HTMLTestRunner_loguru下html没有显示日志内容问题

• 前置条件

  Windows 10

  loguru

  HTMLTestRunner

• 最近将原有logging日志系统替换成了loguru，loguru的好处不用多说，简单好用。配置起来也比lgging方便多了。封装代码如下：

  import time, os
  from loguru import logger
  from settings import LOG_DIR # 日志保存路径
  
  class Log:
      """输出日志到文件和控制台"""
      def __init__(self):
          # 文件的命名
          log_name = f"test_{time.strftime(‘%Y-%m-%d‘, time.localtime()).replace(‘-‘,‘_‘)}.log"
          log_path = os.path.join(LOG_DIR, log_name)
          # 判断日志文件夹是否存在，不存则创建
          if not os.path.exists(LOG_DIR): 
              os.mkdir(LOG_DIR)
          # 日志输出格式
          formatter = "{time:YYYY-MM-DD HH:mm:ss} | {level}: {message}"
          # 日志写入文件
          logger.add(log_path,   # 写入目录指定文件
                 format=formatter,
                 encoding=‘utf-8‘,   
                 retention=‘10 days‘,  # 设置历史保留时长
                 backtrace=True,  # 回溯
                 diagnose=True,   # 诊断
                 enqueue=True,   # 异步写入
                 # rotation="5kb",  # 切割，设置文件大小，rotation="12:00"，rotation="1 week"
                 # filter="my_module"  # 过滤模块
                 # compression="zip"   # 文件压缩
                ) 
      
      def debug(self, msg):
          logger.debug(msg)
  
      def info(self, msg):
          logger.info(msg)
  
      def warning(self, msg):
          logger.warning(msg)
  
      def error(self, msg):
          logger.error(msg) 
  
  log = Log()
  

• 将上述封装代码引入，生成的html仍没有日志。点击通过无法查看日志。

  

  原来使用HTMLTestRunner生成html测试报告时，报告中只有console输出，logging的输出无法保存，如果要在报告中加入每一个测试用例执行的日志信息，则需要改HTMLTestRunner源码。

• 修改_TestResult类，同时别忘了在文件最上面import logging。

  import logging
  

  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, retry=0,save_last_try=False):
          TestResult.__init__(self)
  
          self.stdout0 = None
          self.stderr0 = None
          self.success_count = 0
          self.failure_count = 0
          self.error_count = 0
          self.skip_count = 0
          self.verbosity = verbosity
          # result is a list of result in 4 tuple
          # (
          #   result code (0: success; 1: fail; 2: error;3:skip),
          #   TestCase object,
          #   Test output (byte string),
          #   stack trace,
          # )
          self.result = []
          self.retry = retry
          self.trys = 0
          self.status = 0
  
          self.save_last_try = save_last_try
          self.outputBuffer = StringIO.StringIO()
          self.logger = logging.getLogger() # 新增这一行
  

• startTest函数中初始化logging.Handler

  def startTest(self, test):
          # test.imgs = []
          test.imgs = getattr(test, "imgs", [])
          # TestResult.startTest(self, test)
          self.outputBuffer.seek(0)
          self.outputBuffer.truncate()
          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
  	    
          # 新增如下代码
          self.log_cap = StringIO.StringIO()
          self.ch = logging.StreamHandler(self.log_cap)
          self.ch.setLevel(logging.DEBUG)
          formatter = logging.Formatter(‘[%(levelname)s][%(asctime)s] [%(filename)s]->[%(funcName)s] line:%(lineno)d ---> %(message)s‘)
          self.ch.setFormatter(formatter)
          self.logger.addHandler(self.ch)
  

• complete_output函数的返回值中加入logging存在内存中的输出，用换行符隔开

  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()
          return self.outputBuffer.getvalue()+‘\n‘+self.log_cap.getvalue() # 新增这一行
  

• 每个用例执行完后，清除handler，修改stopTest函数

  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.
          if self.retry and self.retry>=1:
              if self.status == 1:
                  self.trys += 1
                  if self.trys <= self.retry:
                      if self.save_last_try:
                          t = self.result.pop(-1)
                          if t[0]==1:
                              self.failure_count -=1
                          else:
                              self.error_count -= 1
                      test=copy.copy(test)
                      sys.stderr.write("Retesting... ")
                      sys.stderr.write(str(test))
                      sys.stderr.write(‘..%d \n‘ % self.trys)
                      doc = getattr(test,‘_testMethodDoc‘,u"") or u‘‘
                      if doc.find(‘_retry‘)!=-1:
                          doc = doc[:doc.find(‘_retry‘)]
                      desc ="%s_retry:%d" %(doc, self.trys)
                      if not PY3K:
                          if isinstance(desc, str):
                              desc = desc.decode("utf-8")
                      test._testMethodDoc = desc
                      test(self)
                  else:
                      self.status = 0
                      self.trys = 0
          #self.complete_output()
          a = self.complete_output()
          # 清除log的handle，新增如下代码
          self.logger.removeHandler(self.ch)
          return a
  

• 可能各自用的HTMLTestRunner版本内容不一样，均只需按照上述修改即可。

• 按照上述修改，再次运行，生成的html报告还是没有日志内容，会不会是loguru和logging不兼容?

  

• 上网苦苦找寻，终于找到了解决办法，再次修改loguru封装代码如下：

  import time, os, logging
  from loguru import logger
  from settings import LOG_DIR # 日志保存路径
  
  # 新增如下三行代码
  class PropogateHandler(logging.Handler):
      def emit(self, record):
          logging.getLogger(record.name).handle(record)
          
  class Log:
      """输出日志到文件和控制台"""
      def __init__(self):
          # 文件的命名
          log_name = f"test_{time.strftime(‘%Y-%m-%d‘, time.localtime()).replace(‘-‘,‘_‘)}.log"
          log_path = os.path.join(LOG_DIR, log_name)
          # 判断日志文件夹是否存在，不存则创建
          if not os.path.exists(LOG_DIR): 
              os.mkdir(LOG_DIR)
          # 日志输出格式
          formatter = "{time:YYYY-MM-DD HH:mm:ss} | {level}: {message}"
          # 日志写入文件
          logger.add(log_path,   # 写入目录指定文件
                 format=formatter,
                 encoding=‘utf-8‘,   
                 retention=‘10 days‘,  # 设置历史保留时长
                 backtrace=True,  # 回溯
                 diagnose=True,   # 诊断
                 enqueue=True,   # 异步写入
                 # rotation="5kb",  # 切割，设置文件大小，rotation="12:00"，rotation="1 week"
                 # filter="my_module"  # 过滤模块
                 # compression="zip"   # 文件压缩
                ) 
          # 新增代码
          logger.add(PropogateHandler(), format=formatter)
  
      def debug(self, msg):
          logger.debug(msg)
  
      def info(self, msg):
          logger.info(msg)
  
      def warning(self, msg):
          logger.warning(msg)
  
      def error(self, msg):
          logger.error(msg) 
  
  log = Log()
  

  

原文：https://www.cnblogs.com/tahitimoon/p/15229214.html