document_analyzer/tests/acceptance/conftest.py

"""Pytest configuration and shared fixtures for QE acceptance tests.

Usage::

    pytest tests/acceptance/ -v --run-acceptance [--acceptance-runs=3]

Environment variables:
    DASHSCOPE_API_KEY  — LLM API key (required for Layers B/C)
    TEST_IR_PATH       — path to IR JSON to validate (default: ir_final.json sample)
    TEST_PARSED_PATH   — path to _parsed.json or _updated.json for coverage analysis
"""

from __future__ import annotations

import json
import os
import sys
import tempfile
from pathlib import Path
from typing import Any

import pytest

# ── Path setup ──────────────────────────────────────────────────────────────

_PROJECT_ROOT = Path(__file__).resolve().parent.parent.parent
sys.path.insert(0, str(_PROJECT_ROOT))


def _skill_path(skill_name: str) -> str:
    return str(_PROJECT_ROOT / "skills" / skill_name / "scripts")


# ── pytest configuration ────────────────────────────────────────────────────


def pytest_addoption(parser):
    parser.addoption(
        "--run-acceptance",
        action="store_true",
        default=False,
        help="Run QE acceptance tests (requires DASHSCOPE_API_KEY)",
    )
    parser.addoption(
        "--acceptance-runs",
        type=int,
        default=1,
        help="Number of IR generation runs for Layer B stability testing (default: 1 = skip)",
    )
    parser.addoption(
        "--ir-path",
        type=str,
        default=None,
        help="Path to IR JSON file to validate",
    )
    parser.addoption(
        "--parsed-path",
        type=str,
        default=None,
        help="Path to _parsed.json or _updated.json for coverage analysis",
    )


def pytest_configure(config):
    config.addinivalue_line(
        "markers",
        "acceptance: QE acceptance test (requires --run-acceptance flag and DASHSCOPE_API_KEY)",
    )


def pytest_collection_modifyitems(config, items):
    acceptance_dir = str(_PROJECT_ROOT / "tests" / "acceptance")
    acceptance_items = [i for i in items if str(i.fspath).startswith(acceptance_dir)]
    non_acceptance_items = [i for i in items if not str(i.fspath).startswith(acceptance_dir)]

    if not config.getoption("--run-acceptance"):
        skip_msg = pytest.mark.skip(reason="Need --run-acceptance flag to run")
        for item in acceptance_items:
            item.add_marker(skip_msg)
        # Don't skip non-acceptance tests
        return

    if not os.environ.get("DASHSCOPE_API_KEY"):
        skip_msg = pytest.mark.skip(reason="DASHSCOPE_API_KEY not set")
        for item in acceptance_items:
            item.add_marker(skip_msg)


# ── Shared fixtures ─────────────────────────────────────────────────────────


@pytest.fixture(scope="session")
def project_root() -> Path:
    return _PROJECT_ROOT


@pytest.fixture(scope="session")
def ir_path(request) -> str:
    """Path to the IR JSON file under test."""
    path = (
        request.config.getoption("--ir-path")
        or os.environ.get("TEST_IR_PATH")
        or str(
            Path.home()
            / ".openclaw/workspace/skills/doc_parser_skill/output/ir_final.json"
        )
    )
    if not os.path.exists(path):
        pytest.skip(f"IR file not found: {path}")
    return path


@pytest.fixture(scope="session")
def ir_data(ir_path: str) -> dict:
    """Load the IR JSON data."""
    with open(ir_path, "r", encoding="utf-8") as f:
        return json.load(f)


@pytest.fixture(scope="session")
def parsed_path(request) -> str | None:
    """Path to the corresponding _parsed.json or _updated.json."""
    path = (
        request.config.getoption("--parsed-path")
        or os.environ.get("TEST_PARSED_PATH")
        or str(
            _PROJECT_ROOT
            / "skills/ir_generation_skill/车机娱乐系统禁止功能文档_精简_updated.json"
        )
    )
    if os.path.exists(path):
        return path
    return None


@pytest.fixture(scope="session")
def parsed_data(parsed_path: str | None) -> dict | None:
    """Load the parsed document JSON for coverage analysis."""
    if parsed_path is None:
        return None
    with open(parsed_path, "r", encoding="utf-8") as f:
        return json.load(f)


@pytest.fixture(scope="session")
def llm_client():
    """Create an LLMClient instance for acceptance tests.

    Uses the DashScope-compatible LLMClient from the project.
    """
    sys.path.insert(0, _skill_path("doc_parser_skill"))
    from LLM import LLMClient

    return LLMClient()


@pytest.fixture(scope="session")
def acceptance_runs(request) -> int:
    return request.config.getoption("--acceptance-runs", default=1)


# ── Pipeline runner ─────────────────────────────────────────────────────────


@pytest.fixture(scope="session")
def run_ir_pipeline():
    """Return a callable that runs the IR generation pipeline on a parsed JSON.

    Usage::

        ir_data, ir_path = run_ir_pipeline(parsed_json_path, output_dir)
    """
    sys.path.insert(0, _skill_path("ir_generation_skill"))
    from ir_generator import generate_ir

    def _run(parsed_path: str, output_dir: str | None = None) -> tuple[dict, str]:
        """Run IR generation and return (ir_data, ir_path)."""
        out = output_dir or tempfile.mkdtemp(prefix="qe_acceptance_")
        result = generate_ir(parsed_path, out, dry_run=False)
        ir_list = result.get("ir", [])
        ir_path = result.get("path", "")
        # ir_generator produces a list; wrap to match rich format expectations
        # for schema validation we accept both formats
        return ir_list, ir_path

    return _run