "--invariant-actor",
                "anthropic/claude-3",
            ],
        )


class PlanStatusWithExtendedFieldsSuite:
    """Benchmark plan status with automation profile + invariants."""

    def setup(self) -> None:
        self._mock_service = MagicMock()
        self._mock_service.list_plans.return_value = [
            _mock_plan(f"local/plan-{i}", with_profile=True, with_invariants=True)
            for i in range(20)
        ]
        self._patcher = patch(
            "cleveragents.cli.commands.plan._get_lifecycle_service",
            return_value=self._mock_service,
        )
        self._patcher.start()

    def teardown(self) -> None:
        self._patcher.stop()

    def time_plan_status_table(self) -> None:
        """Benchmark plan status table rendering with extended fields."""
        _runner.invoke(plan_app, ["status"])

    def time_plan_list_table(self) -> None:
        """Benchmark list table rendering with extended fields."""
        _runner.invoke(plan_app, ["list"])


class ActionShowExtendedSuite:
    """Benchmark action show with optional actors/invariants/schema."""

    def setup(self) -> None:
        self._mock_service = MagicMock()
        self._mock_service.get_action_by_name.return_value = _mock_action()
        self._patcher = patch(
            "cleveragents.cli.commands.action._get_lifecycle_service",
            return_value=self._mock_service,
        )
        self._patcher.start()

    def teardown(self) -> None:
        self._patcher.stop()

    def time_action_show_rich(self) -> None:
        """Benchmark action show in rich format."""
        _runner.invoke(action_app, ["show", "local/bench-action"])

    def time_action_show_json(self) -> None:
        """Benchmark action show in JSON format."""
        _runner.invoke(action_app, ["show", "local/bench-action", "--format", "json"])