feat: fpw plugins, validation/crawl perf, WS stats, test DB isolation

- Add Free_Proxy_Website-style fpw_* plugins and register them - Per-plugin crawl timeout (crawl_timeout_seconds=120); remove global crawl_timeout setting - Validator: fix connect vs total timeout on save; SOCKS session LRU cache; drop redundant semaphore - Validation handler uses single DB connection; batch upsert after crawl; WorkerPool put_nowait - Remove unused max_retries from settings API/UI; settings maintenance SQL + init_db cleanup of deprecated keys - WebSocket dashboard stats; ProxyList pool_filter and API alignment - POST /api/proxies/delete-one for IPv6-safe deletes; task poll stops on 404 - pytest uses PROXYPOOL_DB_PATH=db/proxies.test.sqlite so tests do not wipe production DB - .gitignore: explicit proxies.test.sqlite patterns; fix plugin_service ValidationException import Made-with: Cursor
2026-04-05 13:39:19 +08:00
parent 92c7fa19e2
commit 0131c8b408
63 changed files with 2331 additions and 531 deletions
--- a/tests/README.md
+++ b/tests/README.md
@@ -5,6 +5,8 @@
 ```
 tests/
 ├── conftest.py              # pytest 配置和 fixtures
+├── task_utils.py            # 异步任务轮询（集成/E2E 共用）
+├── support/                 # 测试专用插件类等（非 mock）
 ├── README.md                # 本文件
 ├── unit/                    # 单元测试
 │   ├── test_models.py       # 模型测试
@@ -12,6 +14,7 @@ tests/
 ├── integration/             # 集成测试
 │   ├── test_proxies_api.py  # 代理 API 测试
 │   ├── test_plugins_api.py  # 插件 API 测试
+│   ├── test_plugins_live_crawl.py # 各插件真实爬取验收（须外网）
 │   ├── test_scheduler_api.py # 调度器 API 测试
 │   ├── test_settings_api.py # 设置 API 测试
 │   └── test_health_api.py   # 健康检查测试
@@ -19,6 +22,25 @@ tests/
    └── test_full_workflow.py # 完整工作流测试
 ```

+## 网络与真实调用
+
+集成测试与 E2E **不再 mock** `PluginRunner` / `ValidatorService`：会发起真实 HTTP 爬取与代理验证（视设置而定）。运行全量 `pytest` 需要 **可用的出站网络**，且含 `network` / `slow` 标记的用例可能耗时数分钟。
+
+跳过需外网的用例（例如离线快速检查）：
+
+```bash
+pytest -m "not network"
+```
+
+**插件爬取验收**（`test_plugins_live_crawl.py`）：
+
+- 核心 8 插件：必须至少 1 条代理且无 Runner 失败。
+- `fpw_*`：对照 [Free_Proxy_Website](https://github.com/cyubuchen/Free_Proxy_Website) 的公开源，允许 0 条（国际网络差异），使用更长超时。
+
+```bash
+pytest tests/integration/test_plugins_live_crawl.py -v
+```
+
 ## 运行测试

 ### 安装测试依赖
--- a/tests/conftest.py
+++ b/tests/conftest.py
@@ -1,5 +1,15 @@
 """pytest 配置文件和 fixtures"""
+# 必须在任何 app.* 导入之前：下方 app fixture 会清空表，不可与生产共用 db/proxies.sqlite
+import os
+
+os.environ["PROXYPOOL_DB_PATH"] = "db/proxies.test.sqlite"
+
 import asyncio
+import sys
+
+if sys.platform == "win32":
+    asyncio.set_event_loop_policy(asyncio.WindowsSelectorEventLoopPolicy())
+
 import pytest
 import pytest_asyncio
 from typing import AsyncGenerator
@@ -17,22 +27,28 @@ from app.plugins import (
    SpeedXPlugin,
    YunDaiLiPlugin,
    ProxyScrapePlugin,
+    FpwProxyListDownloadPlugin,
+    FpwSocksSslProxyPlugin,
+    FpwSpysOnePlugin,
+    FpwProxynovaPlugin,
+    FpwHidemyPlugin,
+    FpwPremproxyPlugin,
+    FpwFreeproxylistsPlugin,
+    FpwGatherproxyPlugin,
+    FpwCheckerproxyPlugin,
 )
 from app.repositories.proxy_repo import ProxyRepository
-from app.models.domain import ProxyRaw


@pytest_asyncio.fixture(scope="function")
 async def app():
    """创建应用实例"""
-    # 初始化测试数据库并清空历史数据
    await init_db()
    async with get_db() as db:
        await db.execute("DELETE FROM proxies")
        await db.execute("DELETE FROM settings")
        await db.commit()
-    
-    # 清理并重新注册插件，防止跨测试污染
+
    registry.clear()
    for plugin_cls in [
        Fate0Plugin,
@@ -43,14 +59,22 @@ async def app():
        SpeedXPlugin,
        YunDaiLiPlugin,
        ProxyScrapePlugin,
+        FpwProxyListDownloadPlugin,
+        FpwSocksSslProxyPlugin,
+        FpwSpysOnePlugin,
+        FpwProxynovaPlugin,
+        FpwHidemyPlugin,
+        FpwPremproxyPlugin,
+        FpwFreeproxylistsPlugin,
+        FpwGatherproxyPlugin,
+        FpwCheckerproxyPlugin,
    ]:
        registry.register(plugin_cls)
-    
+
    test_app = create_app()
    async with test_app.router.lifespan_context(test_app):
        yield test_app
-    
-    # 给 aiosqlite / aiohttp 后台线程留出收尾时间
+
    await asyncio.sleep(0.1)


@@ -80,32 +104,4 @@ async def sample_proxy(db, proxy_repo):
    """创建一个测试代理"""
    await proxy_repo.insert_or_update(db, "192.168.1.1", 8080, "http", 50)
    yield {"ip": "192.168.1.1", "port": 8080, "protocol": "http", "score": 50}
-    # 清理
    await proxy_repo.delete(db, "192.168.1.1", 8080)
-
-
-@pytest_asyncio.fixture(autouse=True)
-async def mock_external_requests(monkeypatch, request):
-    """
-    自动在集成/E2E 测试中 mock 外部网络请求：
-    1. 插件爬取返回固定测试代理，避免真实 HTTP 请求
-    2. 代理验证瞬间成功，避免连接超时等待
-    """
-    if "/unit/" in request.node.nodeid:
-        return
-    from app.services.plugin_runner import PluginRunner
-    from app.services.validator_service import ValidatorService
-
-    async def _mock_run(self, plugin):
-        from app.models.domain import CrawlResult
-        return CrawlResult(
-            plugin_name=plugin.name,
-            proxies=[ProxyRaw("192.168.100.10", 8080, "http")],
-            success_count=1,
-        )
-
-    async def _mock_validate(self, ip: str, port: int, protocol: str = "http"):
-        return True, 1.23
-
-    monkeypatch.setattr(PluginRunner, "run", _mock_run)
-    monkeypatch.setattr(ValidatorService, "validate", _mock_validate)
--- a/tests/e2e/test_full_workflow.py
+++ b/tests/e2e/test_full_workflow.py
@@ -4,10 +4,14 @@
 """
 import pytest

+from tests.task_utils import poll_task_until_terminal
+

 class TestFullWorkflow:
    """测试完整工作流"""

+    @pytest.mark.network
+    @pytest.mark.slow
    @pytest.mark.asyncio
    async def test_proxy_management_workflow(self, client):
        """测试代理管理完整工作流
@@ -35,12 +39,18 @@ class TestFullWorkflow:
        # 3. 触发所有插件爬取
        response = await client.post("/api/plugins/crawl-all")
        assert response.status_code == 200
-        crawl_result = response.json()["data"]
+        task_id = response.json()["data"]["task_id"]
+        task_data = await poll_task_until_terminal(
+            client, task_id, max_rounds=400, interval=0.5
+        )
+        assert task_data is not None
+        assert task_data["status"] in ("completed", "failed", "cancelled")
        
        # 4. 获取更新后的统计
        response = await client.get("/api/proxies/stats")
        updated_stats = response.json()["data"]
-        
+        assert "total" in initial_stats and "total" in updated_stats
+
        # 5. 导出代理（所有格式）
        for fmt in ["csv", "txt", "json"]:
            response = await client.get(f"/api/proxies/export/{fmt}")
@@ -50,6 +60,8 @@ class TestFullWorkflow:
        response = await client.delete("/api/proxies/clean-invalid")
        assert response.status_code == 200

+    @pytest.mark.network
+    @pytest.mark.slow
    @pytest.mark.asyncio
    async def test_plugin_management_workflow(self, client):
        """测试插件管理完整工作流
@@ -93,6 +105,12 @@ class TestFullWorkflow:
        # 6. 触发爬取
        response = await client.post(f"/api/plugins/{plugin_id}/crawl")
        assert response.status_code == 200
+        crawl_task_id = response.json()["data"]["task_id"]
+        crawl_task = await poll_task_until_terminal(
+            client, crawl_task_id, max_rounds=140, interval=0.5
+        )
+        assert crawl_task is not None
+        assert crawl_task["status"] in ("completed", "failed", "cancelled")

    @pytest.mark.asyncio
    async def test_scheduler_workflow(self, client):
--- a/tests/integration/test_plugins_api.py
+++ b/tests/integration/test_plugins_api.py
@@ -1,6 +1,8 @@
 """插件 API 集成测试 - 测试 /api/plugins/* 所有接口"""
 import pytest

+from tests.task_utils import poll_task_until_terminal
+

 class TestPluginsAPI:
    """测试插件相关 API"""
@@ -116,10 +118,11 @@ class TestPluginsAPI:
        data = response.json()
        assert data["code"] == 200

+    @pytest.mark.network
+    @pytest.mark.slow
    @pytest.mark.asyncio
    async def test_crawl_plugin(self, client):
        """测试 POST /api/plugins/{id}/crawl - 异步任务模式"""
-        import asyncio
        response = await client.get("/api/plugins")
        plugins = response.json()["data"]["plugins"]
        if not plugins:
@@ -133,18 +136,11 @@ class TestPluginsAPI:
        assert "task_id" in data["data"]
        
        task_id = data["data"]["task_id"]
-        # 轮询任务状态
-        task_data = None
-        for _ in range(10):
-            await asyncio.sleep(0.3)
-            res = await client.get(f"/api/tasks/{task_id}")
-            assert res.status_code == 200
-            task_data = res.json()["data"]
-            if task_data["status"] in ("completed", "failed", "cancelled"):
-                break
-        
+        task_data = await poll_task_until_terminal(
+            client, task_id, max_rounds=140, interval=0.5
+        )
        assert task_data is not None
-        assert task_data["status"] in ("completed", "cancelled")
+        assert task_data["status"] in ("completed", "failed", "cancelled")

    @pytest.mark.asyncio
    async def test_crawl_nonexistent_plugin(self, client):
@@ -152,10 +148,11 @@ class TestPluginsAPI:
        response = await client.post("/api/plugins/nonexistent_plugin/crawl")
        assert response.status_code == 404

+    @pytest.mark.network
+    @pytest.mark.slow
    @pytest.mark.asyncio
    async def test_crawl_all_plugins(self, client):
        """测试 POST /api/plugins/crawl-all - 异步任务模式"""
-        import asyncio
        response = await client.post("/api/plugins/crawl-all")
        assert response.status_code == 200
        data = response.json()
@@ -163,15 +160,8 @@ class TestPluginsAPI:
        assert "task_id" in data["data"]
        
        task_id = data["data"]["task_id"]
-        # 轮询任务状态
-        task_data = None
-        for _ in range(10):
-            await asyncio.sleep(0.3)
-            res = await client.get(f"/api/tasks/{task_id}")
-            assert res.status_code == 200
-            task_data = res.json()["data"]
-            if task_data["status"] in ("completed", "failed", "cancelled"):
-                break
-        
+        task_data = await poll_task_until_terminal(
+            client, task_id, max_rounds=400, interval=0.5
+        )
        assert task_data is not None
-        assert task_data["status"] in ("completed", "cancelled")
+        assert task_data["status"] in ("completed", "failed", "cancelled")
--- a/tests/integration/test_proxies_api.py
+++ b/tests/integration/test_proxies_api.py
@@ -14,6 +14,7 @@ class TestProxiesAPI:
        assert data["code"] == 200
        assert "data" in data
        assert "total" in data["data"]
+        assert "pending" in data["data"]
        assert "available" in data["data"]
        assert "scheduler_running" in data["data"]

@@ -68,6 +69,17 @@ class TestProxiesAPI:
        # 可能返回 200(有数据) 或 404(无数据)
        assert response.status_code in [200, 404]

+    @pytest.mark.asyncio
+    async def test_delete_proxy_post_json(self, client, sample_proxy):
+        """测试 POST /api/proxies/delete-one（前端默认路径，兼容 IPv6）"""
+        response = await client.post(
+            "/api/proxies/delete-one",
+            json={"ip": sample_proxy["ip"], "port": sample_proxy["port"]},
+        )
+        assert response.status_code == 200
+        data = response.json()
+        assert data["code"] == 200
+
    @pytest.mark.asyncio
    async def test_delete_proxy(self, client, sample_proxy):
        """测试 DELETE /api/proxies/{ip}/{port}"""
@@ -76,6 +88,19 @@ class TestProxiesAPI:
        data = response.json()
        assert data["code"] == 200

+    @pytest.mark.asyncio
+    async def test_delete_one_ipv6(self, client, db, proxy_repo):
+        """POST delete-one 可删除含冒号的 IP（路径 DELETE 无法可靠表达）"""
+        await proxy_repo.insert_or_update(db, "2001:db8::1", 18080, "http", 40)
+        r = await client.post(
+            "/api/proxies/delete-one",
+            json={"ip": "2001:db8::1", "port": 18080},
+        )
+        assert r.status_code == 200
+        assert r.json()["code"] == 200
+        left = await proxy_repo.get_by_ip_port(db, "2001:db8::1", 18080)
+        assert left is None
+
    @pytest.mark.asyncio
    async def test_delete_nonexistent_proxy(self, client):
        """测试 DELETE /api/proxies/{ip}/{port} - 不存在的代理"""
--- a/tests/integration/test_scheduler_api.py
+++ b/tests/integration/test_scheduler_api.py
@@ -1,6 +1,17 @@
 """调度器 API 集成测试 - 测试 /api/scheduler/* 所有接口"""
 import pytest

+from app.api.deps import get_settings_repo
+from app.repositories.settings_repo import SettingsRepository
+
+
+class FailingSettingsRepository(SettingsRepository):
+    """save 恒为 False，用于覆盖「设置保存失败」分支，非 MagicMock。"""
+
+    @staticmethod
+    async def save(db, settings):
+        return False
+

 class TestSchedulerAPI:
    """测试调度器相关 API"""
@@ -93,18 +104,17 @@ class TestSchedulerAPI:
        assert job is not None

    @pytest.mark.asyncio
-    async def test_start_scheduler_db_save_failure(self, client, monkeypatch):
+    async def test_start_scheduler_db_save_failure(self, client, app):
        """测试启动调度器时数据库保存失败应返回 running=False"""
-        from app.repositories.settings_repo import SettingsRepository
-
        # lifespan 启动时调度器可能已自动启动，先停止它
        await client.post("/api/scheduler/stop")

-        async def mock_save(*args, **kwargs):
-            return False
+        app.dependency_overrides[get_settings_repo] = lambda: FailingSettingsRepository()
+        try:
+            response = await client.post("/api/scheduler/start")
+        finally:
+            app.dependency_overrides.pop(get_settings_repo, None)

-        monkeypatch.setattr(SettingsRepository, "save", mock_save)
-        response = await client.post("/api/scheduler/start")
        assert response.status_code == 200
        data = response.json()
        assert data["code"] == 200
--- a/tests/integration/test_settings_api.py
+++ b/tests/integration/test_settings_api.py
@@ -12,7 +12,7 @@ class TestSettingsAPI:
        assert response.status_code == 200
        data = response.json()
        assert data["code"] == 200
-        assert "crawl_timeout" in data["data"]
+        assert "crawl_timeout" not in data["data"]
        assert "validation_timeout" in data["data"]
        assert "auto_validate" in data["data"]

@@ -21,17 +21,16 @@ class TestSettingsAPI:
        """测试 GET /api/settings 返回结构"""
        response = await client.get("/api/settings")
        data = response.json()["data"]
-        
-        # 验证所有预期的设置项
+
        expected_keys = [
-            "crawl_timeout",
            "validation_timeout",
-            "max_retries",
            "default_concurrency",
            "min_proxy_score",
            "proxy_expiry_days",
            "auto_validate",
+            "auto_validate_after_crawl",
            "validate_interval_minutes",
+            "validation_targets",
        ]
        for key in expected_keys:
            assert key in data, f"缺少设置项: {key}"
@@ -40,65 +39,45 @@ class TestSettingsAPI:
    async def test_save_settings(self, client):
        """测试 POST /api/settings"""
        settings = {
-            "crawl_timeout": 45,
            "validation_timeout": 15,
-            "max_retries": 5,
            "default_concurrency": 100,
            "min_proxy_score": 10,
            "proxy_expiry_days": 14,
            "auto_validate": True,
+            "auto_validate_after_crawl": False,
            "validate_interval_minutes": 60,
+            "validation_targets": [
+                "http://httpbin.org/ip",
+            ],
        }
        response = await client.post("/api/settings", json=settings)
        assert response.status_code == 200
        data = response.json()
        assert data["code"] == 200
-        # 验证返回的数据与提交的一致
        for key, value in settings.items():
            assert data["data"][key] == value

    @pytest.mark.asyncio
    async def test_save_settings_partial(self, client):
        """测试 POST /api/settings - 部分更新（实际上会替换所有）"""
-        # 先获取当前设置
        response = await client.get("/api/settings")
        current_settings = response.json()["data"]
-        
-        # 修改部分设置
+
        new_settings = current_settings.copy()
-        new_settings["crawl_timeout"] = 60
+        new_settings["validation_timeout"] = 25
        new_settings["auto_validate"] = False
-        
+
        response = await client.post("/api/settings", json=new_settings)
        assert response.status_code == 200
        data = response.json()
-        assert data["data"]["crawl_timeout"] == 60
+        assert data["data"]["validation_timeout"] == 25
        assert data["data"]["auto_validate"] is False

    @pytest.mark.asyncio
    async def test_save_settings_validation_error(self, client):
        """测试 POST /api/settings - 验证错误"""
-        # crawl_timeout 必须在 5-120 之间
        invalid_settings = {
-            "crawl_timeout": 200,  # 超出范围
-            "validation_timeout": 10,
-            "max_retries": 3,
-            "default_concurrency": 50,
-            "min_proxy_score": 0,
-            "proxy_expiry_days": 7,
-            "auto_validate": True,
-            "validate_interval_minutes": 30,
-        }
-        response = await client.post("/api/settings", json=invalid_settings)
-        assert response.status_code == 422  # 验证错误
-
-    @pytest.mark.asyncio
-    async def test_save_settings_invalid_type(self, client):
-        """测试 POST /api/settings - 无效类型"""
-        invalid_settings = {
-            "crawl_timeout": "invalid",  # 应该是整数
-            "validation_timeout": 10,
-            "max_retries": 3,
+            "validation_timeout": 100,
            "default_concurrency": 50,
            "min_proxy_score": 0,
            "proxy_expiry_days": 7,
@@ -108,31 +87,62 @@ class TestSettingsAPI:
        response = await client.post("/api/settings", json=invalid_settings)
        assert response.status_code == 422

+    @pytest.mark.asyncio
+    async def test_save_settings_invalid_type(self, client):
+        """测试 POST /api/settings - 无效类型"""
+        invalid_settings = {
+            "validation_timeout": 10,
+            "default_concurrency": "invalid",
+            "min_proxy_score": 0,
+            "proxy_expiry_days": 7,
+            "auto_validate": True,
+            "validate_interval_minutes": 30,
+        }
+        response = await client.post("/api/settings", json=invalid_settings)
+        assert response.status_code == 422
+
+    @pytest.mark.asyncio
+    async def test_save_settings_ignores_deprecated_crawl_timeout(self, client):
+        """旧客户端若仍提交 crawl_timeout，应忽略且保存成功"""
+        response = await client.get("/api/settings")
+        base = response.json()["data"]
+        payload = {**base, "crawl_timeout": 999}
+        response = await client.post("/api/settings", json=payload)
+        assert response.status_code == 200
+        again = (await client.get("/api/settings")).json()["data"]
+        assert "crawl_timeout" not in again
+
+    @pytest.mark.asyncio
+    async def test_save_settings_ignores_obsolete_max_retries(self, client):
+        """已移除的 max_retries 键若仍被提交，应忽略。"""
+        response = await client.get("/api/settings")
+        base = response.json()["data"]
+        payload = {**base, "max_retries": 9}
+        response = await client.post("/api/settings", json=payload)
+        assert response.status_code == 200
+        again = (await client.get("/api/settings")).json()["data"]
+        assert "max_retries" not in again
+
    @pytest.mark.asyncio
    async def test_settings_roundtrip(self, client):
        """测试设置读写一致性"""
-        # 生成随机但有效的设置
        import random
+
        test_settings = {
-            "crawl_timeout": random.randint(10, 60),
            "validation_timeout": random.randint(5, 30),
-            "max_retries": random.randint(1, 5),
            "default_concurrency": random.randint(20, 100),
            "min_proxy_score": random.randint(0, 50),
            "proxy_expiry_days": random.randint(1, 14),
            "auto_validate": random.choice([True, False]),
            "validate_interval_minutes": random.randint(10, 120),
        }
-        
-        # 写入设置
+
        response = await client.post("/api/settings", json=test_settings)
        assert response.status_code == 200
-        
-        # 读取设置
+
        response = await client.get("/api/settings")
        saved_settings = response.json()["data"]
-        
-        # 验证一致性
+
        for key, value in test_settings.items():
            assert saved_settings[key] == value, f"设置项 {key} 不一致"

@@ -140,9 +150,7 @@ class TestSettingsAPI:
    async def test_settings_roundtrip_with_validation_targets(self, client):
        """测试设置读写一致性 - 包含数组类型的 validation_targets"""
        test_settings = {
-            "crawl_timeout": 30,
            "validation_timeout": 10,
-            "max_retries": 3,
            "default_concurrency": 50,
            "min_proxy_score": 0,
            "proxy_expiry_days": 7,
@@ -154,13 +162,11 @@ class TestSettingsAPI:
            ],
        }

-        # 写入设置
        response = await client.post("/api/settings", json=test_settings)
        assert response.status_code == 200
        data = response.json()
        assert data["data"]["validation_targets"] == test_settings["validation_targets"]

-        # 读取设置
        response = await client.get("/api/settings")
        saved_settings = response.json()["data"]
        assert saved_settings["validation_targets"] == test_settings["validation_targets"]
@@ -179,7 +185,6 @@ class TestSettingsAPI:
        data = response.json()
        assert data["data"]["validation_targets"] == []

-        # 读取确认
        response = await client.get("/api/settings")
        saved_settings = response.json()["data"]
        assert saved_settings["validation_targets"] == []
--- a/tests/support/init.py
+++ b/tests/support/init.py
@@ -0,0 +1 @@
+# Test support package (non-mock plugin doubles, etc.)
--- a/tests/support/plugins_for_runner.py
+++ b/tests/support/plugins_for_runner.py
@@ -0,0 +1,19 @@
+"""供 PluginRunner 等测试使用的真实插件子类（非 unittest.mock）。"""
+from typing import List
+
+from app.core.plugin_system.base import BaseCrawlerPlugin
+from app.models.domain import ProxyRaw
+
+
+class UnhealthyPlugin(BaseCrawlerPlugin):
+    """health_check 抛错，用于验证 Runner 对异常的统计与落库。"""
+
+    name = "test_unhealthy_runner"
+    display_name = "TestUnhealthy"
+    description = "PluginRunner health_check failure test double"
+
+    async def crawl(self) -> List[ProxyRaw]:
+        return []
+
+    async def health_check(self) -> bool:
+        raise RuntimeError("network down")
--- a/tests/task_utils.py
+++ b/tests/task_utils.py
@@ -0,0 +1,22 @@
+"""测试用异步任务轮询工具"""
+import asyncio
+from typing import Any, Dict, Optional
+
+
+async def poll_task_until_terminal(
+    client,
+    task_id: str,
+    *,
+    max_rounds: int,
+    interval: float,
+) -> Optional[Dict[str, Any]]:
+    """轮询任务直到终态或超时。返回最后一次 task data。"""
+    task_data = None
+    for _ in range(max_rounds):
+        await asyncio.sleep(interval)
+        res = await client.get(f"/api/tasks/{task_id}")
+        assert res.status_code == 200
+        task_data = res.json()["data"]
+        if task_data["status"] in ("completed", "failed", "cancelled"):
+            break
+    return task_data
--- a/tests/unit/test_models.py
+++ b/tests/unit/test_models.py
@@ -106,6 +106,14 @@ class TestProxyListRequest:
        assert request.page_size == 50
        assert request.protocol == "https"

+    def test_pool_filter_pending_available(self):
+        r1 = ProxyListRequest(pool_filter="pending")
+        assert r1.pool_filter == "pending"
+        r2 = ProxyListRequest(pool_filter="all")
+        assert r2.pool_filter is None
+        with pytest.raises(Exception):
+            ProxyListRequest(pool_filter="invalid")
+

 class TestSettingsSchema:
    """测试 SettingsSchema"""
@@ -113,16 +121,22 @@ class TestSettingsSchema:
    def test_default_settings(self):
        """测试默认设置"""
        settings = SettingsSchema()
-        assert settings.crawl_timeout == 30
-        assert settings.validation_timeout == 10
+        assert settings.validation_timeout == 6
+        assert settings.default_concurrency == 120
        assert settings.auto_validate is True
+        assert settings.auto_validate_after_crawl is False

    def test_custom_settings(self):
        """测试自定义设置"""
-        settings = SettingsSchema(crawl_timeout=60, auto_validate=False)
-        assert settings.crawl_timeout == 60
+        settings = SettingsSchema(validation_timeout=25, auto_validate=False)
+        assert settings.validation_timeout == 25
        assert settings.auto_validate is False

+    def test_settings_schema_ignores_unknown_fields(self):
+        s = SettingsSchema.model_validate({"validation_timeout": 10, "crawl_timeout": 99})
+        assert "crawl_timeout" not in s.model_dump()
+        assert s.validation_timeout == 10
+

 class TestBatchDeleteRequest:
    """测试 BatchDeleteRequest"""
--- a/tests/unit/test_repositories.py
+++ b/tests/unit/test_repositories.py
@@ -81,23 +81,25 @@ class TestProxyRepository:

    @pytest.mark.asyncio
    async def test_iter_batches(self, db, proxy_repo):
-        """测试流式分批读取"""
-        # 插入 5 条测试数据
+        """测试流式分批读取（与库内已有数据共存，只校验增量与分批形状）"""
+        async with db.execute("SELECT COUNT(*) FROM proxies") as c:
+            before = (await c.fetchone())[0]
        for i in range(5):
-            await proxy_repo.insert_or_update(db, f"192.168.1.{i}", 8000 + i, "http", 10)
-        
+            await proxy_repo.insert_or_update(db, f"192.168.99.{i}", 8000 + i, "http", 10)
+        async with db.execute("SELECT COUNT(*) FROM proxies") as c:
+            after = (await c.fetchone())[0]
+        assert after == before + 5
+
        batches = []
        async for batch in proxy_repo.iter_batches(db, batch_size=2):
            batches.append(batch)
-        
-        assert len(batches) == 3
-        assert len(batches[0]) == 2
-        assert len(batches[1]) == 2
-        assert len(batches[2]) == 1
-        
-        # 清理
+
+        assert sum(len(b) for b in batches) == after
+        assert len(batches[-1]) in (1, 2)
+        assert all(len(b) <= 2 for b in batches)
+
        for i in range(5):
-            await proxy_repo.delete(db, f"192.168.1.{i}", 8000 + i)
+            await proxy_repo.delete(db, f"192.168.99.{i}", 8000 + i)

    @pytest.mark.asyncio
    async def test_batch_delete(self, db, proxy_repo):
@@ -121,6 +123,38 @@ class TestProxyRepository:
        """测试获取统计信息"""
        stats = await proxy_repo.get_stats(db)
        assert "total" in stats
+        assert "pending" in stats
        assert "available" in stats
        assert "avg_score" in stats
        assert "http_count" in stats
+
+    @pytest.mark.asyncio
+    async def test_get_today_new_count_only_validated_available(self, db, proxy_repo):
+        """今日新增不计待验证；仅今日创建且 validated=1、score>0"""
+        base = await proxy_repo.get_today_new_count(db)
+        await proxy_repo.upsert_from_crawl(db, "192.168.88.20", 9020, "http", 0)
+        assert await proxy_repo.get_today_new_count(db) == base
+
+        await proxy_repo.insert_or_update(db, "192.168.88.21", 9021, "http", 55)
+        assert await proxy_repo.get_today_new_count(db) == base + 1
+
+        await proxy_repo.delete(db, "192.168.88.20", 9020)
+        await proxy_repo.delete(db, "192.168.88.21", 9021)
+
+    @pytest.mark.asyncio
+    async def test_upsert_many_from_crawl(self, db, proxy_repo):
+        from app.models.domain import ProxyRaw
+
+        raws = [
+            ProxyRaw("10.0.0.1", 18080, "http"),
+            ProxyRaw("10.0.0.2", 18081, "socks5"),
+        ]
+        await proxy_repo.upsert_many_from_crawl(db, raws, 0)
+        await db.commit()
+        p1 = await proxy_repo.get_by_ip_port(db, "10.0.0.1", 18080)
+        assert p1 is not None
+        assert p1.validated == 0
+        p2 = await proxy_repo.get_by_ip_port(db, "10.0.0.2", 18081)
+        assert p2.protocol == "socks5"
+        await proxy_repo.delete(db, "10.0.0.1", 18080)
+        await proxy_repo.delete(db, "10.0.0.2", 18081)
				`@@ -0,0 +1 @@`
				`# Test support package (non-mock plugin doubles, etc.)`