深度解析python自动化框架(unnitest+selenium+htmlreport)

1.最基本的错误截图要能提供吧。

2.要是能提供出错页加载的接口信息该多好!(这个是我正在思考的问题)

3.如果可以的话,在自动化录制过程中,能全程录制,这样也利于跟踪bug(这个应该比较容易实现,毕业屏幕录制软件不少,可能需要一个脚本动态启动他们,结束关闭他们。)

在这里我顺便扯一句,可能有的公司认为自动化测试没什么价值,原因:自动化测试根本发现不了什么太多的bug。但是,我的想法是:

一、自动化测试核心不是发现多少bug,而是能确保你的项目主流程么有bug,能正常上线!如果自动化都能发现n个bug,手动人员都不需要测试的了!

二、.理论上讲自动化能覆盖所有手动测试的流程,但是你给自动化投入了多少,靠1到2个人?还更不上迭代的进度!

三、对于我们自动化人员自身而言,也要提高自己,你的价值一定不是只写写脚本,比如搭建环境的能力,搭建web框架的能力,开发桌面应用程序的能力,这些东西都能给手动测试提供便利,也是你的存在感最直接的体现!

回到正题吧,下面讲讲我的思路吧,我的想法如下:

1.因为测试的是流程,前面的test有问题后面的test我要跳过,执行也没有意思,但是报告中要体现!

2.如果出错我要能提供准确的截图。

3.我改写的unnitest要尽量方便,不要在写case上搞一堆的try….except,既不美观,也一直在写相同的代码。根据代码重构的思想,如果你一直在复制黏贴代码,你的代码一定是可以重构的!

我这里继承了unnitest.testcase类,上篇随便我说过,unnitest就是testcase的run方法与testresult的游戏,那么我们就这样想,在执行run的时候如果判断为asserterror或者webdriverexception我们认为是fail,fail就自动截图(前提能拿到webdriver实例);如果是其他的错误我们认为是error。注意fail和error是不同的 ,fail一般是bug(除非你不小心写错代码,导致控件找不到),error是除了以上2中异常的错误,多半是运行错误不太可能是bug。多半是你自己的代码写法有问题。贴出代码如下:

#coding=utf-8
import unittest
import sys
from selenium.common import exceptions
from selenium import webdriver
class skiptest(exception):
“””
raise this exception in a test to skip it.
usually you can use testcase.skiptest() or one of the skipping decorators
instead of raising this directly.
“””
pass
class _expectedfailure(exception):
“””
raise this when a test is expected to fail.
this is an implementation detail.
“””
def __init__(self, exc_info):
super(_expectedfailure, self).__init__()
self.exc_info = exc_info
class iqunxingtest(unittest.testcase):
__screenshot_path =”c:\screenshot”#拿不到截图路径默认路径
def run(self, result=none):
orig_result = result
if result is none:
result = self.defaulttestresult()
starttestrun = getattr(result, ‘starttestrun’, none)
if starttestrun is not none:
starttestrun()
self._resultfordocleanups = result
screenshot_path=getattr(result,”screenshot_path”,false)#尝试从result哪里拿到截图目录
if not screenshot_path:
screenshot_path=self.__screenshot_path
result.starttest(self)
testmethod = getattr(self, self._testmethodname)
if (getattr(self.__class__, “__unittest_skip__”, false) or
getattr(testmethod, “__unittest_skip__”, false)):
# if the class or method was skipped.
try:
skip_why = (getattr(self.__class__, ‘__unittest_skip_why__’, ”)
or getattr(testmethod, ‘__unittest_skip_why__’, ”))
self._addskip(result, skip_why)
finally:
result.stoptest(self)
return
try:
success = false
try:
self.setup()
except skiptest as e:
self._addskip(result, str(e))
except keyboardinterrupt:
raise
except:
result.adderror(self, sys.exc_info())#启动setup失败直接判断出错
else:
try:
testmethod()
except keyboardinterrupt:
raise
except (self.failureexception,exceptions.webdriverexception):#如果是断言错误或webdriverexception,类型为fail,且增加截图
#增加截图
browser=self.getbrowser()#尝试拿浏览器实例
if browser:
filename=self.__class__.__name__+”_”+self._testmethodname+”.png”#格式:类名+方面名称
browser.get_screenshot_as_file(screenshot_path+”\\”+filename)
reback_filename=filename
else:
reback_filename=none
result.addfailure(self, sys.exc_info(),reback_filename)#回传截图名称给report,以便能显示在报告中
except skiptest as e:#如果为跳过的异常,类型为skip异常
self._addskip(result, str(e))
except:#其他类型错误全部为error
result.adderror(self, sys.exc_info())
else:
success = true
try:
self.teardown()
except keyboardinterrupt:
raise
except:
result.adderror(self, sys.exc_info())
success = false
cleanupsuccess = self.docleanups()
success = success and cleanupsuccess
if success:
result.addsuccess(self)
finally:
result.stoptest(self)
if orig_result is none:
stoptestrun = getattr(result, ‘stoptestrun’, none)
if stoptestrun is not none:
stoptestrun()
def getbrowser(self):
#首先从类属性里面找webdriver实例
for value in self.__class__.__dict__.itervalues():
if isinstance(value,webdriver.firefox):#如果value是webdriver.firefox实例
return value
#再从实例属性里面找
for value in self.__dict__.itervalues():
if isinstance(value, webdriver.firefox): # 如果value是webdriver.firefox实例
return value
#实在找不到返回none
return none

代码很简单,可能也有一些么有考虑到的bug,不过思路就是这样!下面贴出我的单元测试代码的demo:

#coding=utf-8
import iqunxingtest
import htmltestrunner
import sys,os
import unittest
from selenium import webdriver
class mydemo(iqunxingtest.iqunxingtest):
u”’测试百度”’
@classmethod
def setupclass(cls):
cls.browser=webdriver.firefox()
@unittest.myskip
def test1(self):
u”’打开百度”’
browser=self.browser
browser.get(“http://www.baidu.com”)
@unittest.myskip
def test2(self):
u”’输入框输入selenium”’
browser = self.browser
browser.find_element_by_id(“kw”).send_keys(“selenium”)
@unittest.myskip
def test3(self):
u”’点击百度一下”’
browser = self.browser
browser.find_element_by_id(“su1”).click()
if __name__ == ‘__main__’:
module_name=os.path.basename(sys.argv[0]).split(“.”)[0]
module=__import__(module_name)
runner=htmltestrunner.htmltestrunner(“new.html”)
all_suite=unittest.defaulttestloader.loadtestsfrommodule(module)
runner.run(all_suite)

注意:由于我们要在报告中显示文档属性,而我们的case使用了装饰器unittest.myskip,所以我们修改上篇中使用到的myskip(我的问题,以前没发现点)增加@wraps(func)即可完整代码如下:

from functools import wraps
def myskip(func):
@wraps(func)
def rebacktest(self):
if self._resultfordocleanups.failures or self._resultfordocleanups.errors:
raise unittest.skiptest(“{} do not excute because {} is failed”.format(func.__name__,self._resultfordocleanups.failures[0][0]._testmethodname))
func(self)
return rebacktest

其实上面的testcase还是挺简洁的,我们把工作放在我们的iqunxingtest的run方法里面,当执行test时,asserterror与webdriverexception错误时,自动截图,当然是找到浏览器的实例对象,找不到就不截图了,回传给result也是none,与普通的单元测试无异。

htmltestrunner的完整代码如下:(这个代码太长了,实在不想写出该了什么完整贴出来吧!)

#coding=utf-8
“””
a testrunner for use with the python unit testing framework. it
generates a html report to show the result at a glance.
the simplest way to use this is to invoke its main method. e.g.
import unittest
import htmltestrunner
… define your tests …
if __name__ == ‘__main__’:
htmltestrunner.main()
for more customization options, instantiates a htmltestrunner object.
htmltestrunner is a counterpart to unittest’s texttestrunner. e.g.
# output to a file
fp = file(‘my_report.html’, ‘wb’)
runner = htmltestrunner.htmltestrunner(
stream=fp,
title=’my unit test’,
description=’this demonstrates the report output by htmltestrunner.’
)
# use an external stylesheet.
# see the template_mixin class for more customizable options
runner.stylesheet_tmpl = ”
# run the test
runner.run(my_test_suite)
————————————————————————
copyright (c) 2004-2007, wai yip tung
all rights reserved.
redistribution and use in source and binary forms, with or without
modification, are permitted provided that the following conditions are
met:
* redistributions of source code must retain the above copyright notice,
this list of conditions and the following disclaimer.
* redistributions in binary form must reproduce the above copyright
notice, this list of conditions and the following disclaimer in the
documentation and/or other materials provided with the distribution.
* neither the name wai yip tung nor the names of its contributors may be
used to endorse or promote products derived from this software without
specific prior written permission.
this software is provided by the copyright holders and contributors “as
is” and any express or implied warranties, including, but not limited
to, the implied warranties of merchantability and fitness for a
particular purpose are disclaimed. in no event shall the copyright owner
or contributors be liable for any direct, indirect, incidental, special,
exemplary, or consequential damages (including, but not limited to,
procurement of substitute goods or services; loss of use, data, or
profits; or business interruption) however caused and on any theory of
liability, whether in contract, strict liability, or tort (including
negligence or otherwise) arising in any way out of the use of this
software, even if advised of the possibility of such damage.
“””
# url: http://tungwaiyip.info/software/htmltestrunner.html
__author__ = “wai yip tung”
__version__ = “0.8.2”
“””
change history
version 0.8.2
* show output inline instead of popup window (viorel lupu).
version in 0.8.1
* validated xhtml (wolfgang borgert).
* added description of test classes and test cases.
version in 0.8.0
* define template_mixin class for customization.
* workaround a ie 6 bug that it does not treat 《script》 block as cdata.
version in 0.7.1
* back port to python 2.3 (frank horowitz).
* fix missing scroll bars in detail log (podi).
“””
# todo: color stderr
# todo: simplify javascript using ,ore than 1 class in the class attribute?
#coding=utf-8
import datetime
import io
import sys
import os
reload(sys)
sys.setdefaultencoding(‘utf8’)
import time
import unittest
import re
from xml.sax import saxutils
# ————————————————————————
# the redirectors below are used to capture output during testing. output
# sent to sys.stdout and sys.stderr are automatically captured. however
# in some cases sys.stdout is already cached before htmltestrunner is
# invoked (e.g. calling logging.basicconfig). in order to capture those
# output, use the redirectors for the cached stream.
#
# e.g.
# >>> logging.basicconfig(stream=htmltestrunner.stdout_redirector)
# >>>
pattren=re.compile(r”(.*)reback_filename:(.*)”,re.s)
screenshot_dir_name=”screen_shot”
class outputredirector(object):
“”” wrapper to redirect stdout or stderr “””
def __init__(self, fp):
self.fp = fp
def write(self, s):
self.fp.write(s)
def writelines(self, lines):
self.fp.writelines(lines)
def flush(self):
self.fp.flush()
stdout_redirector = outputredirector(sys.stdout)
stderr_redirector = outputredirector(sys.stderr)
# ———————————————————————-
# template
class template_mixin(object):
“””
define a html template for report customerization and generation.
overall structure of an html report
html
+————————+
| |
| |
| |
| stylesheet |
| +—————-+ |
| | | |
| +—————-+ |
| |
| |
| |
| |
| |
| heading |
| +—————-+ |
| | | |
| +—————-+ |
| |
| report |
| +—————-+ |
| | | |
| +—————-+ |
| |
| ending |
| +—————-+ |
| | | |
| +—————-+ |
| |
| |
| |
+————————+
“””
status = {
0: ‘pass’,
1: ‘fail’,
2: ‘error’,
3:’skip’
}
default_title = ‘unit test report’
default_description = ”
# ————————————————————————
# html template
html_tmpl = r”””

%(title)s

%(stylesheet)s

《script》
%(heading)s
%(report)s
%(ending)s

“””
# variables: (title, generator, stylesheet, heading, report, ending)
# ————————————————————————
# stylesheet
#
# alternatively use a for external style sheet, e.g.
#
stylesheet_tmpl = “””

“””
# ————————————————————————
# heading
#
heading_tmpl = “””

%(title)s
%(parameters)s

%(description)s

“”” # variables: (title, parameters, description)
heading_attribute_tmpl = “””

%(name)s: %(value)s

“”” # variables: (name, value)
# ————————————————————————
# report
#
report_tmpl = “””

show
summary
failed
all

test group/test case
count
pass
fail
error
skip
view
screenshot

%(test_list)s

total
%(count)s
%(pass)s
%(fail)s
%(error)s
%(skip)s

“”” # variables: (test_list, count, pass, fail, error)
report_class_tmpl = r”””

%(desc)s
%(count)s
%(pass)s
%(fail)s
%(error)s
%(skip)s
detail

“”” # variables: (style, desc, count, pass, fail,skip, error, cid)
report_test_with_output_tmpl = r”””

%(desc)s

%(status)s

[x]

%(script)s

picture_shot

“”” # variables: (tid, class, style, desc, status)
report_test_no_output_tmpl = r”””

%(desc)s

%(status)s

picture_shot

“”” # variables: (tid, class, style, desc, status)
report_test_output_tmpl = r”””
%(id)s: %(output)s
“””
# variables: (id, output)
# ————————————————————————
# ending
#
ending_tmpl = “””

“””
# ——————– the end of the template class ——————-
testresult = unittest.testresult
class _testresult(testresult):
# note: _testresult is a pure representation of results.
# it lacks the output and reporting ability compares to unittest._texttestresult.
def __init__(self, verbosity=1):
testresult.__init__(self)
self.stdout0 = none
self.stderr0 = none
self.success_count = 0
self.skipped_count=0#add skipped_count
self.failure_count = 0
self.error_count = 0
self.verbosity = verbosity
# result is a list of result in 4 tuple
# (
# result code (0: success; 1: fail; 2: error),
# testcase object,
# test output (byte string),
# stack trace,
# )
self.result = []
def starttest(self, test):
testresult.starttest(self, test)
# just one buffer for both stdout and stderr
self.outputbuffer = io.bytesio()
stdout_redirector.fp = self.outputbuffer
stderr_redirector.fp = self.outputbuffer
self.stdout0 = sys.stdout
self.stderr0 = sys.stderr
sys.stdout = stdout_redirector
sys.stderr = stderr_redirector
def complete_output(self):
“””
disconnect output redirection and return buffer.
safe to call multiple times.
“””
if self.stdout0:
sys.stdout = self.stdout0
sys.stderr = self.stderr0
self.stdout0 = none
self.stderr0 = none
return self.outputbuffer.getvalue()
def stoptest(self, test):
# usually one of addsuccess, adderror or addfailure would have been called.
# but there are some path in unittest that would bypass this.
# we must disconnect stdout in stoptest(), which is guaranteed to be called.
self.complete_output()
def addsuccess(self, test):
self.success_count += 1
testresult.addsuccess(self, test)
output = self.complete_output()
self.result.append((0, test, output, ”))
if self.verbosity > 1:
sys.stderr.write(‘ok ‘)
sys.stderr.write(str(test))
sys.stderr.write(‘\n’)
else:
sys.stderr.write(‘.’)
def addskip(self, test, reason):
self.skipped_count+= 1
testresult.addskip(self, test,reason)
output = self.complete_output()
self.result.append((3, test,”,reason))
if self.verbosity > 1:
sys.stderr.write(‘skip ‘)
sys.stderr.write(str(test))
sys.stderr.write(‘\n’)
else:
sys.stderr.write(‘s’)
def adderror(self, test, err):
self.error_count += 1
testresult.adderror(self, test, err)
_, _exc_str = self.errors[-1]
output = self.complete_output()
self.result.append((2, test, output, _exc_str))
if self.verbosity > 1:
sys.stderr.write(‘e ‘)
sys.stderr.write(str(test))
sys.stderr.write(‘\n’)
else:
sys.stderr.write(‘e’)
def addfailure(self, test, err,reback_filename):
self.failure_count += 1
testresult.addfailure(self, test, err)
_, _exc_str = self.failures[-1]
if reback_filename:
_exc_str=_exc_str+”reback_filename:”+reback_filename
output = self.complete_output()
self.result.append((1, test, output, _exc_str))
if self.verbosity > 1:
sys.stderr.write(‘f ‘)
sys.stderr.write(str(test))
sys.stderr.write(‘\n’)
else:
sys.stderr.write(‘f’)
# def addfailure(self, test, err):
# self.failure_count += 1
# testresult.addfailure(self, test, err)
# _, _exc_str = self.failures[-1]
# # if reback_filename:
# # _exc_str = _exc_str + “reback_filename:” + reback_filename
# output = self.complete_output()
# self.result.append((1, test, output, _exc_str))
# if self.verbosity > 1:
# sys.stderr.write(‘f ‘)
# sys.stderr.write(str(test))
# sys.stderr.write(‘\n’)
# else:
# sys.stderr.write(‘f’)
class htmltestrunner(template_mixin):
“””
“””
def __init__(self,reportpath,verbosity=1, title=none, description=none,name=none):
dirname=os.path.dirname(reportpath)
if dirname==”.”or dirname==””:
self.screen_shot_path =screenshot_dir_name
else:
self.screen_shot_path=dirname+”/”+screenshot_dir_name
if not os.path.exists(self.screen_shot_path):
os.mkdir(self.screen_shot_path)
self.stream =open(reportpath,”wb”)
self.verbosity = verbosity
if title is none:
self.title = self.default_title
else:
self.title = title
if name is none:
self.name =”
else:
self.name = name
if description is none:
self.description = self.default_description
else:
self.description = description
self.starttime = datetime.datetime.now()
def run(self, test):
“run the given test case or test suite.”
result = _testresult(self.verbosity)
result.__dict__.update({“screenshot_path”:self.screen_shot_path})
test(result)
self.stoptime = datetime.datetime.now()
self.generatereport(test, result)
# print (sys.stderr, ‘\ntime elapsed: %s’ % (self.stoptime-self.starttime))
return result
def sortresult(self, result_list):
# unittest does not seems to run in any particular order.
# here at least we want to group them together by class.
rmap = {}
classes = []
for n,t,o,e in result_list:
cls = t.__class__
if not cls in rmap:
rmap[cls] = []
classes.append(cls)
rmap[cls].append((n,t,o,e))
r = [(cls, rmap[cls]) for cls in classes]
return r
def getreportattributes(self, result):
“””
return report attributes as a list of (name, value).
override this to add custom attributes.
“””
starttime = str(self.starttime)[:19]
duration = str(self.stoptime – self.starttime)
status = []
if result.success_count: status.append(‘pass %s’ % result.success_count)
if result.failure_count: status.append(‘failure %s’ % result.failure_count)
if result.skipped_count: status.append(‘skip %s’ % result.skipped_count)
if result.error_count: status.append(‘error %s’ % result.error_count )
if status:
status = ‘ ‘.join(status)
else:
status = ‘none’
return [
(‘start time’, starttime),
(‘duration’, duration),
(‘status’, status),
]
def generatereport(self, test, result):
report_attrs = self.getreportattributes(result)#报告的头部
generator = ‘htmltestrunner %s’ % __version__
stylesheet = self._generate_stylesheet()#拿到css文件
heading = self._generate_heading(report_attrs)
report = self._generate_report(result)
ending = self._generate_ending()
output = self.html_tmpl % dict(
title = saxutils.escape(self.title),
generator = generator,
stylesheet = stylesheet,
heading = heading,
report = report,
ending = ending,
)
self.stream.write(output.encode(‘utf8’))
def _generate_stylesheet(self):
return self.stylesheet_tmpl
def _generate_heading(self, report_attrs):
a_lines = []
for name, value in report_attrs:
line = self.heading_attribute_tmpl % dict(
name = saxutils.escape(name),
value = saxutils.escape(value),
)
a_lines.append(line)
heading = self.heading_tmpl % dict(
title = saxutils.escape(self.title),
parameters = ”.join(a_lines),
description = saxutils.escape(self.description),
)
return heading
#根据result收集报告
def _generate_report(self, result):
rows = []
sortedresult = self.sortresult(result.result)
i = 0
for cid, (cls, cls_results) in enumerate(sortedresult):
# subtotal for a class
np = nf =ns=ne = 0#np代表pass个数,nf代表fail,ns代表skip,ne,代表error
for n,t,o,e in cls_results:
if n == 0: np += 1
elif n == 1: nf += 1
elif n==3:ns+=1
else: ne += 1
# format class description
# if cls.__module__ == “__main__”:
# name = cls.__name__
# else:
# name = “%s.%s” % (cls.__module__, cls.__name__)
name = cls.__name__
try:
core_name=self.name[i]
except exception,e:
core_name =”
# doc = (cls.__doc__)+core_name and (cls.__doc__+core_name).split(“\n”)[0] or “”
doc = (cls.__doc__) and cls.__doc__ .split(“\n”)[0] or “”
desc = doc and ‘%s: %s’ % (name, doc) or name
i=i+1
#生成每个testcase类的汇总数据,对于报告中的
row = self.report_class_tmpl % dict(
style = ne > 0 and ‘errorclass’ or nf > 0 and ‘failclass’ or ‘passclass’,
desc = desc,
count = np+nf+ne+ns,
pass = np,
fail = nf,
error = ne,
skip=ns,
cid = ‘c%s’ % (cid+1),
)
rows.append(row)
#生成每个testcase类中所有方法的测试结果
for tid, (n,t,o,e) in enumerate(cls_results):
self._generate_report_test(rows, cid, tid, n, t, o, e)
report = self.report_tmpl % dict(
test_list = ”.join(rows),
count = str(result.success_count+result.failure_count+result.error_count+result.skipped_count),
pass = str(result.success_count),
fail = str(result.failure_count),
error = str(result.error_count),
skip=str(result.skipped_count)
)
return report
def _generate_report_test(self, rows, cid, tid, n, t, o, e):
# e.g. ‘pt1.1’, ‘ft1.1’, etc
has_output = bool(o or e)
tid = (n == 0 and ‘p’ or n==3 and ‘s’ or ‘f’) + ‘t%s.%s’ % (cid+1,tid+1)
name = t.id().split(‘.’)[-1]
doc = t.shortdescription() or “”
desc = doc and (‘%s: %s’ % (name, doc)) or name
tmpl = has_output and self.report_test_with_output_tmpl or self.report_test_no_output_tmpl
uo1=””
# o and e should be byte string because they are collected from stdout and stderr?
uo =o
# else:
# uo = e
# if isinstance(e,str):
# # todo: some problem with ‘string_escape’: it escape \n and mess up formating
# # ue = unicode(e.encode(‘string_escape’))
# ue = e
# else:
ue =e
hidde_status = ”’hidden=”hidden””’
image_url = ”
if “reback_filename” in str(saxutils.escape(str(ue))):
try:
pattren_result=re.search(pattren,str(saxutils.escape(str(ue))))
ue=pattren_result.group(1)
file_name=pattren_result.group(2)
hidde_status = ”
image_url=screenshot_dir_name+”/”+file_name
except exception,e:
pass
script = self.report_test_output_tmpl % dict(
id = tid,
output = saxutils.escape(str(uo) + str(ue))
)
row = tmpl % dict(
tid = tid,
class = (n == 0 and ‘hiddenrow’ or ‘none’),
errorcase’ or (n == 1 and ‘failcase’) or (n == 3 and ‘skipcase’ or ‘none’),
desc = desc,
script = script,
hidde=hidde_status,
image=image_url,
status = self.status[n],
)
rows.append(row)
if not has_output:
return
def _generate_ending(self):
return self.ending_tmpl
##############################################################################
# facilities for running tests from the command line
##############################################################################
# note: reuse unittest.testprogram to launch test. in the future we may
# build our own launcher to support more specific command line
# parameters like test title, css, etc.
# class testprogram(unittest.testprogram):
# “””
# a variation of the unittest.testprogram. please refer to the base
# class for command line parameters.
# “””
# def runtests(self):
# # pick htmltestrunner as the default test runner.
# # base class’s testrunner parameter is not useful because it means
# # we have to instantiate htmltestrunner before we know self.verbosity.
# if self.testrunner is none:
# self.testrunner = htmltestrunner(verbosity=self.verbosity)
# unittest.testprogram.runtests(self)
#
# main = testprogram
##############################################################################
# executing this module from the command line
##############################################################################
if __name__ == “__main__”:
main(module=none)

运行一下生成的report如下:

当然我们也能点开截图!我在实际中大概就是这么应用的,当然你也可以实现自己的好多想法,我只是说下我思路,下个话题还是关于selenium吧!wxpython暂时搁浅一段时间

以上就是深度解析python自动化框架(unnitest+selenium+htmlreport) 的详细内容,更多请关注 第一php社区 其它相关文章!

Posted in 未分类