Function 参数最佳实践：dict vs kwargs

Question

I have a Python best practices question that I haven't been able to find a good answer to on the web (possibly due to deficient Googling skills). When defining a function that takes a variable number of key-value pairs, I could either define a parameter that takes a dict , like:

def func(dict_param):
    for key, value in dict_param.items():
        # do stuff

or allow the function to take **kwargs :

def func(**kwargs):
    for key, value in kwargs.items():
        # do stuff

My question is this: Are there any reasons to use a dict over **kwargs and vice versa? Does it depend? What are the caveats of each? What are best practices here?

Answer 1

Following the Zen of Python , which says:

Explicit is better than implicit.

either should not be used as a replacement for actually listing the function parameters in its signature.

Assuming that your use case is to pass around a large number of parameters, you would typically prefer the first form. Instead, if are dealing with function for which you want to support a variable number of keyword arguments, you should use the second form.

Note that, in general the two forms are not the same, in that the keys in the second form must be strings, ie:

def func(dict_param):
    for key, value in dict_param.items():
        pass


def func2(**kwargs):
    for key, value in kwargs.items():
        pass


d = {1: 2, 3: 4}
func(d)
# Runs smoothly
func2(**d)
# TypeError: func2() keywords must be strings

Answer 2

When def func(dict_param) is used, this function will error if the parameter dict_param is not passed as an argument.

However when def func(**kwargs) is used the dictionary paramter is optional and the function can run without being passed an argument (unless there are other arguments)

But as norok2 said,

Explicit is better than implicit.

Therfore, you should always design your functions to try and avoid these ambigious parameters to avoid any potential type errors.

Hope this helps:)