Importing inside functions

Question

Beginner here. I'm trying to learn about global/local variables, circular imports, and how to avoid them. I have three files:

#globaltest import pandas as pd x = 0 dic = {'1995': [3,5,2,1,2,7], '1996': [6,3,1,1,1,8], } df = pd.DataFrame(dic, columns = ['1995','1996'], index=['AT1','AT2','AT3','FR1','FR5','FR7']) df.index.name='Geo'

Second file is:

 #prueba from globaltest import x,df def abcde(): #from globaltest import x,df while x<5: print("Not yet") x=x+1 print("Now ",df)

And the third file is:

#main from prueba import abcde ab=int(input("Option: ")) if ab==1: abcde()

The code works well if in the second file, "prueba", I import the variables from "globaltest" inside of the function abcde(). If I import the variables at the top, however, it doesn't work (UnboundLocalError: local variable 'x' referenced before assignment). I've read that you should not import anything inside functions, so I would like to know what would be the solution to avoid doing it in this case. Thanks in advance!

@ombk,I think it means the function is trying to use a variable that hasn't been assigned. As I said, if I import the variables from the first program inside the function, the program works. If the import takes place outside, I get the error. I'm trying to understand how it works. — JavierSando
– JavierSando, Commented Nov 25, 2020 at 9:32
You get yourself 2 answers, however, I suggest as a first step to check the definition of global scope variables and local scope variables. This way you will understand why it works in prueba if it is inside and not outside. — ombk
– ombk, Commented Nov 25, 2020 at 9:40

Mike Scotty · Accepted Answer · 2020-11-25 09:37:21Z

From the docs (emphasis mine):

If a name binding operation occurs anywhere within a code block, all uses of the name within the block are treated as references to the current block. This can lead to errors when a name is used within a block before it is bound. This rule is subtle. Python lacks declarations and allows name binding operations to occur anywhere within a code block. The local variables of a code block can be determined by scanning the entire text of the block for name binding operations.

The name binding happens here: x=x+1.

To simplify your issue, we can use the follwing code, as importing the variables in this case is the same as simply having them in the global scope:

x = 0; df = None def abcde(): while x<5: print("Not yet") x=x+1 # <- this is where the name binding happens, turning x into a local variable print("Now", df) abcde()

Running this code will yield the following error:

UnboundLocalError: local variable 'x' referenced before assignment

To overcome this issue, you should tell python that you want to access the global variable instead of a local reference by sing the global keyword:

x = 0; df = None def abcde(): global x while x<5: print("Not yet") x=x+1 print("Now", df) abcde()

Output:

Not yet Not yet Not yet Not yet Not yet Now None

However, note hat this will only modify x in the global scope of the current module. It will not affect x in the globaltest module, because it's encapsulated in it's own (module) namespace.

To use the very same x, you should reference it with module.variable:

import globaltest def abcde(): while globaltest.x < 5: print("Not yet") globaltest.x += 1 print("Now", globaltest.df) abcde()

If you access the variable via module.variable you do not have to use the global keyword.

Yash Kanojia · Accepted Answer · 2020-11-25 09:36:50Z

You should do the following changes

 #prueba import globaltest def abcde(): #from globaltest import x,df while globaltest.x<5: print("Not yet") globaltest.x=globaltest.x+1 print("Now ",globaltest.df)

Here we define that the x we are referring is from globaltest module.

Collectives™ on Stack Overflow

Importing inside functions

2 Answers 2

Comments

Comments

Linked

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

Comments

Comments

Linked

Related