DF Test Framework

企业级 Python 测试自动化框架，支持 HTTP/UI/数据库/消息队列多场景测试

License

核心特性

多协议客户端 — HTTP/GraphQL/gRPC，同步/异步双模式，中间件拦截器链（重试、签名、Bearer Token）
异步高性能 — AsyncHttpClient 并发性能提升 10-30 倍，AsyncDatabase/AsyncRedis/异步 MQ 完整双轨支持
UI 自动化 — Playwright，同步/异步，浏览器会话复用，失败自动截图 + trace 取证
数据库访问 — SQLAlchemy 2.0，多方言（MySQL/PostgreSQL/SQLite/MSSQL），Repository + Unit of Work 模式
测试数据管线 — CaseContext[T] + YAML 驱动参数化，支持 skip/xfail(strict)/seed/tags，CLI 精准过滤复现
响应契约 — http.response_contract 配置驱动业务错误检查，无需自定义基类；expect_business_error 断言
命名与清理 — NamingStandard 命名协议 + df-test cleanup 兜底残留清理（SQL/Redis 双载体）
基准数据池 — BaselinePool + df-test baseline check|ensure，共享环境受管数据幂等保障与巡检
闸门与竞态 — requires_env/requires_flag 统一 skip、wait_until/assert_consistently 轮询原语、race() 竞态栅栏
多客户端模型 — @service_client / @ui_client / http_service() 零子类声明，多服务/多前端各自独立
CLI 脚手架 — df-test init 一键生成 api/ui/full 最小骨架，df-test doctor 项目体检
可观测性 — 语义事件驱动：OpenTelemetry 链路追踪 + Prometheus 监控 + Allure 报告自动集成
消息队列 — Kafka/RabbitMQ/RocketMQ 统一接口
存储客户端 — 本地文件 / AWS S3 / 阿里云 OSS
脱敏服务 — HTTP/UI/DB 日志中敏感字段自动遮盖，集中配置
Mock 工具 — HTTP/数据库/Redis/时间 Mock + Playwright 路由 Mock（page_mock 轮换/延迟应答），开箱即用

安装

# 使用 uv（推荐）
uv add df-test-framework

# 使用 pip
pip install df-test-framework

可选依赖（按需安装，核心功能无需额外依赖）：

依赖组	功能	安装命令
`ui`	Playwright UI 测试	`uv add "df-test-framework[ui]"`
`database-async`	异步数据库驱动（MySQL/PG/SQLite）	`uv add "df-test-framework[database-async]"`
`database-mssql`	MSSQL 驱动	`uv add "df-test-framework[database-mssql]"`
`mq`	消息队列（Kafka + RabbitMQ + RocketMQ）	`uv add "df-test-framework[mq]"`
`observability`	OpenTelemetry + Prometheus	`uv add "df-test-framework[observability]"`
`storage`	对象存储（S3 + 阿里云 OSS）	`uv add "df-test-framework[storage]"`
`crypto`	AES 加解密	`uv add "df-test-framework[crypto]"`
`gm`	SM2 签名/验签与加解密、SM3/HMAC-SM3、SM4	`uv add "df-test-framework[gm]"`
`all`	所有可选功能	`uv add "df-test-framework[all]"`

快速开始

脚手架建项目（推荐）

df-test init my-project              # API 测试项目
df-test init my-project --type ui    # UI 测试项目
df-test init my-project --type full  # API + UI 混合

cd my-project && pytest -v

API 测试

from df_test_framework import api_class, BaseAPI
from pydantic import BaseModel

class User(BaseModel):
    id: int
    name: str

@api_class()
class UserAPI(BaseAPI):
    def get_user(self, user_id: int) -> User:
        # 便捷方法返回解析后的数据：传 model= 得 Pydantic 模型，不传得 dict；
        # HTTP 4xx/5xx 自动抛 HttpError，配置 http.response_contract 后业务失败自动抛 BusinessError
        return self.get(f"/users/{user_id}", model=User)

def test_get_user(user_api: UserAPI):   # fixture 由 @api_class 自动注册
    user = user_api.get_user(1)
    assert user.id == 1

异步并发测试

import asyncio
from df_test_framework import AsyncHttpClient

async def test_concurrent():
    async with AsyncHttpClient("https://api.example.com") as client:
        tasks = [client.get(f"/users/{i}") for i in range(100)]
        responses = await asyncio.gather(*tasks)
        assert all(r.status_code == 200 for r in responses)

YAML 驱动参数化

# tests/data/login.yaml
tests:
  valid_user:
    title: "正常登录"
    marks: [smoke]
    data:
      username: admin
      password: secret
  locked_user:
    title: "锁定账号登录失败"
    xfail: "已知 Bug #123"
    data:
      username: locked
      password: secret

import pytest
from df_test_framework.testing.data.pipeline import case_dataset, CaseContext

@pytest.mark.parametrize("ctx", case_dataset("tests/data/login.yaml"))
def test_login(ctx: CaseContext, login_api):
    result = login_api.login(ctx.data["username"], ctx.data["password"])
    assert result["token"]

架构

Layer 4 ─── bootstrap/          # 引导层：Bootstrap、Providers、RuntimeContext
Layer 3 ─── testing/ + cli/     # 门面层：pytest Fixtures、CLI 脚手架
Layer 2 ─── capabilities/       # 能力层：HTTP/UI/DB/MQ/Storage
Layer 1 ─── infrastructure/     # 基础设施：config/logging/events/plugins
Layer 0 ─── core/               # 核心层：纯抽象（无第三方依赖）
横切  ───── plugins/             # 插件：MonitoringPlugin（Allure 走 Fixture 订阅事件）

文档

通过 df-test --help 查看命令入口，初始化项目后参考生成的 README。完整使用指南、国密说明及版本发布文档由维护团队提供，暂不公开源码仓库地址。

许可证

MIT License

Metadata

Release files for df-test-framework 4.10.2

For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.

Source distribution (sdist)

Source distribution for df-test-framework 4.10.2
File	Size	Uploaded
df_test_framework-4.10.2.tar.gz	2.8 MB	Oct 10, 2026	Details

Built distribution (wheel)

Table of built distributions (wheels) for df-test-framework 4.10.2
File	Interpreter	ABI	Platform	Reset
df_test_framework-4.10.2-py3-none-any.whl 812.3 kB Oct 10, 2026	Python 3	none	any	Details

Total release size: 3.6 MB

Release files / df_test_framework-4.10.2.tar.gz

Download URL	df_test_framework-4.10.2.tar.gz
Size	2.8 MB
Tags	Source
SHA-256 checksum How to use checksums	`c2d35dd403c650d94b18fa603ddd254744eb5b09f4b401215e90496227b94566`
BLAKE2b-256 checksum How to use checksums	`e899120f672d01ff7ae37166f10e03447f9f6fb7cb360eadc917ff407755596e`
Upload date	Oct 10, 2026
Uploaded using Trusted Publishing? What is trusted publishing?	No
Uploaded via	`twine/7.0.0 CPython/3.13.14`

Release files / df_test_framework-4.10.2-py3-none-any.whl

Download URL	df_test_framework-4.10.2-py3-none-any.whl
Size	812.3 kB
Tags	Python 3
SHA-256 checksum How to use checksums	`aee97621e19a07971d6928dc4dcb1e7fad9084710b93056e8e501373f20177c9`
BLAKE2b-256 checksum How to use checksums	`cd2cf26b5938828a474ea71d2b34eaef7980afd1c69fa212834bb7c3650e9aaf`
Upload date	Oct 10, 2026
Uploaded using Trusted Publishing? What is trusted publishing?	No
Uploaded via	`twine/7.0.0 CPython/3.13.14`