Member Avatar for pink_872

I am trying to read from a txt file and counts the number of times each word appears. The problem is that it counts the EOL characters as well. I tried to use the rstrip, still it didn't do anything. So how can I handle these end-of-line characters?
Please help.

Object= open('w.txt','r')
L= Object.read().rstrip()
occurrenences={}

for word in L.split():
    occurrenences[word] = occurrenences.get(word,0)+1
    
for word in occurrenences:
    print(occurrenences[word],word)

Object.close()
  • 3 Contributors
  • 3 Replies
  • 185 Views
  • 9 Hours Discussion Span
  • Latest Post Latest Post by griswolf

Dani AI

Generated

For , a concise diagnostic and a couple of robust fixes.

First, check whether the file actually contains control characters (CR/LF) or literal backslash sequences \r / \n. A quick binary peek shows the difference: if you see b'\r' / b'\n' those are real newlines; if you see b'\\r' / b'\\n' the file contains the two-character sequences and they will be treated as normal text. Example quick check:

with open('w.txt', 'rb') as f:
    print(repr(f.read(200)))

A reliable way to count words while ignoring line breaks and stray control characters is to extract tokens with a regular expression and use a Counter. This treats newlines and carriage returns as separators and avoids counting them as words:

import re
from collections import Counter

with open('w.txt', 'r', encoding='utf-8', newline=None) as f:
    text = f.read()

words = re.findall(r"[A-Za-z0-9']+", text.lower())
counts = Counter(words)
for word, n in counts.most_common():
    print(n, word)

If the file contains literal \r/\n (backslash + letter), replace those sequences before tokenizing:

text = text.replace('\\r', ' ').replace('\\n', ' ')

Other practical notes: open text files in text mode (Python 3 default) so newline translation happens; in Python 2 use 'rU' for universal newlines. For very large files process line-by-line to avoid high memory use. Finally, choose your token regex to match your definition of "word" (the example above keeps digits and apostrophes but excludes underscores; adjust if needed). This addresses both real control characters and the common case where the file contains literal escape sequences.

Recommended Answers

All 3 Replies

Member Avatar for GDICommander

On my machine (Windows), when providing a file that contains "a b c d e", I have this output:

(1, 'a')
(1, 'c')
(1, 'b')
(1, 'e')
(1, 'd')

There is no EOL character in the output, which is expected. So, the code haven't took the EOL character on my machine.

Are you running your script on another OS?

Member Avatar for pink_872

On my machine (Windows), when providing a file that contains "a b c d e", I have this output:

(1, 'a')
(1, 'c')
(1, 'b')
(1, 'e')
(1, 'd')

There is no EOL character in the output, which is expected. So, the code haven't took the EOL character on my machine.

Are you running your script on another OS?

I am running on Windows 7 OS. Lets say that I had the following line in my txt file:
My \r name \r is Nana \n

I want my program to skip \n and \r. I don't want them to be counted.
I tried everything, still it didn't work.

PLEASE HELP :(

Member Avatar for griswolf

Your code should just work because split() should handle all kinds of white space including various newline characters. However, you can try

import string
# ...
for word in L.split(string.whitespace):
    if word:
        occurrenences[word] = occurrenences.get(word,0)+1
#...

The extra test is because when you specify the split characters, you get empty strings if there are adjacent split characters as you have in your example.

Be a part of the DaniWeb community

We're a friendly, industry-focused community of developers, IT pros, digital marketers, and technology enthusiasts meeting, networking, learning, and sharing knowledge.