infoxtractor/tests/integration/test_rest_adapter.py

"""Integration tests for the FastAPI REST adapter (spec §5).

Uses ``fastapi.testclient.TestClient`` against a real DB. Ollama / OCR probes
are stubbed via the DI hooks the routes expose for testing — in Chunk 4 the
production probes swap in.
"""

from __future__ import annotations

from collections.abc import Iterator
from uuid import uuid4

import pytest
from fastapi.testclient import TestClient
from sqlalchemy.ext.asyncio import async_sessionmaker, create_async_engine

from ix.adapters.rest.routes import Probes, get_probes, get_session_factory_dep
from ix.app import create_app


def _factory_for_url(postgres_url: str):  # type: ignore[no-untyped-def]
    """Build a TestClient-compatible session factory.

    TestClient runs the ASGI app on its own dedicated event loop (the one it
    creates in its sync wrapper), distinct from the per-test loop
    pytest-asyncio gives direct tests. Session factories must therefore be
    constructed from an engine that was itself created on that inner loop.
    We do this lazily: each dependency resolution creates a fresh engine +
    factory on the current running loop, which is the TestClient's loop at
    route-invocation time. Engine reuse would drag the cross-loop futures
    that asyncpg hates back in.
    """

    def _factory():  # type: ignore[no-untyped-def]
        eng = create_async_engine(postgres_url, pool_pre_ping=True)
        return async_sessionmaker(eng, expire_on_commit=False)

    return _factory


@pytest.fixture
def app(postgres_url: str) -> Iterator[TestClient]:
    """Spin up the FastAPI app wired to the test DB + stub probes."""

    app_obj = create_app(spawn_worker=False)
    app_obj.dependency_overrides[get_session_factory_dep] = _factory_for_url(
        postgres_url
    )
    app_obj.dependency_overrides[get_probes] = lambda: Probes(
        ollama=lambda: "ok",
        ocr=lambda: "ok",
    )
    with TestClient(app_obj) as client:
        yield client


def _valid_request_body(client_id: str = "mammon", request_id: str = "r-1") -> dict:
    return {
        "use_case": "bank_statement_header",
        "ix_client_id": client_id,
        "request_id": request_id,
        "context": {"texts": ["hello world"]},
    }


def test_post_jobs_creates_pending(app: TestClient) -> None:
    resp = app.post("/jobs", json=_valid_request_body())
    assert resp.status_code == 201, resp.text
    body = resp.json()
    assert body["status"] == "pending"
    assert len(body["ix_id"]) == 16
    assert body["job_id"]


def test_post_jobs_idempotent_returns_200(app: TestClient) -> None:
    first = app.post("/jobs", json=_valid_request_body("m", "dup"))
    assert first.status_code == 201
    first_body = first.json()

    second = app.post("/jobs", json=_valid_request_body("m", "dup"))
    assert second.status_code == 200
    second_body = second.json()
    assert second_body["job_id"] == first_body["job_id"]
    assert second_body["ix_id"] == first_body["ix_id"]


def test_get_job_by_id(app: TestClient) -> None:
    created = app.post("/jobs", json=_valid_request_body()).json()
    resp = app.get(f"/jobs/{created['job_id']}")
    assert resp.status_code == 200
    body = resp.json()
    assert body["job_id"] == created["job_id"]
    assert body["request"]["use_case"] == "bank_statement_header"
    assert body["status"] == "pending"


def test_get_job_404(app: TestClient) -> None:
    resp = app.get(f"/jobs/{uuid4()}")
    assert resp.status_code == 404


def test_get_by_correlation_query(app: TestClient) -> None:
    created = app.post("/jobs", json=_valid_request_body("mammon", "corr-1")).json()
    resp = app.get("/jobs", params={"client_id": "mammon", "request_id": "corr-1"})
    assert resp.status_code == 200
    assert resp.json()["job_id"] == created["job_id"]

    missing = app.get("/jobs", params={"client_id": "mammon", "request_id": "nope"})
    assert missing.status_code == 404


def test_healthz_all_ok(app: TestClient) -> None:
    resp = app.get("/healthz")
    assert resp.status_code == 200, resp.text
    body = resp.json()
    assert body["postgres"] == "ok"
    assert body["ollama"] == "ok"
    assert body["ocr"] == "ok"


def test_healthz_503_on_postgres_fail(postgres_url: str) -> None:
    """Broken postgres probe → 503. Ollama/OCR still surface in the body."""

    app_obj = create_app(spawn_worker=False)

    def _bad_factory():  # type: ignore[no-untyped-def]
        def _raise():  # type: ignore[no-untyped-def]
            raise RuntimeError("db down")

        return _raise

    app_obj.dependency_overrides[get_session_factory_dep] = _bad_factory
    app_obj.dependency_overrides[get_probes] = lambda: Probes(
        ollama=lambda: "ok", ocr=lambda: "ok"
    )

    with TestClient(app_obj) as client:
        resp = client.get("/healthz")
    assert resp.status_code == 503
    body = resp.json()
    assert body["postgres"] == "fail"


def test_healthz_degraded_ollama_is_503(postgres_url: str) -> None:
    """Per spec §5: degraded flips HTTP to 503 (only all-ok yields 200)."""

    app_obj = create_app(spawn_worker=False)
    app_obj.dependency_overrides[get_session_factory_dep] = _factory_for_url(
        postgres_url
    )
    app_obj.dependency_overrides[get_probes] = lambda: Probes(
        ollama=lambda: "degraded", ocr=lambda: "ok"
    )

    with TestClient(app_obj) as client:
        resp = client.get("/healthz")
    assert resp.status_code == 503
    assert resp.json()["ollama"] == "degraded"


def test_metrics_shape(app: TestClient) -> None:
    # Submit a couple of pending jobs to populate counters.
    app.post("/jobs", json=_valid_request_body("mm", "a"))
    app.post("/jobs", json=_valid_request_body("mm", "b"))

    resp = app.get("/metrics")
    assert resp.status_code == 200
    body = resp.json()
    for key in (
        "jobs_pending",
        "jobs_running",
        "jobs_done_24h",
        "jobs_error_24h",
        "by_use_case_seconds",
    ):
        assert key in body
    assert body["jobs_pending"] == 2
    assert body["jobs_running"] == 0
    assert isinstance(body["by_use_case_seconds"], dict)