深刻解读Python的unittest并拓展HTMLTestRunner

unnitest是Python的一个重要的单元测试框架，对于用Python进行开发的同事们可能不须要对他有过深刻的了解会用就行，可是，对于自动化测试人员我以为是要熟知unnitest的执行原理以及相关模块的做用。我这边提几个简单的需求以下：

1.如何利用unnitest执行流程测试而非单元测试。好比咱们可能利用selenium+unnitest来跑一段流程，好比test1里面咱们实现登录，test2在test1成功登录的基础上，实现一个查询的测试，test3咱们查询一些数据后，页面选择性提交数据。这个你总不能在test1里面登录完而后关闭浏览器；在test2里面再登录再执行查询后关闭浏览器；test3再登录执行提交数据。你或者能够这样想把这几个步骤写在一个test里面。可是，若是流程过长怎么办？本身难道感受不到剪不断理还乱的纠结吗....

2.如何控制unnitest的执行顺序。unnitest里面tests数组里面存放的TestCase默认是以首字母排序的，这对于test1,test2....test9这样的执行顺序是没用问题的，可是对于多个test好比test1,test2.........test18,这样unnitest可不是按照这个顺序，我说了是按照首字母排序来的，他会这样执行test1,test10,....test18,test2,..........test9,固然我说的这些对于单元测试是影响不大的(除非各个test之间有数据依赖关系，后面提到)，对于流程多是颠覆性的。

3.流程测试中如何动态的控制是否跳过某个test的执行。对于流程来讲，这也是常见的一种想法，好比test1我连登录都没有成功，还有意义执行后面的test吗，以后报出来的都是一些NosuchElement的错误，这些错误没有任何意义，并且纯粹浪费时间....出来的报告也是不"人性的"。那么咱们如今一个好的想法是：若是test1没有执行成功，后面的test能动态所有跳过，其实不止是test1没成功，后面test跳过，准确的说是，test1,test2........testN中若是任意某个test没有经过，后面的能动态的所有跳过。固然，若是后面的test和前面test没什么关系，也可能选择无论前面是否成功均不跳过；也能够是只和test1登陆有关，只要登陆成功了我就不跳过，若是登陆不成功我就跳过...还有不少.....更重要的是，报告中有所展示,不能说skip了某个用例，你报告就不显示了，这样老板认为你偷懒，用例写这么少？ 你还要瑟瑟发抖的去解释，是由于前面的用例没经过因此，没显示了...
咱们高大上的是这样的，test1执行失败了，test2.......testN,报告中都体现，标注是skip的case，并且点开还有缘由解释:"test1没有执行成功，因此跳过此case"。这就是我拓展HTMLTestRunner的缘由，后面逐行解释若是拓展它。

.......还有不少

补充一下，为何往流程上扯呢，由于公司的模块太多，并且复杂，单元测试机会不太会用，只要保证各个业务的主流程没问题便可，可是不影响咱们解析unnitest。

咱们的想法不少，可是如何来实现呢？那就让咱们来深刻探讨下python的unnitest吧！

关于unnitest看似复杂我给出来就是unnitest=TestCase+TestResult,只要熟知这2个模块，你就能"随心所欲"！！可能有人说不对不是有什么TestSuite吗还有TextTestRunner等等吗，不错确实咱们平时用到的大可能是这些模块，可是，到其实最终执行的是TestCase中的run方法，并把结果给TestResult(或它的子类)。咱们先来看一个简单的unnitest例子，并以此来拓展！例子以下：

import unittest
class Mydemo(unittest.TestCase):
    def setUp(self):
        self.a=1
    def test1(self):
        print "i am test1 the value of a is {}".format(self.a)
    def test2(self):
        print "i am test2 the value of a is {}".format(self.a)
    def  test3(self):
        print "i am test3 the value of a is {}".format(self.a)
if __name__ == '__main__':
    unittest.main()

运行结果以下：

i am test1 the value of a is 1
...
i am test2 the value of a is 1
i am test3 the value of a is 1
----------------------------------------------------------------------
Ran 3 tests in 0.000s

OK

这个是没有问题的，那么咱们可能要想这个unnitest.main()是什么东西，还有其余的写法来执行吗，能只执行test1,test2,不执行test3吗(暂时不用skip)？那么咱们从unittest.main()看起来。debug进入其实最终执行的是TestProgram这类，贴出构造函数部分代码：

if argv is None:
            argv = sys.argv#获得当前模块的绝对路径

        self.exit = exit
        self.failfast = failfast
        self.catchbreak = catchbreak
        self.verbosity = verbosity
        self.buffer = buffer
        self.defaultTest = defaultTest
        self.testRunner = testRunner
        self.testLoader = testLoader
        self.progName = os.path.basename(argv[0])
        self.parseArgs(argv)#查找当前module的Testsuite
        self.runTests()#执行测试

好了，从上面咱们能够看出来其实也就2个主要的步骤就是第一:找出要测试的testcase，并加入到Testsuite，第二：运行Testsuite并把结果给TestResult。

首先，第一:了解什么是TestCase？什么是TestSuite？第二:若是找出这些Testcase,或者TestSuite？

什么是TestCase？

有人说TesetCase就是以test开头的就叫一个testcase，我只能这样说太偏面的，准确的说：是实例了一个TesetCase类的叫一个TestCase，好比这样：

import unittest
class Mydemo(unittest.TestCase):
    def setUp(self):
        self.a=1
    def Mytest1(self):
        print "i am Mytest1 the value of a is {}".format(self.a)
    def Mytest2(self):
        print "i am Mytest2 the value of a is {}".format(self.a)
    def Mytest3(self):
        print "i am Mytest3 the value of a is {}".format(self.a)
if __name__ == '__main__':
    test_runner=unittest.TextTestRunner()
    test_suit=unittest.TestSuite()
    test_suit.addTests(map(Mydemo,["Mytest1","Mytest2","Mytest3"]))
    test_runner.run(test_suit)

运行结果以下：

...
i am Mytest1 the value of a is 1
----------------------------------------------------------------------
i am Mytest2 the value of a is 1
Ran 3 tests in 0.000s
i am Mytest3 the value of a is 1

OK

 上面3个Testcase可并无以test开头...那么为何你们都要默认以test开头来写呢，咱们打开C:\Python27\Lib\unittest\loader.py这个模块在296行有写defaultTestLoader = TestLoader()，咱们来看看TestLoader这个类第一行就看见testMethodPrefix = 'test'，也就是说若是你使用到defaultTestLoader，那么默认是以test开头的方法为一个用例，具体能够在TestLoader类中的getTestCaseNames获得实现,红字注释部分为何testCaseClass要有__call__方法，咱们后面提到。（不知道__call__这个魔法属性的用法自行百度）

def getTestCaseNames(self, testCaseClass):
        """Return a sorted sequence of method names found within testCaseClass
        """
        def isTestMethod(attrname, testCaseClass=testCaseClass,
                         prefix=self.testMethodPrefix):
            return attrname.startswith(prefix) and \
                hasattr(getattr(testCaseClass, attrname), '__call__')#返回一个testCaseClass有__call__方法且attrname以prefix开头的为一个testcase
        testFnNames = filter(isTestMethod, dir(testCaseClass))
        if self.sortTestMethodsUsing:
            testFnNames.sort(key=_CmpToKey(self.sortTestMethodsUsing))
        return testFnNames

原来是这样啊，咱们上文提到的unittest.main()其实用的就是defaultTestLoader，固然你把if __name__ == '__main__'下面的代码换成unittest.main()确定不成功，除非你把上文提到的testMethodPrefix 换成"Mytest"。有了对TestCase的见解，咱们具体来看看这个类。

这个类里面包含了咱们所能用的方法。我列出来一些主要的吧。

setUp（）在每一个test执行前都要执行的方法。

tearDown()在每一个test执行后都要执行的方法。（不论是否执行成功）

setUpClass()在一个测试类中在全部test开始以前，执行一次且必须使用到Testsuite(只有在TestSuite的run方法里面才对其调用)

tearDownClass()在一个测试类中在全部test结束以后，执行一次且必须使用到Testsuite(只有在TestSuite的run方法里面才对其调用)

run()这是unnitest的核心，逻辑也相对复杂，可是很好理解，具体本身看源码。全部最终case的执行都会归结到该run方法。

还有一个重要的_resultForDoCleanups私有变量，存储TestResult的执行结果，这个在构建后面的skip用到。

咱们要明确TestCase类中全部的测试用例是独立的，我上面说过了，其实每一个testcase就是一个个TestCase类的实例对象，因此不要企图在某个test存储或改变一个变量，下个test中能用到，除非利用到setUpClass。咱们看个例子：

import unittest
class Mydemo(unittest.TestCase):
    def test1(self):
        self.a=1
        print "i am test1 the value of a is {}".format(self.a)
    def test2(self):
        print "i am test2 the value of a is {}".format(self.a)
if __name__ == '__main__':
    unittest.main()

 结果：

C:\Python27\python.exe D:/Moudle/module_1/test4.py
i am test1 the value of a is 1
.E
======================================================================
ERROR: test2 (__main__.Mydemo)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "D:/Moudle/module_1/test4.py", line 7, in test2
    print "i am test2 the value of a is {}".format(self.a)
AttributeError: 'Mydemo' object has no attribute 'a'

----------------------------------------------------------------------
Ran 2 tests in 0.001s

FAILED (errors=1)

上面就是说明TestCase类中全部的测试用例是独立的，每一个testcase就是由TestCase实例化的一个独立的实例。那是否是就是每一个TestCase不能共享数据呢？答案是否认的，不能共享的缘由是咱们上面用到的是self（实例对象属性），能共享咱们就必须使用类属性，好比下个例子：

import unittest
class Mydemo(unittest.TestCase):
    def test1(self):
        Mydemo.a=1
        print "i am test1 the value of a is {}".format(self.a)
    def test2(self):
        print "i am test2 the value of a is {}".format(Mydemo.a)
if __name__ == '__main__':
    unittest.main()

 运行结果以下：

i am test1 the value of a is 1
..
i am test2 the value of a is 1
----------------------------------------------------------------------
Ran 2 tests in 0.000s

OK

 这些东西实际上是python类的一些动态行为，可是既然和unnitest关联，就随便提下。咱们运行test1的时候，给Mydemo加了一个新的属性a（值为1），当咱们运行test2时，咱们就能拿到Mydemo类的属性了。说了TaseCase咱们不得不说下TestSuite。TestSuite是有一个个TestCase组成的，固然TestSuite里面能够再嵌套TestSuite。咱们打开C:\Python27\Lib\unittest\suite.py找到TestSuite，它继承于BaseTestSuite，其实主要的一些属性就那么几个：

1.self._tests这个私有变量里面方的是全部的TestCase或者TestSuite。

2.run()方法，方法以下：

 

def run(self, result, debug=False):
        topLevel = False
        if getattr(result, '_testRunEntered', False) is False:
            result._testRunEntered = topLevel = True

        for test in self:#这个循环会一直遍历_tests中的变量
            if result.shouldStop:
                break
            if _isnotsuite(test):
                self._tearDownPreviousClass(test, result)
                self._handleModuleFixture(test, result)
                self._handleClassSetUp(test, result)#这一句提到了调用setUpClass的规则
                result._previousTestClass = test.__class__

                if (getattr(test.__class__, '_classSetupFailed', False) or
                    getattr(result, '_moduleSetUpFailed', False)):
                    continue

            if not debug:
                test(result)#若是是TestSuit继续调用该方法，若是是TestCase则调用TestCase中的run方法
            else:
                test.debug()

        if topLevel:
            self._tearDownPreviousClass(None, result)
            self._handleModuleTearDown(result)
            result._testRunEntered = False
        return result

 

 注释1：self是个迭代对象，一直遍历上文提到的self._tests变量

注释2：咱们看看_handleClassSetUp中的方法，发如今在用例的执行过程当中，每一个TestCase类只会调用一次setUpClass方法，同理tearDownClass。对用这一点咱们举个例子：

import unittest
class Mydemo(unittest.TestCase):
    @classmethod
    def setUpClass(cls):
        print "I am setUpClass"
    def test1(self):
        print "i am test1 "
    def test2(self):
        print "i am test2"
    @classmethod
    def tearDownClass(cls):
        print "I am tearDownClass"
if __name__ == '__main__':
    unittest.main()

运行结果是：

C:\Python27\python.exe D:/Moudle/module_1/test4.py
I am setUpClass
..
i am test1 
----------------------------------------------------------------------
i am test2
Ran 2 tests in 0.001s
I am tearDownClass

OK

说明类方法setUpClass与tearDownClass只执行了一遍了，这就回答了咱们第一个问题了：在setUpClass中启动浏览器，执行完全部流程后关闭浏览器，举一个简单的demo就是：

 

#coding=utf-8
import unittest
from  selenium import webdriver
class Mydemo(unittest.TestCase):
    @classmethod
    def setUpClass(cls):
        cls.browser=webdriver.Firefox()
    def test1(self):
        '''登陆'''
        browser=self.browser
        #do someting about login
    def test2(self):
        '''查询'''
        browser = self.browser
        # do someting about search
    def test3(self):
        '''提交数据'''
        browser = self.browser
        # do someting about submmit
    @classmethod
    def tearDownClass(cls):
        browser=self.browser
        browser.close()
　　　　
if __name__ == '__main__':
    unittest.main()

 

 上面就会在全部的case执行以前启动firefox,由于每一个test中拿到的都是Mydemo类中同一个webdriver对象，因此能保证操做的都是同一个浏览器句柄。关于这个setUpClass若是想要动态的改变某个值必定要使用python的可变的对象好比list,dict等...这些其实都是一些python类的一些知识,算我啰嗦吧我仍是想举个例子，嫌烦的同窗，绕过这一部分吧。

#coding=utf-8
import unittest
from  selenium import webdriver
class Mydemo(unittest.TestCase):
    @classmethod
    def setUpClass(cls):
        cls.a=1
    def test1(self):
        print "before update the a in test1 is:{}".format(self.a)
        self.a=self.a+1
        print "after update the a in test1 is:{}".format(self.a)
    def test2(self):
        print "the value in test2 is:{}".format(self.a)
    @classmethod
    def tearDownClass(cls):
        print "I am tearDownClass"
if __name__ == '__main__':
    unittest.main()

 运行结果：

C:\Python27\python.exe D:/Moudle/module_1/test4.py
before update the a in test1 is:1
..
after update the a in test1 is:2
----------------------------------------------------------------------
the value in test2 is:1
I am tearDownClass
Ran 2 tests in 0.001s

OK

 咱们想在test1中改变a的值，可是test2中的结果说明a没有被改变，这其实也很好理解。若是咱们想要改变怎么办，看看下面的例子：

#coding=utf-8
import unittest
from  selenium import webdriver
class Mydemo(unittest.TestCase):
    @classmethod
    def setUpClass(cls):
        cls.a=[0]
    def test1(self):
        print "before update the a in test1 is:{}".format(self.a[0])
        self.a[0]=self.a[0]+1
        print "after update the a in test1 is:{}".format(self.a[0])
    def test2(self):
        print "the value in test2 is:{}".format(self.a[0])
    @classmethod
    def tearDownClass(cls):
        print "I am tearDownClass"
if __name__ == '__main__':
    unittest.main()

 运行结果：

C:\Python27\python.exe D:/Moudle/module_1/test4.py
..
before update the a in test1 is:0
----------------------------------------------------------------------
after update the a in test1 is:1
Ran 2 tests in 0.000s
the value in test2 is:1

I am tearDownClass
OK

 咱们把a变成一个list，发现a的值在test2中改变了。好了这一部分就这样了。

注释三：这个其实也是python类的一些知识可能有的人没有关注就是__call__这个魔法属性，咱们看到在这个循环中test若是是testsuite对象，那么会调用中TestSuite类中的__call__方法（在其父类BaseTestSuite中），该方法中会再次调用run方法。一直到test是个testcase对象，那么就会调用咱们上文提到的TestCase中的__call__（这就是咱们上面提到为何找有__call__属性类实例的方法），同样该__call__中的方法也是调用TestCase中的run。因此最终全部的执行其实都是执行TestCase中的run方法。

上面大体讲了一些TestCase与TestSuit的知识，可能穿插的比较多。

如何建立这些Testcase或者TestSuite？

1.本身手动实例化TestCase

这个上面已经有例子，与普通类无异，这中在自动化领域用处不大，咱们不能一个个的实例化吧...

2.利用C:\Python27\Lib\unittest\loader.py模块的TestLoader，该类提供了多种不一样情境find testcase。

1.loadTestsFromTestCase利用给出的TestCase类名称返回找到全部的suite。

2.loadTestsFromMoudle利用给出的Moudle返回找到全部的suite。

3.loadTestsFromName利用给出的Moudle名称返回找到全部的suite。

4.discover返回给定目录下符合pattern类型(默认test*.py)全部的suite。

其实这些方法最终都要归结到loadTestsFromTestCase，可能官方不提供咱们也能写，既然有了就直接用吧。

通过上面的说明，我以为你们对一TestCase，TestSuite应该有一个比较清楚的认识了，也解决了我本身的提问。问题一：咱们能够用类方法setUpClass实现。对于问题二：咱们能够利用TestLoader类中的方法返回suite，而后对这些suite按照本身的想法进行一些排序，而后再调用run方法。说完了TestCase咱们再说下TestResult。

什么是TestResult？

顾名思义，testresult就是存储测试结果的，不过经过何种方式调用run函数，最终到Testcase中的run方法时必须传一个result（若是为None则本身实例化一个TestResult对象)。这个result就是TestResult对象或者是其子类的对象，咱们每次执行的结果都会调用其addFailure，addSuccess，addSkip....等方法将执行结果保存到TestResult实例属性中。咱们仍是来看看TestCase的run方法：

def run(self, result=None):
        orig_result = result
        if result is None:#若是没有传入result对象本身实例化一个TestResult对象
            result = self.defaultTestResult()
            startTestRun = getattr(result, 'startTestRun', None)
            if startTestRun is not None:
                startTestRun()

        self._resultForDoCleanups = result
        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)#调用addSkip
            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())#调用addError
            else:
                try:
                    testMethod()
                except KeyboardInterrupt:
                    raise
                except self.failureException:
                    result.addFailure(self, sys.exc_info())#调用addFailure
                except _ExpectedFailure as e:
                    addExpectedFailure = getattr(result, 'addExpectedFailure', None)
                    if addExpectedFailure is not None:
                        addExpectedFailure(self, e.exc_info)#调用addExpectedFailure
                    else:
                        warnings.warn("TestResult has no addExpectedFailure method, reporting as passes",
                                      RuntimeWarning)
                        result.addSuccess(self)#调用addSuccess
                except _UnexpectedSuccess:
                    addUnexpectedSuccess = getattr(result, 'addUnexpectedSuccess', None)
                    if addUnexpectedSuccess is not None:
                        addUnexpectedSuccess(self)
                    else:
                        warnings.warn("TestResult has no addUnexpectedSuccess method, reporting as failures",
                                      RuntimeWarning)
                        result.addFailure(self, sys.exc_info())
                except SkipTest as e:
                    self._addSkip(result, str(e))
                except:
                    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()

 经过注释部分咱们能够看出，每次执行用例时，都会把执行结果保存到TestResult中。咱们再看看TextTestRunner这个类，在开始就使用了类TextTestResult，而这个类也是继承TestResult，然后在执行的过程当中最终把TextTestResult实例对象传递给TestCase的run方法。因此我上文说了，不过你是用什么方式执行unnitest,到最后都是TestCase的run方法与TestResult的游戏。而咱们的HTMLTestRunner模块也是在继承在TestResult类的基础上的。

说完了TestCase咱们来看看第三个问题吧，也是比较有实际意义的话题，开始我是这样跳过某些test的，代码是这样的：

 

#coding=utf-8
import unittest
a=[False]
class Mydemo(unittest.TestCase):
    def test1(self):
        try:
            print "i am test1"
            #test 1 do some thing
        except Exception,e:
            a[0] = True
            raise e
    @unittest.skipIf(a[0],"test1 fail skip test2")
    def test2(self):
        try:
            print "i am test2"
            raise  AssertionError("error")
            # test2 do some thing
        except Exception,e:
            a[0] = True
            raise e
    @unittest.skipIf(a[0], "test1 fail skip test2")
    def test3(self):
        try:
            print "i am test3"
            # test2 do some thing
        except Exception, e:
            a[0] =True
            raise e
if __name__ == '__main__':
    unittest.main()

 

 想法很简单：就是利用一个全局的数组，若是某个test执行出错我就更改这个数组元素，到下一个case执行的时候就会判断是否要跳过。上面由于test2出错了，本来咱们想跳过test3，可是很遗憾并无跳过test3！结果以下：

C:\Python27\python.exe D:/Moudle/module_1/test4.py
i am test1
.F.
i am test2
======================================================================
i am test3
FAIL: test2 (__main__.Mydemo)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "D:/Moudle/module_1/test4.py", line 20, in test2
    raise e
AssertionError: error

----------------------------------------------------------------------
Ran 3 tests in 0.000s

FAILED (failures=1)

缘由很简单：python在建立Mydemo这个类的时候，因为实例方法都使用了装饰器unittest.skipIf，因此每一个方法都向unittest.skipIf这个装饰传递传递参数a[0]，可是这个a[0]是没用执行过任何case以前的a[0],也就是咱们刚开始定义的a[0]=Flase，因此不可能跳过的。退一万步讲，即便这样可行，也太不美观了吧。咱们想的是当执行当前的test时能判断前面是否有出错的case，有的话就跳过了。可行吗?我以为可行。主要就是用到我上面提到的TestCase中的_resultForDoCleanups的变量，这个其实就是TestResult一个引用。那么咱们能够这样写：

#coding=utf-8
import unittest
class Mydemo(unittest.TestCase):
    def test1(self):
        print "excute test1"
    def test2(self):
        if self._resultForDoCleanups.failures or self._resultForDoCleanups.errors:
            raise unittest.SkipTest("{} do not excute because {} is failed".format(self._testMethodName,self._resultForDoCleanups.failures[0][0]._testMethodName))
        print "excute test2"
        raise AssertionError("test2 fail")

    def test3(self):
        if self._resultForDoCleanups.failures or self._resultForDoCleanups.errors:
            raise unittest.SkipTest("{} do not excute because {} is failed".format(self._testMethodName,self._resultForDoCleanups.failures[0][0]._testMethodName))
        print "excute test3"
if __name__ == '__main__':
    unittest.main()

 运行结果以下：

.Fs
======================================================================
FAIL: test2 (__main__.Mydemo)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "D:/Moudle/module_1/test4.py", line 10, in test2
    raise AssertionError("test2 fail")
AssertionError: test2 fail

----------------------------------------------------------------------
Ran 3 tests in 0.001s

FAILED (failures=1, skipped=1)
excute test1
excute test2

 能够了，咱们看出，test2失败了，test3跳过;固然test2若是正确，test3会执行。目的是达到了，但是每一个case都这样写不太好，咱们想到了装饰器(不会自行百度)，在C:\Python27\Lib\unittest\case.py中新增以下代码：

def Myskip(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

 而后C:\Python27\Lib\unittest\__init__.py中新增：

__all__ = ['TestResult', 'TestCase', 'TestSuite',
           'TextTestRunner', 'TestLoader', 'FunctionTestCase', 'main',
           'defaultTestLoader', 'SkipTest', 'skip', 'skipIf', 'skipUnless',
           'expectedFailure', 'TextTestResult', 'installHandler',
           'registerResult', 'removeResult', 'removeHandler','Myskip']
......
from .case import (TestCase, FunctionTestCase, SkipTest, skip, skipIf,Myskip,
                   skipUnless, expectedFailure)

 最终咱们这样写：

#coding=utf-8
import unittest
class Mydemo(unittest.TestCase):
    def test1(self):
        print "excute test1"
    @unittest.Myskip
    def test2(self):
        print "excute test2"
        raise AssertionError("test2 fail")
    @unittest.Myskip
    def test3(self):
        print "excute test3"
if __name__ == '__main__':
    unittest.main()

 好了，看上去还不错....关于其余的unnitest相关知识，不想再扯了，最后拓展HTMLTestRunner报告，这多是你们关心的！写这个HTMLTestRunner的大神是在很久以前的了,基本能知足你们需求。可是，目前对于web自动化，我以为至少要新增2个东西。第一个新增skip列：由于我可能会skip某些case；第二新增截图列,若是有错误我可能要截图。
打了这么久字不想再多说了....我给出所有代码，而后代码中我改变的地方我给出标记并加注释吧，完整代码以下：(可能有点长，可是要有点耐心)

#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 = '<link rel="stylesheet" href="my_stylesheet.css" type="text/css">'

    # 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
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)
#   >>>

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
    +------------------------+
    |<html>                  |
    |  <head>                |
    |                        |
    |   STYLESHEET           |
    |   +----------------+   |
    |   |                |   |
    |   +----------------+   |
    |                        |
    |  </head>               |
    |                        |
    |  <body>                |
    |                        |
    |   HEADING              |
    |   +----------------+   |
    |   |                |   |
    |   +----------------+   |
    |                        |
    |   REPORT               |
    |   +----------------+   |
    |   |                |   |
    |   +----------------+   |
    |                        |
    |   ENDING               |
    |   +----------------+   |
    |   |                |   |
    |   +----------------+   |
    |                        |
    |  </body>               |
    |</html>                 |
    +------------------------+
    """

    STATUS = {
    0: 'pass',
    1: 'fail',
    2: 'error',
    3:'skip'
    }

    DEFAULT_TITLE = 'Unit Test Report'
    DEFAULT_DESCRIPTION = ''

    # ------------------------------------------------------------------------
    # HTML Template

    HTML_TMPL = r"""<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd">
<html xmlns="http://www.w3.org/1999/xhtml">
<head>
    <title>%(title)s</title>
    <meta name="generator" content="%(generator)s"/>
    <meta http-equiv="Content-Type" content="text/html; charset=UTF-8"/>
    %(stylesheet)s
</head>
<body>
<script language="javascript" type="text/javascript"><!--
output_list = Array();

/* level - 0:Summary; 1:Failed; 2:All */
function showCase(level) {
    trs = document.getElementsByTagName("tr");
    for (var i = 0; i < trs.length; i++) {
        tr = trs[i];
        id = tr.id;
        if (id.substr(0,2) == 'ft') {
            if (level < 1) {
                tr.className = 'hiddenRow';
            }
            else {
                tr.className = '';
            }
        }
        if (id.substr(0,2) == 'pt') {
            if (level > 1) {
                tr.className = '';
            }
            else {
                tr.className = 'hiddenRow';
            }
        }
        if (id.substr(0,2) == 'st') {
 if (level > 1) {
 tr.className = '';
 }
 else {
 tr.className = 'hiddenRow';
  }
        }
    }
}

function showClassDetail(cid, count) {
    var id_list = Array(count);
    var toHide = 1;
    for (var i = 0; i < count; i++) {
        tid0 = 't' + cid.substr(1) + '.' + (i+1);
        tid = 'f' + tid0;
        tr = document.getElementById(tid);
        if (!tr) {
            tid = 'p' + tid0;
            tr = document.getElementById(tid);
        }
         if (!tr) { tid = 's' + tid0; tr = document.getElementById(tid); }
        id_list[i] = tid;
        if (tr.className) {
            toHide = 0;
        }

    }
    for (var i = 0; i < count; i++) {
        tid = id_list[i];
        if (toHide) {
            document.getElementById('div_'+tid).style.display = 'none'
            document.getElementById(tid).className = 'hiddenRow';
        }
        else {
            document.getElementById(tid).className = '';
        }
    }
}


function showTestDetail(div_id){
    var details_div = document.getElementById(div_id)
    var displayState = details_div.style.display
    // alert(displayState)
    if (displayState != 'block' ) {
        displayState = 'block'
        details_div.style.display = 'block'
    }
    else {
        details_div.style.display = 'none'
    }
}


function html_escape(s) {
    s = s.replace(/&/g,'&');
    s = s.replace(/</g,'<');
    s = s.replace(/>/g,'>');
    return s;
}

/* obsoleted by detail in <div>
function showOutput(id, name) {
    var w = window.open("", //url
                    name,
                    "resizable,scrollbars,status,width=800,height=450");
    d = w.document;
    d.write("<pre>");
    d.write(html_escape(output_list[id]));
    d.write("\n");
    d.write("<a href='javascript:window.close()'>close</a>\n");
    d.write("</pre>\n");
    d.close();
}
*/
--></script>

%(heading)s
%(report)s
%(ending)s

</body>
</html>
"""
    # variables: (title, generator, stylesheet, heading, report, ending)


    # ------------------------------------------------------------------------
    # Stylesheet
    #
    # alternatively use a <link> for external style sheet, e.g.
    #   <link rel="stylesheet" href="$url" type="text/css">

    STYLESHEET_TMPL = """
<style type="text/css" media="screen">
body        { font-family: verdana, arial, helvetica, sans-serif; font-size: 80%; }
table       { font-size: 100%; }
pre         { word-wrap:break-word;word-break:break-all;overflow:auto;}

/* -- heading ---------------------------------------------------------------------- */
h1 {
	font-size: 16pt;
	color: gray;
}
.heading {
    margin-top: 0ex;
    margin-bottom: 1ex;
}

.heading .attribute {
    margin-top: 1ex;
    margin-bottom: 0;
}

.heading .description {
    margin-top: 4ex;
    margin-bottom: 6ex;
}

/* -- css div popup ------------------------------------------------------------------------ */
a.popup_link {
}

a.popup_link:hover {
    color: red;
}

.popup_window {
    display: none;
    position: relative;
    left: 0px;
    top: 0px;
    /*border: solid #627173 1px; */
    padding: 10px;
    background-color: 00;
    font-family: "Lucida Console", "Courier New", Courier, monospace;
    text-align: left;
    font-size: 8pt;
    width: 600px;
}

}
/* -- report ------------------------------------------------------------------------ */
#show_detail_line {
    margin-top: 3ex;
    margin-bottom: 1ex;
}
#result_table {
    width: 80%;
    border-collapse: collapse;
    border: 1px solid #777;
}
#header_row {
    font-weight: bold;
    color: white;
    background-color: #777;
}
#result_table td {
    border: 1px solid #777;
    padding: 2px;
}
#total_row  { font-weight: bold; }
.passClass  { background-color: #6c6; }
.failClass  { background-color: #c60; }
.errorClass { background-color: #c00; }
.passCase   { color: #6c6; }
.failCase   { color: #c60; font-weight: bold; }
.errorCase  { color: #c00; font-weight: bold; }
.hiddenRow  { display: none; }
.testcase   { margin-left: 2em; }


/* -- ending ---------------------------------------------------------------------- */
#ending {
}

</style>
"""



    # ------------------------------------------------------------------------
    # Heading
    #

    HEADING_TMPL = """<div class='heading'>
<h1>%(title)s</h1>
%(parameters)s
<p class='description'>%(description)s</p>
</div>

""" # variables: (title, parameters, description)

    HEADING_ATTRIBUTE_TMPL = """<p class='attribute'><strong>%(name)s:</strong> %(value)s</p>
""" # variables: (name, value)



    # ------------------------------------------------------------------------
    # Report
    #

    REPORT_TMPL = """
<p id='show_detail_line'>Show
<a href='javascript:showCase(0)'>Summary</a>
<a href='javascript:showCase(1)'>Failed</a>
<a href='javascript:showCase(2)'>All</a>
</p>
<table id='result_table'>
<colgroup>
<col align='left' />
<col align='right' />
<col align='right' />
<col align='right' />
<col align='right' />
<col align='right' />
</colgroup>
<tr id='header_row'>
    <td>Test Group/Test case</td>
    <td>Count</td>
    <td>Pass</td>
    <td>Fail</td>
    <td>Error</td>
    <td>Skip</td>
    <td>View</td>
    <td>Screenshot</td>
</tr>
%(test_list)s
<tr id='total_row'>
    <td>Total</td>
    <td>%(count)s</td>
    <td>%(Pass)s</td>
    <td>%(fail)s</td>
    <td>%(error)s</td>
    <td>%(skip)s</td>
    <td> </td>
    <td> </td>

</tr>
</table>
""" # variables: (test_list, count, Pass, fail, error)

    REPORT_CLASS_TMPL = r"""
<tr class='%(style)s'>
    <td>%(desc)s</td>
    <td>%(count)s</td>
    <td>%(Pass)s</td>
    <td>%(fail)s</td>
    <td>%(error)s</td>
    <td>%(skip)s</td>
    <td><a href="javascript:showClassDetail('%(cid)s',%(count)s)">Detail</a></td>
	<td> </td>
</tr>
""" # variables: (style, desc, count, Pass, fail,skip, error, cid)


    REPORT_TEST_WITH_OUTPUT_TMPL = r""" <tr id='%(tid)s' class='%(Class)s'> <td class='%(style)s'><div class='testcase'>%(desc)s</div></td> <td colspan='6' align='center'> <!--css div popup start--> <a class="popup_link" onfocus='this.blur();' href="javascript:showTestDetail('div_%(tid)s')" > %(status)s</a> <div id='div_%(tid)s' class="popup_window" > <div style='text-align: right; color:red;cursor:pointer'> <a onfocus='this.blur();' onclick="document.getElementById('div_%(tid)s').style.display = 'none' " > [x]</a> </div> <pre> %(script)s </pre> </div> <!--css div popup end--> </td> <td align='center'> <a %(hidde)s href="%(image)s">picture_shot</a> </td> </tr>
""" # variables: (tid, Class, style, desc, status)
    REPORT_TEST_NO_OUTPUT_TMPL = r""" <tr id='%(tid)s' class='%(Class)s'> <td class='%(style)s'><div class='testcase'>%(desc)s</div></td> <td colspan='6' align='center'>%(status)s</td> <td align='center'> <a %(hidde)s href="%(image)s">picture_shot</a> </td> </tr>
""" # variables: (tid, Class, style, desc, status)


    REPORT_TEST_OUTPUT_TMPL = r"""
%(id)s: %(output)s
"""
    
    # variables: (id, output)



    # ------------------------------------------------------------------------
    # ENDING
    #

    ENDING_TMPL = """<div id='ending'> </div>"""

# -------------------- 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):
        self.failure_count += 1
        TestResult.addFailure(self, test, err)
        _, _exc_str = self.failures[-1]
        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, stream=sys.stdout, verbosity=1, title=None, description=None,name=None):
        self.stream = stream
        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)
        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表明pass个数，nf表明fail,ns表明skip,ne,表明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
            #生成每一个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)
            #生成每一个TestCase类中全部方法的测试结果
            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?
        if isinstance(o,str):
            uo = str(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 = o
        script = self.REPORT_TEST_OUTPUT_TMPL % dict(
            id = tid,
            output = saxutils.escape(str(uo) + str(ue))
        )
       
        if "shot_picture_name" in str(saxutils.escape(str(ue))): hidde_status='' pattern = re.compile(r'AssertionError:.*?shot_picture_name=(.*)',re.S) shot_name =re.search(pattern,str(saxutils.escape(str(e)))) try: image_url="http://192.168.99.105/contractreport/screenshot/"+time.strftime("%Y-%m-%d", time.localtime(time.time()))+"/"+shot_name.group(1)+".png" except Exception,e: image_url = "http://192.168.99.105/contractreport/screenshot/" + time.strftime("%Y-%m-%d",time.localtime(time.time())) else: hidde_status = '''hidden="hidden"''' image_url=''
        row = tmpl % dict(
            tid = tid,
            Class = (n == 0 and 'hiddenRow' or 'none'),
            style=n == 2 and '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)

 把上面代码复制覆盖原来的HTMLTestRunner就好，截图那块我是把错误的图像放在apache服务器的某个路径下的，若是有错误就显示图片超连接，没有就隐藏这超连接。

关于上面的改动其实很简单，熟悉必定的前端语言(html.javascript)便可。HTMLTestRunner原理就是咱们上文提到的利用_TestResult继承unnitest中的TestResult类，并重写了addSuccess，addSkip，addError等方法，把测试结果放在一个self.result里面，最后遍历这个result利用前端的一些知识生成一个html报告。这边贴图贴一下生成的样式吧，执行testcase的代码：

#coding=utf-8
import unittest
import HTMLTestRunner
import sys,os
class Mydemo(unittest.TestCase):
    def test1(self):
        print "excute test1"
    @unittest.Myskip
    def test2(self):
        print "excute test2"
        raise AssertionError("test2 fail")
    @unittest.Myskip
    def test3(self):
        print "excute test3"
    @unittest.Myskip
    def test4(self):
        print "excute test4"
if __name__ == '__main__':
    module_name=os.path.basename(sys.argv[0]).split(".")[0]
    module=__import__(module_name)
    fp=file("./new.html","wb")
    runner=HTMLTestRunner.HTMLTestRunner(fp)
    all_suite=unittest.defaultTestLoader.loadTestsFromModule(module)
    runner.run(all_suite)

最后生成的报告以下：说了这么多只是但愿你们能对unnitest有更多的了解，固然若是你已经懂的更多或者认为我某些地方说错了，请一笑而过....