从文本文件中的每一行中删除第一个字符

Question

im new to Python, to programming in general.

I want to remove first char from each line in a text file and write the changes back to the file. For example i have file with 36 lines, and the first char in each line contains a symbol or a number, and i want it to be removed.

I made a little code here, but it doesn't work as expected, it only duplicates whole liens. Any help would be appreciated in advance!

from sys import argv

run, filename = argv

f = open(filename, 'a+')
f.seek(0)
lines = f.readlines()
for line in lines:
    f.write(line[1:])
f.close()

Answer 1

Your code already does remove the first character. I saved exactly your code as both dupy.py and dupy.txt , then ran python dupy.py dupy.txt , and the result is:

from sys import argv

run, filename = argv

f = open(filename, 'a+')
f.seek(0)
lines = f.readlines()
for line in lines:
    f.write(line[1:])
f.close()
rom sys import argv
un, filename = argv
 = open(filename, 'a+')
.seek(0)
ines = f.readlines()
or line in lines:
   f.write(line[1:])
.close()

It's not copying entire lines; it's copying lines with their first character stripped.

---

But from the initial statement of your problem, it sounds like you want to overwrite the lines, not append new copies. To do that, don't use append mode. Read the file, then write it:

from sys import argv

run, filename = argv

f = open(filename)
lines = f.readlines()
f.close()
f = open(filename, 'w')
for line in lines:
    f.write(line[1:])
f.close()

Or, alternatively, write a new file, then move it on top of the original when you're done:

import os
from sys import argv

run, filename = argv

fin = open(filename)
fout = open(filename + '.tmp', 'w')
lines = f.readlines()
for line in lines:
    fout.write(line[1:])
fout.close()
fin.close()
os.rename(filename + '.tmp', filename)

(Note that this version will not work as-is on Windows, but it's simpler than the actual cross-platform version; if you need Windows, I can explain how to do this.)

---

You can make the code a lot simpler, more robust, and more efficient by using with statements, looping directly over the file instead of calling readlines , and using tempfile :

import tempfile
from sys import argv

run, filename = argv

with open(filename) as fin, tempfile.NamedTemporaryFile(delete=False) as fout:
    for line in fin:
        fout.write(line[1:])
    os.rename(fout.name, filename)

On most platforms, this guarantees an "atomic write"—when your script finishes, or even if someone pulls the plug in the middle of it running, the file will end up either replaced by the new version, or untouched; there's no way it can end up half-way overwritten into unrecoverable garbage.

Again this version won't work on Windows. Without a whole lot of work, there is no way to implement this "write-temp-and-rename" algorithm on Windows. But you can come close with only a bit of extra work:

with open(filename) as fin, tempfile.NamedTemporaryFile(delete=False) as fout:
    for line in fin:
        fout.write(line[1:])
    outname = fout.name
os.remove(filename)
os.rename(outname, filename)

This does prevent you from half-overwriting the file, but it leaves a hole where you may have deleted the original file, and left the new file in a temporary location that you'll have to search for. You can make this a little nicer by putting the file somewhere easier to find (see the NamedTemporaryFile docs to see how). Or renaming the original file to a temporary name, then writing to the original filename, then deleting the original file. Or various other possibilities. But to actually get the same behavior as on other platforms is very difficult.

Answer 2

You can either read all lines in memory then recreate file,

from sys import argv

run, filename = argv

with open(filename, 'r') as f:
    data = [i[1:] for i in f
with open(filename, 'w') as f:
    f.writelines(i+'\n' for i in data) # this is for linux. for win use \r\n

or You can create other file and move data from first file to second line by line. Then You can rename it If You'd like

from sys import argv

run, filename = argv

new_name = filename + '.tmp'
with open(filename, 'r') as f_in, open(new_name, 'w') as f_out:
    for line in f_in:
        f_out.write(line[1:])

os.rename(new_name, filename)

Answer 3

At its most basic, your problem is that you need to seek back to the beginning of the file after you read its complete contents into the array f . Since you are making the file shorter, you also need to use truncate to adjust the official length of the file after you're done. Furthermore, open mode a+ (a is for append ) overrides seek and forces all writes to go to the end of the file. So your code should look something like this:

import sys

def main(argv):
    filename = argv[1]
    with open(filename, 'r+') as f:
        lines = f.readlines()
        f.seek(0)
        for line in lines:
            f.write(line[1:])
        f.truncate()

if __name__ == '__main__': main(sys.argv)

It is better , when doing something like this, to write the changes to a new file and then rename it over the old file when you're done. This causes the update to happen "atomically" - a concurrent reader sees either the old file or the new one, not some mangled combination of the two. That looks like this:

import os
import sys
import tempfile

def main(argv):
    filename = argv[1]
    with open(filename, 'r') as inf:
        with tempfile.NamedTemporaryFile(dir=".", delete=False) as outf:
            tname = outf.name
            for line in inf:
                outf.write(line[1:])
    os.rename(tname, filename)

if __name__ == '__main__': main(sys.argv)

(Note: Atomically replacing a file via rename does not work on Windows; you have to os.remove the old name first. This unfortunately does mean there is a brief window (no pun intended) where a concurrent reader will find that the file does not exist. As far as I know there is no way to avoid this.)

Answer 4

import re

with open(filename,'r+') as f:
    modified = re.sub('^.','',f.read(),flags=re.MULTILINE)
    f.seek(0,0)
    f.write(modified)

In the regex pattern:
^ means 'start of string'
^ with flag re.MULTILINE means 'start of line'

^. means 'the only one character at the start of a line'

The start of a line is the start of the string or any position after a newline (a newline is \\n )
So, we may fear that some newlines in sequences like \\n\\n\\n\\n\\n\\n\\n could match with the regex pattern.
But the dot symbolizes any character EXCEPT a newline, then all the newlines don't match with this regex pattern.

During the reading of the file triggered by f.read() , the file's pointer goes until the end of the file.

f.seek(0,0) moves the file's pointer back to the beginning of the file

f.truncate() puts a new EOF = end of file at the point where the writing has stopped. It's necessary since the modified text is shorter than the original one.
Compare what it does with a code without this line

Answer 5

To be hones, i'm really not sure how good/bad is an idea of nesting with open() , but you can do something like this.

with open(filename_you_reading_lines_FROM, 'r') as f0:
    with open(filename_you_appending_modified_lines_TO, 'a') as f1:
        for line in f0:
            f1.write(line[1:])

Answer 6

While there seemed to be some discussion of best practice and whether it would run on Windows or not, being new to Python, I was able to run the first example that worked and get it to run in my Win environment that has cygwin binaries in my environmental variables Path and remove the first 3 characters (which were line numbers from a sample file):

import os
from sys import argv

run, filename = argv

fin = open(filename)
fout = open(filename + '.tmp', 'w')
lines = fin.readlines()
for line in lines:
    fout.write(line[3:])
fout.close()
fin.close()

I chose not to automatically overwrite since I wanted to be able to eyeball the output.

python c:\bin\remove1st3.py sampleCode.txt