DocHub/backend/tests/test_main.py

from unittest.mock import MagicMock

import pytest
from fastapi.testclient import TestClient

import app.main as main_module
from app.main import app
from app.document_service import DocumentNotFoundError
from app.dependencies import get_document_service, get_search_client

client = TestClient(app)


def test_health_returns_ok():
    response = client.get("/health")
    assert response.status_code == 200
    assert response.json() == {"status": "ok"}


def test_startup_creates_bucket_when_missing(monkeypatch):
    fake_garage = MagicMock()
    fake_garage.list_buckets.return_value = {"Buckets": []}
    monkeypatch.setattr(main_module, "_get_garage_sdk", lambda: fake_garage)

    with TestClient(app):
        pass

    fake_garage.list_buckets.assert_called()
    fake_garage.create_bucket.assert_called()


def test_startup_does_not_create_bucket_when_it_already_exists(monkeypatch):
    fake_garage = MagicMock()
    fake_garage.list_buckets.return_value = {"Buckets": [{"Name": "dochub"}]}
    monkeypatch.setattr(main_module, "_get_garage_sdk", lambda: fake_garage)

    with TestClient(app):
        pass

    fake_garage.create_bucket.assert_not_called()


def test_startup_does_not_crash_app_when_garage_is_unreachable(monkeypatch):
    def raise_connection_error():
        raise ConnectionError("garage unreachable")

    monkeypatch.setattr(main_module, "_get_garage_sdk", raise_connection_error)

    with TestClient(app) as test_client:
        response = test_client.get("/health")
        assert response.status_code == 200


class FakeDocumentService:
    def __init__(self):
        self.storage = {}

    def list_documents(self, path):
        prefix = path if not path or path.endswith("/") else f"{path}/"
        files = [
            {"key": key, "name": key[len(prefix):], "size": len(data), "modifiedAt": "2026-08-01T10:00:00Z"}
            for key, data in self.storage.items()
            if key.startswith(prefix)
        ]
        return {"folders": [], "files": files}

    def get_document(self, key):
        if key not in self.storage:
            raise DocumentNotFoundError(key)
        return {"key": key, "content": self.storage[key].decode("utf-8"), "contentType": "text/markdown"}

    def save_document(self, key, content, content_type):
        self.storage[key] = content
        return {"key": key, "updatedAt": "2026-08-01T10:05:00Z"}

    def delete_document(self, key):
        self.storage.pop(key, None)

    def upload_document(self, key, data, content_type):
        self.storage[key] = data
        return {"key": key, "name": key.rsplit("/", 1)[-1], "size": len(data)}


class FakeSearchClient:
    def search(self, query):
        return [
            {
                "key": "产品文档/架构设计.md",
                "title": "架构设计.md",
                "path": "产品文档",
                "snippet": "全文<em>检索</em>示例",
            }
        ]


@pytest.fixture
def e2e_client():
    fake_service = FakeDocumentService()
    fake_search = FakeSearchClient()
    app.dependency_overrides[get_document_service] = lambda: fake_service
    app.dependency_overrides[get_search_client] = lambda: fake_search
    with TestClient(app) as test_client:
        yield test_client
    app.dependency_overrides.clear()


def test_end_to_end_upload_get_list_search_delete_flow(e2e_client):
    upload_response = e2e_client.post(
        "/api/upload",
        data={"path": "产品文档"},
        files={"file": ("架构设计.md", "# 架构设计\n\nDocHub 支持全文检索。".encode("utf-8"), "text/markdown")},
    )
    assert upload_response.status_code == 201
    key = upload_response.json()["key"]
    assert key == "产品文档/架构设计.md"

    get_response = e2e_client.get(f"/api/documents/{key}")
    assert get_response.status_code == 200
    assert "架构设计" in get_response.json()["content"]

    list_response = e2e_client.get("/api/documents", params={"path": "产品文档"})
    assert list_response.status_code == 200
    assert any(f["key"] == key for f in list_response.json()["files"])

    search_response = e2e_client.get("/api/search", params={"q": "检索"})
    assert search_response.status_code == 200
    assert search_response.json()["results"][0]["key"] == "产品文档/架构设计.md"

    delete_response = e2e_client.delete(f"/api/documents/{key}")
    assert delete_response.status_code == 204

    missing_response = e2e_client.get(f"/api/documents/{key}")
    assert missing_response.status_code == 404