Undefined variable error in function

Question

I don't understand why I have the error

NameError: name 'corpus' is not defined

def pretreatment(fileName):
 with open(fileName, 'r') as file:
 global corpus 
 text = file.readlines()
 corpus = []
 for occurrence in text:
 occurrence = occurrence.replace('.',' ')
 occurrence = occurrence.replace(',',' ')
 occurrence = occurrence.replace(';',' ')
 occurrence = occurrence.replace('(',' ')
 occurrence = occurrence.replace(')',' ')
 occurrence = occurrence.replace('?',' ')
 occurrence = occurrence.replace('!',' ')
 occurrence = occurrence.replace(':',' ')
 corpus.append(occurrence)
 return corpus
def lexical_analysis(corpus):
 global corpus
 lexical_corpus = pretreatment(corpus)
 tokens = nltk.word_tokenize(lexical_corpus)
 return tokens
print(pretreatment("blabla.txt"))
print(lexical_analysis(corpus))

I have called the pretreatment function in the lexical_analysis function, but I still have an undefined variable error.

I would like to take advantage of and ask if there is a way to use the replace function in a more elegant way.

EDIT: Thank you for all the explanations. I just managed to declare the variables as global, I understood what the problem was and everything worked perfectly

global variableName

• Martijn Pieters: corpus is a local variable name in pretreatment. Local variables are not shared; the name corpus in the global namespace is independent. You need to assign the returned value from the function to a global name.
• Mangohero1: Read about how variables work and variable scope.

Answer 1

corpus is a local variable inside the function 'pretreatment`. Just because you have called the function doesn't mean it would (or should) begin to exist in the global scope. Try reading about scopes in Python.

print(lexical_analysis(pretreatment('blabla.txt')) # should work

Answer 2

corpus is not defined in global scope, so you are getting error.

you can use like this,

 corpus = pretreatment("blabla.txt")
 print(lexical_analysis(corpus)) 

or

print(lexical_analysis(pretreatment("blabla.txt")))