Counting characters in a file in Python

Question

I am trying to write a function that counts the number of characters in a text file and returns the result. I have the following code;

def file_size(filename):
    """Function that counts the number of characters in a file"""
    filename = "data.txt"
    with open(filename, 'r') as file:
        text = file.read()
        len_chars = sum(len(word) for word in text)
        return len_chars

This seemed to be working fine in my IDE when I test ran it with a text file that I created. However when I submit the code to a doctest program I get an error saying it always gives the output of 10. Any help?

Attached is a screenshot of the error message Error screen.

Possible duplicate of [counting characters and lines from a file python 2.7](https://stackoverflow.com/questions/14416522/counting-characters-and-lines-from-a-file-python-2-7) — ziMtyth, Aug 31 '17 at 08:16
You are counting the same thing every time... `filename = "data.txt"` — juanpa.arrivillaga, Aug 31 '17 at 08:17
BTW: `len_chars = sum(len(word) for word in text)` is a over-engineered and I don't think is doing what you think it is doing, going by your names. You see, iterating over a string like you do: `for word in text:` *iterates over the characters*, and doesn't split on word-boundaries just because you name the iterator variable `word`. So you see, there is *no need* to call `len` on each `word` then `sum` it... since `len` is always `1`. But you could have just called `len(text)` and have gotten the same answer! — juanpa.arrivillaga, Aug 31 '17 at 08:21

score 4 · Answer 1 · answered Aug 31 '17 at 08:17

4

You don't use the argument of the function but overwrite filename with the constant "data.txt":

def file_size(filename):
    """Function that counts the number of characters in a file"""
    with open(filename, 'r') as file:
        return len(file.read())

answered Aug 31 '17 at 08:17

Daniel

42,087
4
55
81

2

Not memory efficient – Chris_Rands Aug 31 '17 at 08:18
@Chris_Rands: not the topic here. – Daniel Aug 31 '17 at 08:24
Depending on the particulars, one might prefer raw speed. – juanpa.arrivillaga Aug 31 '17 at 08:52
@juanpa.arrivillaga Agreed, but Python (especially Python 3) is normally based on the philosophy that, unless otherwise stated memory efficiency should be the priority, for example, this is how `itertools` is implemented – Chris_Rands Aug 31 '17 at 08:55
@Chris_Rands: Python's philosophy is simplicity. And there is no equally simple memory efficient solution. – Daniel Aug 31 '17 at 12:39

Sam Chats · Answer 2 · 2017-08-31T08:41:12.490

1

Super efficient solution for ASCII files (runs in theta(1)):

import os
print(os.stat(filename).st_size)

edited Aug 31 '17 at 08:41

answered Aug 31 '17 at 08:26

Sam Chats

2,271
1
12
34

1

Not necessarily what you want, unless you assume the file is in some 8-bit encoding, like ascii. At least, it won't be equivalent to the OPs answer. To be equivalent, the file would need to be in `rb` mode. – juanpa.arrivillaga Aug 31 '17 at 08:32
What juanpa said. This counts bytes, not chars, so it won't always give the correct number, eg if the file is encoded as UTF-8 and it contains non-ASCII chars. – PM 2Ring Aug 31 '17 at 08:38

Eric Duminil · Answer 3 · 2017-08-31T08:25:38.273

0

If you just want the file size of an ASCII file, you should use os.stat :

import os

def file_size(filename):
    st = os.stat(filename)
    return st.st_size

The big advantage with this function is that there's not need to read the file. Python simply asks the filesystem for the file size.

edited Aug 31 '17 at 08:25

answered Aug 31 '17 at 08:21

Eric Duminil

52,989
9
71
124

2

Note, not necessarily the number of characters, unless it is encoded in an 8-bit encoding like ascii. – juanpa.arrivillaga Aug 31 '17 at 08:24

Chris_Rands · Answer 4 · 2017-08-31T11:59:53.877

You could use sum() with a generator expression around iter(partial(f.read, 1), ''), taking inspiration from this answer:

from functools import partial

def num_chars(filename):
    """Function that counts the number of characters in a file"""
    with open(filename) as f:
        return sum(1 for _ in iter(partial(f.read, 1), ''))

The main advantage of this is approach compared to using f.read() is that it is lazy, so you don't read the whole file into memory.

Counting characters in a file in Python

4 Answers4