写在前面
根据pypistats统计,tep在pypi的下载量达到了1w,对于纯个人研发的一款测试小工具来说,已经算不错了,要知道HttpRunner也才6w啊。tep可以说是我在接口自动化测试这个领域的技术沉淀,凝结了个人经验和所见所闻的精华之作,它基于Pytest,借鉴了JMeter、RobotFramework、HttpRunner、京东接口测试平台等各种优秀自动化设计思想,小小工具,蕴含大大能量。相信它也已经影响了不少人,让初学者知道Pytest该怎么玩,让入门者知道Pytest工程化是什么样子,让熟练者可以参考对照优化代码。然而当我把tep优化到1.0.0正式版以后,为什么却选择停止维护呢? 一、 小工具的表达力不够。当我试图用tep来描绘更多自动化设计思想时,瞬间感觉到了一丝苍白,我不一定讲的清楚,别人也不一定能够理解,用代码来交流始终存在着一定门槛。二、每个人对Pytest使用方式不同 。Pytest本身是测试框架,很多人用它来做二次开发,设计”测试框架“,有好的,有差的,不管白猫黑猫能逮到耗子就是好猫,不管设计的如何,能实现接口自动化项目落地就是好框架。tep要想在这个方向上,建立一套标准,几乎是不可能的。这不并意味我会就此放弃Pytest,相反,我将致力于Pytest平台化,从做小工具改为做测试平台。 测试平台具有非常直观的强大表现力,并且具有工程化的规范性,一看就懂,一用就会,一点就通。测试平台也是能更好的做技术沉淀的,如果说写小工具是玩玩而已,那么开发测试平台就是认真搞技术了。比如,如何提高Pytest并行执行的效率,我相信测试平台会比小工具,更能给出一个比较完整的解决方案。下次使用Pytest,也许就不是从tep startproject
开始了,而是docker run
。
作为归档,我也把tep1.0.0正式版的完整教程放在这篇文章了,欢迎大家阅读。
正式版教程: https://dongfanger.gitee.io/blog/tep.html
正式版源码: https://github.com/dongfanger/tep
tep简介
tep
是Try Easy Pytest的首字母缩写,是一款基于pytest测试框架的测试工具,集成了各种实用的第三方包和优秀的自动化测试设计思想,帮你快速实现自动化项目落地。
快速入门
安装
pip install tep
Mac用户建议创建虚拟环境并激活:
python3 -m venv venv
source venv/bin/activate
验证安装成功:
tep -V
新建项目
tep startproject demo
带上-venv
参数,可创建单个项目的Python虚拟环境,并在该项目的虚拟环境中安装tep:
tep startproject demo -venv
启动FastAPI示例应用
运行utils/fastapi_mock.py
脚本。
执行示例用例
执行examples/tests/test_login_to_pay_flow.py
用例。
查看日志
============================= test session starts ==============================
collecting ... collected 1 item
test_login_to_pay_flow.py::test
============================== 1 passed in 0.14s ===============================
Process finished with exit code 0
2022-12-27 15:19:37.223 | INFO | fixtures.fixture_login:produce_expensive_data:15 - ----------------开始登录----------------
2022-12-27 15:19:37.234 | INFO | utils.http_client:request:37 -
Request URL: http://127.0.0.1:5000/login
Request Method: post
Request Headers: {"Content-Type": "application/json"}
Request Payload: {"json": {"username": "dongfanger", "password": "123456"}}
Status Code: 200
Response: {"token":"de2e3ffu29"}
Elapsed: 0.003s
2022-12-27 15:19:37.234 | INFO | fixtures.fixture_login:produce_expensive_data:23 - ----------------登录成功----------------
PASSED [100%]2022-12-27 15:19:37.235 | INFO | utils.step:__init__:12 - ----------------搜索商品----------------
2022-12-27 15:19:37.250 | INFO | utils.http_client:request:37 -
Request URL: http://127.0.0.1:5000/searchSku
Request Method: get
Request Headers: {"token": "de2e3ffu29"}
Request Payload: {"params": {"skuName": "\u7535\u5b50\u4e66"}}
Status Code: 200
Response: {"skuId":"222","price":"2.3"}
Elapsed: 0.001s
2022-12-27 15:19:37.250 | INFO | utils.step:__init__:12 - ----------------添加购物车----------------
2022-12-27 15:19:37.254 | INFO | utils.http_client:request:37 -
Request URL: http://127.0.0.1:5000/addCart
Request Method: post
Request Headers: {"token": "de2e3ffu29"}
Request Payload: {"json": {"skuId": "222", "skuNum": 2}}
Status Code: 200
Response: {"skuId":"222","price":"2.3","skuNum":"3","totalPrice":"6.9"}
Elapsed: 0.001s
2022-12-27 15:19:37.254 | INFO | utils.step:__init__:12 - ----------------下单----------------
2022-12-27 15:19:37.257 | INFO | utils.http_client:request:37 -
Request URL: http://127.0.0.1:5000/order
Request Method: post
Request Headers: {"token": "de2e3ffu29"}
Request Payload: {"json": {"orderId": 222, "payAmount": "0.2", "skuId": "222", "price": "2.3", "skuNum": "3", "totalPrice": "6.9"}}
Status Code: 200
Response: {"orderId":"333"}
Elapsed: 0.001s
2022-12-27 15:19:37.257 | INFO | utils.step:__init__:12 - ----------------支付----------------
2022-12-27 15:19:37.259 | INFO | utils.http_client:request:37 -
Request URL: http://127.0.0.1:5000/pay
Request Method: post
Request Headers: {"token": "de2e3ffu29"}
Request Payload: {"json": {"skuId": 123, "price": 0.1, "skuNum": 2, "totalPrice": 0.2, "orderId": "333"}}
Status Code: 200
Response: {"success":"true"}
Elapsed: 0.001s
能在本地跑起来看到日志且没有报错,恭喜您,上手成功!
目录结构说明
examples:示例代码,可无顾虑全删;
fixtures:Pytest fixture,自动导入;
resources:环境变量、全局变量;
tests:测试用例;
utils:工具包;
conftest.py:Pytest挂载;
pytest.ini:Pytest配置;
reports:测试报告,默认不显示,生成报告后会出现;
用例组织形式
推荐MVC分层设计和数据代码分离。
小提示:tep老版本的极速写法,即接口、数据、代码都放在一个文件的一个函数的写法,仍然适合于新手或追求效率时使用。
用例集
在tests目录下将测试用例按功能模块分成多个用例集:
tests
user
teacher
student
测试用例
必须遵循用例解耦原则,每条用例都是单独可运行的。用例由2个文件组成,一个文件存放纯粹的yaml数据,一个文件存放逻辑代码:
test_login_to_pay_flow.yaml
test_login_to_pay_flow.py
测试数据
存放在yaml文件中,第一层为说明文字,第二层为请求json:
"查询SKU": {
"skuName": "电子书"
}
"添加购物车": {
"skuId": 123,
"skuNum": 2
}
"下单": {
"orderId": 222,
"payAmount": "0.2"
}
"支付": {
"skuId": 123,
"price": 0.1,
"skuNum": 2,
"totalPrice": 0.2
}
测试标题
测试标题采用了@allure.title("")
:
@allure.title("从登录到下单支付")
def test(login, env_vars, case_vars):
测试步骤
一条测试用例由多个测试步骤组成:
@allure.title("从登录到下单支付")
def test(login, env_vars, case_vars):
case_vars.put("token", login["token"])
cache = TepCache(env_vars=env_vars, case_vars=case_vars)
Step("搜索商品", step_search_sku, cache)
Step("添加购物车", step_add_cart, cache)
Step("下单", step_order, cache)
Step("支付", step_pay, cache)
Step第一个参数为步骤描述,第二个参数为步骤实现函数,第三个参数为cache缓存。
步骤实现函数定义在用例文件中:
def step_add_cart(cache: TepCache):
url = cache.env_vars["domain"] + "/addCart"
headers = {"token": cache.case_vars.get("token")}
body = data("添加购物车")
body["skuId"] = cache.case_vars.get("skuId")
response = request("post", url=url, headers=headers, json=body)
assert response.status_code < 400
cache.case_vars.put("skuNum", response.jsonpath("$.skuNum"))
cache.case_vars.put("totalPrice", response.jsonpath("$.totalPrice"))
从上往下依次为url、headers、body、参数化、请求调用、断言、数据提取。
小技巧:打开PyCharm的Show Members,就能快速定位到某个step函数,编辑测试步骤。
变量
环境变量:在resources/env_vars
下预填变量,在resources/tep.yaml
中激活某个环境,在代码中引入env_vars fixture读取变量值:
def test(env_vars):
logger.info(env_vars["domain"])
全局变量:在resources/global_vars.yaml
预填变量,在代码中引入global_vars fixture读取变量值:
def test(global_vars):
print(global_vars["desc"])
用例变量:在用例中引入case_vars fixture,在步骤函数间通过cache传递:
@allure.title("从登录到下单支付")
def test(login, env_vars, case_vars):
case_vars.put("token", login["token"])
cache = TepCache(env_vars=env_vars, case_vars=case_vars)
Step("搜索商品", step_search_sku, cache)
Step("添加购物车", step_add_cart, cache)
Step("下单", step_order, cache)
Step("支付", step_pay, cache)
def step_search_sku(cache: TepCache):
url = cache.env_vars["domain"] + "/searchSku"
headers = {"token": cache.case_vars.get("token")}
body = data("查询SKU")
response = request("get", url=url, headers=headers, params=body)
assert response.status_code < 400
cache.case_vars.put("skuId", response.jsonpath("$.skuId"))
cache.case_vars.put("skuPrice", response.jsonpath("$.price"))
def step_add_cart(cache: TepCache):
url = cache.env_vars["domain"] + "/addCart"
headers = {"token": cache.case_vars.get("token")}
body = data("添加购物车")
body["skuId"] = cache.case_vars.get("skuId")
response = request("post", url=url, headers=headers, json=body)
assert response.status_code < 400
cache.case_vars.put("skuNum", response.jsonpath("$.skuNum"))
cache.case_vars.put("totalPrice", response.jsonpath("$.totalPrice"))
接口关联
如上所述,通过case_vars和cache实现了步骤函数里面的接口关联,上一个接口的响应,提取后存入cache,下一个接口的入参,从cache取值。
数据提取
utils/http_client.py
封装了requests.Response,添加了jsonpath方法,支持简单取值:
response.jsonpath("$.skuNum")
断言
采用Python原生的assert断言。16种常用断言如下:
import allure
@allure.title("等于")
def test_assert_equal():
assert 1 == 1
@allure.title("不等于")
def test_assert_not_equal():
assert 1 != 2
@allure.title("大于")
def test_assert_greater_than():
assert 2 > 1
@allure.title("小于")
def test_assert_less_than():
assert 1 < 2
@allure.title("大于等于")
def test_assert_less_or_equals():
assert 2 >= 1
assert 2 >= 2
@allure.title("小于等于")
def test_assert_greater_or_equals():
assert 1 <= 2
assert 1 <= 1
@allure.title("长度相等")
def test_assert_length_equal():
assert len("abc") == len("123")
@allure.title("长度大于")
def test_assert_length_greater_than():
assert len("hello") > len("123")
@allure.title("长度小于")
def test_assert_length_less_than():
assert len("hi") < len("123")
@allure.title("长度大于等于")
def test_assert_length_greater_or_equals():
assert len("hello") >= len("123")
assert len("123") >= len("123")
@allure.title("长度小于等于")
def test_assert_length_less_or_equals():
assert len("123") <= len("hello")
assert len("123") <= len("123")
@allure.title("字符串相等")
def test_assert_string_equals():
assert "dongfanger" == "dongfanger"
@allure.title("以...开头")
def test_assert_startswith():
assert "dongfanger".startswith("don")
@allure.title("以...结尾")
def test_assert_startswith():
assert "dongfanger".endswith("er")
@allure.title("正则匹配")
def test_assert_regex_match():
import re
assert re.findall(r"don.*er", "dongfanger")
@allure.title("包含")
def test_assert_contains():
assert "fang" in "dongfanger"
assert 2 in [2, 3]
assert "x" in {"x": "y"}.keys()
@allure.title("类型匹配")
def test_assert_type_match():
assert isinstance(1, int)
assert isinstance(0.2, float)
assert isinstance(True, bool)
assert isinstance(3e+26j, complex)
assert isinstance("hi", str)
assert isinstance([1, 2], list)
assert isinstance((1, 2), tuple)
assert isinstance({"a", "b", "c"}, set)
assert isinstance({"x": 1}, dict)
测试报告
allure下载地址: https://github.com/allure-framework/allure2/releases
解压后将bin目录添加到系统环境变量Path。
在pytest命令行添加参数--tep-reports
就能一键生成Allure测试报告,并且会把请求入参和响应出参,记录在测试报告中。
pytest --tep-reports
若想在资源管理器中打开,需要执行命令allure open 报告所在文件夹名
才能正常打开。
用例执行
串行
使用pytest
命令即可。
并行
使用pytest -n auto
,由pytest-xdist提供支持。
特色功能
fixtures自动导入
不是必须在conftest.py里面定义fixture。只要在fixtures目录下,创建以fixture_
开头的文件,fixture都会自动加载到pytest中,方便管理维护。
全局执行一次登录
预置了fixtures/fixture_login.py
登录接口,且全局仅执行一次,解决token复用问题:
import pytest
from loguru import logger
from utils.http_client import request
@pytest.fixture(scope="session")
def login(tep_context_manager, env_vars):
"""
tep_context_manager是为了兼容pytest-xdist分布式执行的上下文管理器
该login只会在整个运行期间执行一次
"""
def produce_expensive_data(variable):
logger.info("----------------开始登录----------------")
response = request(
"post",
url=variable["domain"] + "/login",
headers={"Content-Type": "application/json"},
json={"username": "dongfanger", "password": "123456"}
)
assert response.status_code < 400
logger.info("----------------登录成功----------------")
return response.json()
return tep_context_manager(produce_expensive_data, env_vars)
即便在xdist分布式场景下,也只会执行一次登录。
工具包
cache.py,提供缓存。
class TepCache:
"""
提供缓存服务,包括全局变量、环境变量、用例变量
"""
def __init__(self, global_vars=None, env_vars=None, case_vars=None):
self.global_vars = global_vars
self.env_vars = env_vars
self.case_vars = case_vars
dao.py,目前支持访问MySQL,需要安装pymysql、sqlalchemy和pandas。
#!/usr/bin/python
# encoding=utf-8
"""
@Author : dongfanger
@Date : 9/2/2020 11:32 AM
@Desc : 访问数据库
"""
from loguru import logger
try:
from sqlalchemy import create_engine
from texttable import Texttable
except ModuleNotFoundError:
pass
def mysql_engine(host, port, user, password, db):
try:
engine = create_engine(f"mysql+pymysql://{user}:{password}@{host}:{port}/{db}")
except NameError:
return ""
return engine
def print_db_table(data_frame):
"""以表格形式打印数据表"""
tb = Texttable()
tb.header(data_frame.columns.array)
tb.set_max_width(0)
# text * cols
tb.set_cols_dtype(['t'] * data_frame.shape[1])
tb.add_rows(data_frame.to_numpy(), header=False)
logger.info(tb.draw())
fastapi_mock.py,示例应用。
#!/usr/bin/python
# encoding=utf-8
import uvicorn
from fastapi import FastAPI, Request
app = FastAPI()
@app.post("/login")
async def login(req: Request):
body = await req.json()
if body["username"] == "dongfanger" and body["password"] == "123456":
return {"token": "de2e3ffu29"}
return ""
@app.get("/searchSku")
def search_sku(req: Request):
if req.headers.get("token") == "de2e3ffu29" and req.query_params.get("skuName") == "电子书":
return {"skuId": "222", "price": "2.3"}
return ""
@app.post("/addCart")
async def add_cart(req: Request):
body = await req.json()
if req.headers.get("token") == "de2e3ffu29" and body["skuId"] == "222":
return {"skuId": "222", "price": "2.3", "skuNum": "3", "totalPrice": "6.9"}
return ""
@app.post("/order")
async def order(req: Request):
body = await req.json()
if req.headers.get("token") == "de2e3ffu29" and body["skuId"] == "222":
return {"orderId": "333"}
return ""
@app.post("/pay")
async def pay(req: Request):
body = await req.json()
if req.headers.get("token") == "de2e3ffu29" and body["orderId"] == "333":
return {"success": "true"}
return ""
if __name__ == '__main__':
uvicorn.run("fastapi_mock:app", host="127.0.0.1", port=5000)
func.py,常用函数,比如pairwise自动生成用例等。
#!/usr/bin/python
# encoding=utf-8
"""
@Author : dongfanger
@Date : 7/24/2020 5:41 PM
@Desc : tep函数库
"""
import copy
import inspect
import itertools
import json
import os
import time
from sys import stdout
import yaml
from loguru import logger
from utils.project import Project
def current_time():
"""
当前时间,年-月-日 时-分-秒
:return:
"""
return time.strftime("%Y-%m-%d %H:%M:%S", time.localtime(time.time()))
def current_date():
"""
当前日期 年-月-日
:return:
"""
return time.strftime("%Y-%m-%d", time.localtime(time.time()))
def print_progress_bar(i):
"""
进度条
"""
c = int(i / 10)
progress = '\r %2d%% [%s%s]'
a = '■' * c
b = '□' * (10 - c)
msg = progress % (i, a, b)
stdout.write(msg)
stdout.flush()
def case_pairwise(option):
"""
pairwise算法
"""
cp = [] # 笛卡尔积
s = [] # 两两拆分
for x in eval('itertools.product' + str(tuple(option))):
cp.append(x)
s.append([i for i in itertools.combinations(x, 2)])
logger.info('笛卡尔积:%s' % len(cp))
del_row = []
print_progress_bar(0)
s2 = copy.deepcopy(s)
for i in range(len(s)): # 对每行用例进行匹配
if (i % 100) == 0 or i == len(s) - 1:
print_progress_bar(int(100 * i / (len(s) - 1)))
t = 0
for j in range(len(s[i])): # 对每行用例的两两拆分进行判断,是否出现在其他行
flag = False
for i2 in [x for x in range(len(s2)) if s2[x] != s[i]]: # 找同一列
if s[i][j] == s2[i2][j]:
t = t + 1
flag = True
break
if not flag: # 同一列没找到,不用找剩余列了
break
if t == len(s[i]):
del_row.append(i)
s2.remove(s[i])
res = [cp[i] for i in range(len(cp)) if i not in del_row]
logger.info('过滤后:%s' % len(res))
return res
def load_yaml(path: str) -> dict:
with open(path, encoding="utf8") as f:
return yaml.load(f.read(), Loader=yaml.FullLoader)
def jwt_headers(token):
"""
jwt请求头
"""
return {"Content-Type": "application/json", "authorization": f"Bearer {token}"}
def data(first_node: str) -> dict:
"""
读用例同名的yaml文件
取首节点的值
"""
caller = inspect.stack()[1]
case_path = os.path.dirname(caller.filename)
basename = os.path.basename(caller.filename)
data_path_yml = os.path.join(case_path, basename.rstrip(".py") + ".yml")
data_path_yaml = os.path.join(case_path, basename.rstrip(".py") + ".yaml")
node_value = {}
if not os.path.exists(data_path_yml) and not os.path.exists(data_path_yaml):
logger.error("数据文件不存在")
return node_value
data_path = data_path_yml if os.path.exists(data_path_yml) else data_path_yaml
try:
return load_yaml(data_path)[first_node]
except KeyError:
logger.error(f"数据文件{data_path}不存在首节点{first_node}")
http_client.py,requests库的猴子补丁,可自定义。
#!/usr/bin/python
# encoding=utf-8
import decimal
import json
import time
import allure
import jsonpath
import requests
import urllib3
from loguru import logger
from requests import Response
urllib3.disable_warnings(urllib3.exceptions.InsecureRequestWarning)
def request(method, url, **kwargs):
template = """\n
Request URL: {}
Request Method: {}
Request Headers: {}
Request Payload: {}
Status Code: {}
Response: {}
Elapsed: {}
"""
start = time.process_time()
response = requests.request(method, url, **kwargs) # requests.request原生用法
end = time.process_time()
elapsed = str(decimal.Decimal("%.3f" % float(end - start))) + "s"
headers = kwargs.get("headers", {})
kwargs.pop("headers")
payload = kwargs
log = template.format(url, method, json.dumps(headers), json.dumps(payload), response.status_code, response.text,
elapsed)
logger.info(log)
allure.attach(log, f'request & response', allure.attachment_type.TEXT)
return TepResponse(response)
class TepResponse(Response):
"""
二次封装requests.Response,添加额外方法
"""
def __init__(self, response):
super().__init__()
for k, v in response.__dict__.items():
self.__dict__[k] = v
def jsonpath(self, expr):
"""
此处强制取第一个值,便于简单取值
如果复杂取值,建议直接jsonpath原生用法
"""
return jsonpath.jsonpath(self.json(), expr)[0]
mitm.py,流量录制,做的不是很好,将就看看。
#!/usr/bin/python
# encoding=utf-8
# mitmproxy录制流量自动生成用例
import os
import time
from mitmproxy import ctx
project_dir = os.path.dirname(os.path.dirname(os.path.abspath(__file__)))
tests_dir = os.path.join(project_dir, "tests")
# tests/mitm
mitm_dir = os.path.join(tests_dir, "mitm")
if not os.path.exists(mitm_dir):
os.mkdir(mitm_dir)
# 当前时间作为文件名
filename = f'test_{time.strftime("%Y%m%d_%H%M%S", time.localtime())}.py'
case_file = os.path.join(mitm_dir, filename)
# 生成用例文件
template = """import allure
from utils.http_client import request
@allure.title("")
def test(env_vars):
"""
if not os.path.exists(case_file):
with open(case_file, "w", encoding="utf8") as fw:
fw.write(template)
class Record:
def __init__(self, domains):
self.domains = domains
def response(self, flow):
if self.match(flow.request.url):
# method
method = flow.request.method.lower()
ctx.log.error(method)
# url
url = flow.request.url
ctx.log.error(url)
# headers
headers = dict(flow.request.headers)
ctx.log.error(headers)
# body
body = flow.request.text or {}
ctx.log.error(body)
with open(case_file, "a", encoding="utf8") as fa:
fa.write(self.step(method, url, headers, body))
def match(self, url):
if not self.domains:
ctx.log.error("必须配置过滤域名")
exit(-1)
for domain in self.domains:
if domain in url:
return True
return False
def step(self, method, url, headers, body):
if method == "get":
body_grammar = f"params={body}"
else:
body_grammar = f"json={body}"
return f"""
# 描述
# 数据
# 请求
response = request(
"{method}",
url="{url}",
headers={headers},
{body_grammar}
)
# 提取
# 断言
assert response.status_code < 400
"""
# ==================================配置开始==================================
addons = [
Record(
# 过滤域名
[
"http://www.httpbin.org",
"http://127.0.0.1:5000"
],
)
]
# ==================================配置结束==================================
"""
==================================命令说明开始==================================
# 正向代理(需要手动打开代理)
mitmdump -s mitm.py
# 反向代理
mitmdump -s mitm.py --mode reverse:http://127.0.0.1:5000 --listen-host 127.0.0.1 --listen-port 8000
==================================命令说明结束==================================
"""
project.py,项目基本信息,比如根目录路径。
import os
class Project:
root_dir = os.path.dirname(os.path.dirname(os.path.abspath(__file__)))
data_dir = os.path.join(root_dir, "data")
step.py,测试步骤的泛化调用。
from loguru import logger
from utils.cache import TepCache
class Step:
"""
测试步骤,泛化调用
"""
def __init__(self, name: str, action, cache: TepCache):
logger.info("----------------" + name + "----------------")
action(cache)
关于tep的更多技术细节,请在源码中一探究竟吧。也可以添加微信cekaigang,随时与我联系。
结束与开始
tep小工具发布了1.0.0正式版,我也将不再对其进行维护。对Pytest做接口测试自动化的探索远没有结束,我将从EasyPytest测试平台开始,继续研究Pytest框架的自动化落地实践,那些对测试技术的热情,终将使我们再次相遇。
EasyPytest平台开发版: https://gitee.com/dongfanger/easy-pytest
EasyPytest平台正式版: https://github.com/dongfanger/EasyPytest
标签:case,vars,cache,import,正式版,维护,path,tep1.0,def From: https://www.cnblogs.com/df888/p/17009073.html