带错误代码和错误信息的自定义Python异常

Question

class AppError(Exception):
    pass

class MissingInputError(AppError):
    pass

class ValidationError(AppError):
    pass

...

def validate(self):
    """ Validate Input and save it """

    params = self.__params

    if 'key' in params:
        self.__validateKey(escape(params['key'][0]))
    else:
        raise MissingInputError

    if 'svc' in params:
        self.__validateService(escape(params['svc'][0]))
    else:
        raise MissingInputError

    if 'dt' in params:
        self.__validateDate(escape(params['dt'][0]))
    else:
        raise MissingInputError


def __validateMulti(self, m):
    """ Validate Multiple Days Request"""

    if m not in Input.__validDays:
        raise ValidationError

    self.__dCast = int(m)

validate() 和 __validateMulti() 是一个类的方法，用来验证和存储传入的参数。代码中可以看到，当某些输入参数缺失或验证失败时，我会抛出一些自定义的异常。

我想为我的应用定义一些特定的错误代码和错误信息，比如：

错误 1100: "找不到关键参数。请检查你的输入。"

错误 1101: "找不到日期参数。请检查你的输入。"

...

错误 2100: "多个日期参数无效。接受的值是 2、5 和 7。"

并将这些信息反馈给用户。

1. 我该如何在自定义异常中定义这些错误代码和错误信息？
2. 我该如何抛出或捕获异常，以便知道该显示哪个错误代码或信息？

(附注：这是针对 Python 2.4.3 的内容)。

---

Bastien Léonard 在这个 SO 评论 中提到，你不需要总是定义一个新的 __init__ 或 __str__；默认情况下，参数会被放在 self.args 中，并会通过 __str__ 打印出来。

因此，我更倾向于的解决方案是：

class AppError(Exception): pass

class MissingInputError(AppError):
    
    # define the error codes & messages here
    em = {1101: "Some error here. Please verify.", \
          1102: "Another here. Please verify.", \
          1103: "One more here. Please verify.", \
          1104: "That was idiotic. Please verify."}

用法：

try:
    # do something here that calls
    # raise MissingInputError(1101)

except MissingInputError, e
    print "%d: %s" % (e.args[0], e.em[e.args[0]])

Answer 1

这是我创建的一个自定义异常的例子，它使用了一些预定义的错误代码：

class CustomError(Exception):
"""
Custom Exception
"""

  def __init__(self, error_code, message='', *args, **kwargs):

      # Raise a separate exception in case the error code passed isn't specified in the ErrorCodes enum
      if not isinstance(error_code, ErrorCodes):
          msg = 'Error code passed in the error_code param must be of type {0}'
          raise CustomError(ErrorCodes.ERR_INCORRECT_ERRCODE, msg, ErrorCodes.__class__.__name__)

      # Storing the error code on the exception object
      self.error_code = error_code

      # storing the traceback which provides useful information about where the exception occurred
      self.traceback = sys.exc_info()

      # Prefixing the error code to the exception message
      try:
          msg = '[{0}] {1}'.format(error_code.name, message.format(*args, **kwargs))
      except (IndexError, KeyError):
          msg = '[{0}] {1}'.format(error_code.name, message)

      super().__init__(msg)


# Error codes for all module exceptions
@unique
class ErrorCodes(Enum):
    ERR_INCORRECT_ERRCODE = auto()      # error code passed is not specified in enum ErrorCodes
    ERR_SITUATION_1 = auto()            # description of situation 1
    ERR_SITUATION_2 = auto()            # description of situation 2
    ERR_SITUATION_3 = auto()            # description of situation 3
    ERR_SITUATION_4 = auto()            # description of situation 4
    ERR_SITUATION_5 = auto()            # description of situation 5
    ERR_SITUATION_6 = auto()            # description of situation 6

这里的枚举类型 ErrorCodes 用来定义错误代码。这个异常的创建方式是，把传入的错误代码放在异常信息的前面。

Answer 2

这里有一个简单的例子，展示了一个自定义的 Exception 类，并且它带有特别的代码：

class ErrorWithCode(Exception):
    def __init__(self, code):
        self.code = code
    def __str__(self):
        return repr(self.code)

try:
    raise ErrorWithCode(1000)
except ErrorWithCode as e:
    print("Received error with code:", e.code)

因为你问到了如何使用 args，这里还有一个额外的例子...

class ErrorWithArgs(Exception):
    def __init__(self, *args):
        # *args is used to get a list of the parameters passed in
        self.args = [a for a in args]

try:
    raise ErrorWithArgs(1, "text", "some more text")
except ErrorWithArgs as e:
    print("%d: %s - %s" % (e.args[0], e.args[1], e.args[2]))