通过函数在python 3中创建全局变量

Question

I was wondering why I can't access the variable: "variable_for_raw_data" after the function ends. The code is like this:

def htmlfrom(Website_URL):
    import urllib.request
    response = urllib.request.urlopen(Website_URL)
    variable_for_raw_data =(input("What will this data be saved as: "))
    global variable_for_raw_data
    variable_for_raw_data = response.read()

Now why can't I access the variable "variable_for_raw_data" after the functions ends?

Things to note:

Python 3.3 urllib NOT urllib2

Answer 1

It looks like you're trying to dynamically create variables, I would imagine that your code looks something like this.

def htmlfrom(website_url):
    import urllib.request
    response = urllib.request.urlopen(website_url)
    variable_for_raw_data =(input("What will this data be saved as: "))
    global variable_for_raw_data
    variable_for_raw_data = response.read()


if __name__ == "__main__":

    htmlfrom("www.stackoverflow.com")

    #html_stackoverflow is never created it is the value
    #of variable_for_raw_data before variable_for_raw_data
    #is overridden by response.read()

    #entering information into input doesn't create a variable
    print(html_stackoverflow)

Here's how I would do it:

import urllib.request

def htmlfrom(website_url): 
    '''
       docstrings
    '''

    response = urllib.request.urlopen(website_url)
    variable_for_raw_data = response.read()
    return variable_for_raw_data

if __name__ == "__main__":

    file_name = input("What will this data be saved as: ")
    html_from_website = htmlfrom("www.stackoverflow.com")

        with open(file_name, 'w') as f: 
        f.write(html_from_website)

Explanation

if you have your import statement inside the function it is only accessable inside the function (ie other functions can't access it)

import urllib.request

PEP 8 has guidelines on how things should be named in python CamelCase is usually reserved for class names

def htmlfrom(website_url): 
    '''
        docstring 
    '''

Docstrings are usually a good idea.

Check out this question for more information on the proper use of globals . Based on what I know about your situation I don't think you need to use them.

    response = urllib.request.urlopen(website_url)
    variable_for_raw_data = response.read()
    return variable_for_raw_data

If you don't know about `if name == ' main ': you should read up on it.

if __name__ == "__main__":

Don't forget to use meaningful variable names and not override the builtins (ie file = "foo.txt" would override the file builtin)

    file_name = input("What will this data be saved as: ")
    html_from_website = htmlfrom("www.stackoverflow.com")

You can learn more about context managers here

    with open(file_name, 'w') as f: 
        f.write(html_from_website)

An edit using globals() , FOR WHICH NO USE CASE EXISTS AT ALL .

def htmlfrom(website_url):
    import urllib.request
    response = urllib.request.urlopen(website_url)
    variable_for_raw_data =(input("What will this data be saved as: "))
    globals()[variable_for_raw_data] = response.read()