Writing better functions-CodePudding

I've been using these two functions to write my CSV headers:

def get_csv_headers():
    return ['col 1', 'col 2', 'col 3', 'col 4']
    
def write_csv_headers():
    with open(csv_path, 'w', newline='', encoding="utf-8") as f:
        writer = csv.DictWriter(f, fieldnames=csv_headers)
        writer.writeheader()
    
csv_headers = get_csv_headers()
write_csv_headers()

However, I recently learned about *args and **kwargs (I'm a Python autodidact) and modified my code to:

def write_csv_headers(*args):
    with open(csv_path, 'w', newline='', encoding="utf-8") as f:
        writer = csv.DictWriter(f, fieldnames=[*args])
        writer.writeheader()
    
csv_headers = write_csv_headers('col 1', 'col 2', 'col 3', 'col 4')

I'm trying to learn more about Pythonic syntax and apply Python best practices. Which one is better?

CodePudding user response：

Neither. You shouldn't hard-code the header names, it's good to use a function, like in the first version.

But you shouldn't use a global variable, you should pass the header names as an argument, as in the second version.

There's no need to use *args for this, it can be an ordinary parameter.

def get_csv_headers():
    return ['col 1', 'col 2', 'col 3', 'col 4']

def write_csv_headers(path, headers):
    with open(path, 'w', newline='', encoding="utf-8") as f:
        writer = csv.DictWriter(f, fieldnames=headers)
        writer.writeheader()

csv_headers = get_csv_headers()
write_csv_headers(csv_path, csv_headers)

CodePudding user response：

*args and **kwargs are usually more for passing the args and kwargs along to another function down the road. In this case you're using the args directly. I think the pythonic way would be

headers = get_csv_headers()
write_csv_headers(headers)

CodePudding user response：

If function get_csv_headers returns a list I think it is better to first create a variable to get data from funtion and then pass it to second one as an arguments
code:

#let's assume that this func does some logic (like filtering) to get headers
def get_csv_headers():
    #filter code
    return ['col 1', 'col 2', 'col 3', 'col 4']

def write_csv_headers(pathDir:str,headersList:list):
    with open(pathDir, 'w', newline='', encoding="utf-8") as f:
        writer=csv.DictWriter(f, fieldnames=headersList)
        writer.writeheader()


filtered_headers=get_csv_headers()
write_csv_headers(pathDir='path', headerList=filterd_headers)
#or
write_csv_headers(pathDir='path', headerList=get_csv_headers())