AttributeError: 'list' object has no attribute 'text'

Question

I am working with Vader from the nltk package. I've imported my dataset following the vader tutorial:

    list = []
    for line in open("C:\Users\Luca\Desktop\Uni\Tesi\PythonTest\paolo.txt","r").readlines():
        for value in line.split(","):
            list.append(value)

Then I've created the function to remove punctuation:

     def _words_only(self):
    text_mod = REGEX_REMOVE_PUNCTUATION.sub('', self.text)
    words_only = text_mod.split()
    words_only = [word for word in words_only if len(word) > 1]
    return words_only

But when I try to use the "words only" function I get this error

    AttributeError                            Traceback (most recent call last)
    <ipython-input-14-cbc12179c890> in <module>()
    ----> 1 _words_only(list)

    <ipython-input-13-68a545bbbaa4> in _words_only(self)
  1 def _words_only(self):
    ----> 2        text_mod = REGEX_REMOVE_PUNCTUATION.sub('', self.text)
      3        words_only = text_mod.split()


    AttributeError: 'list' object has no attribute 'text'

I am really new to Python. Is it a problem in the importing process or is it something else? Thanks for your help.

Answer 1

You don't show where/how you created the function _words_only() , but the self argument indicates that you patterned it on a class method. You're evidently using it as a stand-alone function, like this:

_words_only(list)

I would advise you not to tackle classes yet if you can avoid it. Write your function like this:

def words_only(text):
    text_mod = REGEX_REMOVE_PUNCTUATION.sub('', text)
    words_only = text_mod.split()
    words_only = [word for word in words_only if len(word) > 1]
    return words_only

You should also know that your function is designed to process one string, not a list of them. In addition, don't use builtin names like list as variable names-- you're asking for a very confusing error in a day or two. Use a more informative name, or an abbreviation like lst :

lines = [] 
...

some_words = words_only(lines[0])

Since you actually want to work with the list of lines, apply the revised function to each one like this:

filtered_lines = [ words_only(line) for line in lines ]

If you had wanted to work with the entire contents of the file, you would read in your text like this:

myfile = open(r"C:\Users\Luca\Desktop\Uni\Tesi\PythonTest\paolo.txt","r")
text = myfile.read()
myfile.close()
some_words = words_only(text)