docqa/tests/conftest.py

"""Shared pytest fixtures."""

from __future__ import annotations

import json
import sys
from pathlib import Path

import pytest

ROOT = Path(__file__).resolve().parent.parent
sys.path.insert(0, str(ROOT / "src"))

CORPUS = ROOT / "corpus"


@pytest.fixture(scope="session")
def corpus_dir() -> Path:
    if not (CORPUS / "ground_truth.json").exists():
        pytest.skip("corpus not generated; run: python tests/corpus.py corpus 12")
    return CORPUS


@pytest.fixture(scope="session")
def ground_truth(corpus_dir: Path) -> list[dict]:
    return json.loads((corpus_dir / "ground_truth.json").read_text())


@pytest.fixture(scope="session")
def settings():
    from docxextract.config import Settings
    return Settings(torch_threads=4, parse_cache_size=32)


@pytest.fixture(scope="session")
def sample_pdf(ground_truth) -> bytes:
    return Path(ground_truth[0]["path"]).read_bytes()


@pytest.fixture(scope="session")
def engine(settings):
    """Session-scoped warmed engine. Loading a model per test is too slow."""
    from docxextract.engine import Engine, reset_engine
    eng = Engine(settings)
    eng.warmup()
    yield eng
    reset_engine()