FabledCurator/tests/test_api_import_admin.py

from datetime import UTC, datetime, timedelta

import pytest

from backend.app.celery_app import celery
from backend.app.models import ImportBatch, ImportTask

pytestmark = pytest.mark.integration


@pytest.fixture(autouse=True)
def eager():
    celery.conf.task_always_eager = True
    yield
    celery.conf.task_always_eager = False


@pytest.mark.asyncio
async def test_status_when_idle(client):
    resp = await client.get("/api/import/status")
    body = await resp.get_json()
    assert body["active_batch"] is None


@pytest.mark.asyncio
async def test_list_tasks_with_status_filter(client, db):
    batch = ImportBatch(triggered_by="manual", source_path="/import", scan_mode="quick")
    db.add(batch)
    await db.flush()
    for status in ("complete", "failed", "complete"):
        db.add(ImportTask(
            batch_id=batch.id, source_path="/x", task_type="media", status=status,
            finished_at=datetime.now(UTC),
        ))
    await db.commit()

    resp = await client.get("/api/import/tasks?status=failed")
    body = await resp.get_json()
    assert len(body["tasks"]) == 1
    assert body["tasks"][0]["status"] == "failed"


@pytest.mark.asyncio
async def test_retry_failed(client, db):
    batch = ImportBatch(triggered_by="manual", source_path="/import", scan_mode="quick")
    db.add(batch)
    await db.flush()
    db.add(ImportTask(batch_id=batch.id, source_path="/x", task_type="media", status="failed"))
    await db.commit()

    resp = await client.post("/api/import/retry-failed")
    body = await resp.get_json()
    assert body["retried"] == 1


@pytest.mark.asyncio
async def test_clear_completed(client, db):
    batch = ImportBatch(triggered_by="manual", source_path="/import", scan_mode="quick")
    db.add(batch)
    await db.flush()
    long_ago = datetime.now(UTC) - timedelta(days=30)
    db.add(ImportTask(
        batch_id=batch.id, source_path="/x", task_type="media",
        status="complete", finished_at=long_ago,
    ))
    db.add(ImportTask(
        batch_id=batch.id, source_path="/y", task_type="media",
        status="complete", finished_at=datetime.now(UTC),
    ))
    await db.commit()

    resp = await client.post("/api/import/clear-completed", json={"age_days": 7})
    body = await resp.get_json()
    assert body["deleted"] == 1


@pytest.mark.asyncio
async def test_clear_stuck_fails_non_terminal_and_finalizes_orphan_batch(client, db):
    """Operator-flagged 2026-05-25: 3 large PNGs got stuck in 'processing'
    for 2 days, the active ImportBatch never finalized, and the UI's
    'Scanning...' banner persisted with 0/0 files. /api/import/clear-stuck
    is the escape hatch to break the autoretry loop manually."""
    from sqlalchemy import select as _select

    batch = ImportBatch(triggered_by="manual", source_path="/import", scan_mode="quick")
    db.add(batch)
    await db.flush()
    # Three stuck rows in mixed non-terminal states.
    db.add(ImportTask(
        batch_id=batch.id, source_path="/p1", task_type="media", status="processing",
    ))
    db.add(ImportTask(
        batch_id=batch.id, source_path="/p2", task_type="media", status="queued",
    ))
    db.add(ImportTask(
        batch_id=batch.id, source_path="/p3", task_type="media", status="pending",
    ))
    # One already-complete row should be untouched.
    db.add(ImportTask(
        batch_id=batch.id, source_path="/done", task_type="media",
        status="complete", finished_at=datetime.now(UTC),
    ))
    await db.commit()

    resp = await client.post("/api/import/clear-stuck")
    body = await resp.get_json()
    assert resp.status_code == 200
    assert body["tasks_failed"] == 3
    assert body["batches_finalized"] == 1

    statuses = {
        row.status for row in
        (await db.execute(_select(ImportTask).where(ImportTask.batch_id == batch.id)))
        .scalars().all()
    }
    assert statuses == {"failed", "complete"}

    batch_status = (await db.execute(
        _select(ImportBatch.status).where(ImportBatch.id == batch.id)
    )).scalar_one()
    assert batch_status == "complete"


@pytest.mark.asyncio
async def test_clear_stuck_no_op_when_nothing_stuck(client, db):
    resp = await client.post("/api/import/clear-stuck")
    body = await resp.get_json()
    assert resp.status_code == 200
    assert body == {"tasks_failed": 0, "batches_finalized": 0}


@pytest.mark.asyncio
async def test_trigger_accepts_deep(client, monkeypatch):
    # Stub the task dispatch — assert the API accepts 'deep' and forwards
    # mode, without running the real (eager) scan_directory.
    from backend.app.tasks import scan as scan_mod

    seen = {}

    class _Res:
        id = "celery-deep"

    def _fake_delay(*, triggered_by, mode):
        seen["triggered_by"] = triggered_by
        seen["mode"] = mode
        return _Res()

    monkeypatch.setattr(scan_mod.scan_directory, "delay", _fake_delay)
    resp = await client.post("/api/import/trigger", json={"mode": "deep"})
    assert resp.status_code == 202
    body = await resp.get_json()
    assert body["mode"] == "deep"
    assert seen == {"triggered_by": "manual", "mode": "deep"}


@pytest.mark.asyncio
async def test_trigger_still_rejects_unknown_mode(client):
    resp = await client.post("/api/import/trigger", json={"mode": "wat"})
    assert resp.status_code == 400


@pytest.mark.asyncio
async def test_refetch_404_for_unknown_task(client):
    resp = await client.post("/api/import/tasks/999999/refetch")
    assert resp.status_code == 404


@pytest.mark.asyncio
async def test_refetch_400_for_non_failed_task(client, db):
    batch = ImportBatch(triggered_by="manual", source_path="/import", scan_mode="quick")
    db.add(batch)
    await db.flush()
    task = ImportTask(
        batch_id=batch.id, source_path="/x.jpg", task_type="media",
        status="complete", finished_at=datetime.now(UTC),
    )
    db.add(task)
    await db.commit()
    resp = await client.post(f"/api/import/tasks/{task.id}/refetch")
    assert resp.status_code == 400
    assert (await resp.get_json())["status"] == "not_failed"


@pytest.mark.asyncio
async def test_refetch_no_source_when_unresolvable(client, db):
    """A failed task whose file has no sidecar / no resolvable Source
    returns no_source (filesystem-only import — nothing to re-poll)."""
    batch = ImportBatch(triggered_by="manual", source_path="/import", scan_mode="quick")
    db.add(batch)
    await db.flush()
    task = ImportTask(
        batch_id=batch.id, source_path="/import/nowhere/x.jpg",
        task_type="media", status="failed", finished_at=datetime.now(UTC),
    )
    db.add(task)
    await db.commit()
    resp = await client.post(f"/api/import/tasks/{task.id}/refetch")
    assert resp.status_code == 200
    assert (await resp.get_json())["status"] == "no_source"


@pytest.mark.asyncio
async def test_refetch_queued_with_resolvable_source(client, db, tmp_path, monkeypatch):
    """A failed task whose file resolves (via sidecar → artist+platform)
    to an enabled, real-URL Source: the file is deleted, the task is
    marked refetched, and ONE source re-check is queued."""
    import json as _json

    from sqlalchemy import update as _update

    from backend.app.models import Artist, ImportSettings, Source
    from backend.app.tasks import download as download_mod

    # Stub the downloader so the eager test doesn't run a real fetch.
    dispatched = []
    monkeypatch.setattr(download_mod.download_source, "delay", dispatched.append)

    # import_root/<ArtistName>/post.jpg + sidecar identifying the platform.
    import_root = tmp_path / "import"
    artist_dir = import_root / "Maewix"
    artist_dir.mkdir(parents=True)
    media = artist_dir / "post.jpg"
    media.write_bytes(b"corrupt-bytes")
    (artist_dir / "post.jpg.json").write_text(
        _json.dumps({"category": "patreon", "post_id": 123})
    )

    # import_settings(id=1) is migration-seeded; point its scan path at
    # our tmp import root rather than inserting a conflicting row.
    await db.execute(
        _update(ImportSettings).where(ImportSettings.id == 1)
        .values(import_scan_path=str(import_root))
    )
    artist = Artist(name="Maewix", slug="maewix")
    db.add(artist)
    await db.flush()
    db.add(Source(
        artist_id=artist.id, platform="patreon",
        url="https://www.patreon.com/maewix", enabled=True,
        config_overrides={},
    ))
    batch = ImportBatch(triggered_by="manual", source_path=str(import_root), scan_mode="quick")
    db.add(batch)
    await db.flush()
    task = ImportTask(
        batch_id=batch.id, source_path=str(media), task_type="media",
        status="failed", finished_at=datetime.now(UTC),
    )
    db.add(task)
    await db.commit()

    resp = await client.post(f"/api/import/tasks/{task.id}/refetch")
    assert resp.status_code == 200
    body = await resp.get_json()
    assert body["status"] == "refetch_queued"
    assert len(dispatched) == 1
    assert not media.exists()  # corrupt copy removed for re-fetch

    from sqlalchemy import select as _select
    refetched = (await db.execute(
        _select(ImportTask.refetched).where(ImportTask.id == task.id)
    )).scalar_one()
    assert refetched is True

    # Second attempt is a no-op (bounded to one).
    resp2 = await client.post(f"/api/import/tasks/{task.id}/refetch")
    assert (await resp2.get_json())["status"] == "already_refetched"
    assert len(dispatched) == 1


@pytest.mark.asyncio
async def test_trigger_accepts_verify(client, monkeypatch):
    # Stub the verify task's dispatch so the API contract is asserted
    # without running the real (eager) verify_integrity end-to-end.
    from backend.app.tasks import maintenance as m

    seen = {}

    class _Res:
        id = "celery-verify"

    def _fake_delay():
        seen["called"] = True
        return _Res()

    monkeypatch.setattr(m.verify_integrity, "delay", _fake_delay)
    resp = await client.post("/api/import/trigger", json={"mode": "verify"})
    assert resp.status_code == 202
    body = await resp.get_json()
    assert body["mode"] == "verify"
    assert seen == {"called": True}