Question

我想做类似于此的事情，这可以理解为导致SyntaxError：

@api_view(['GET'])
if settings.USE_API_KEY:
    @authentication_classes((TokenAuthentication, ))
    @permission_classes((IsAuthenticated, ))
def all_announcements(request):
    announcements = Announcement.objects.all()
    serializer = AnnouncementSerializer(announcements, many=True)
    return Response(serializer.data)

我如何实现类似的东西？

Answer 1

我不会为此编写装饰器 - 请注意，装饰器语法允许返回装饰器的函数调用（实际上，这是decorators-with-parameters实际上做的）。所以：

def maybe_decorate(condition, decorator):
    return decorator if condition else lambda x: x

@api_view(['GET'])
@maybe_decorate(settings.USE_API_KEY, authentication_classes((TokenAuthentication, )))
@maybe_decorate(settings.USE_API_KEY, permission_classes((IsAuthenticated, )))
def all_announcements(request):
    ...

Answer 2

当您使用这样的装饰器时：

@decorator
def f():
    ...

它等同于：

def f():
    ...
f = decorator(f)

考虑到这一点，你可以像这样完成你想要的东西：

def all_announcements(request):
    announcements = Announcement.objects.all()
    serializer = AnnouncementSerializer(announcements, many=True)
    return Response(serializer.data)

if settings.USE_API_KEY:
    all_announcements = permission_classes((IsAuthenticated,))(all_announcements)
    all_announcements = authentication_classes((TokenAuthentication,))(all_announcements)
all_announcements = api_view(['GET'])(all_announcements)

Answer 3

这样做：

def my_decorator(condition):
    def wrap(fn):
        def new_f(*args,**kwargs):
            # some code that uses the function fn (fn holds a reference to the function you have decorated)
            pass

        if condition:
            return new_f
        else:
            return fn

    return wrap

@my_decorator(settings.USE_API_KEY)
def all_announcements(request):
    ...

（显然是一个一般的例子，你需要申请你的特定代码，其中很多你没有包含在你的问题中，所以我不能提供一个完整的例子供你复制粘贴！）

您可以将此视为my_decorator(condition)函数，返回一个新的装饰器wrap，然后装饰您的函数。装饰器wrap将使用条件来决定是否用new_f替换你的函数。

Answer 4

这样的事情：

def my_conditional_wrapper(f):

    if not condition():
        return f

    def wrapper(*args, **kwargs):
        #wrap function here
        pass

    return wrapper

Answer 5

如何？

def timeit(method):
    def timed(*args, **kw):
        if 'usetimer' not in kw:
            return method(*args, **kw)
        elif ('usetimer' in kw and kw.get('usetimer') is None):
            return method(*args, **kw)
        else:
            import time
            ts = time.time()
            result = method(*args, **kw)
            te = time.time()
            if 'log_time' in kw:
                name = kw.get('log_name', method.__name__.upper())
                kw['log_time'][name] = int((te - ts) * 1000)
            else:
                print '%r took %2.2f ms' % \
                      (method.__name__, (te - ts) * 1000)
            return result
    return timed

def some_func(arg1, **kwargs):
    #do something here

some_func(param1, **{'usetimer': args.usetimer})

如果条件为True，则装饰一个函数？

5 个答案: