mjhjmtu

I have multiple files containing content like the following:

File 1

NC_12548 og789 |nd784 -2 -54 -6

NC_12548 og789 |nd784 -2 -54 -6

NC_12548 og789 |nd784 -2 -54 -6

File2

NC_54456 og789 |nd784 -5 -56 -6

NC_98123 og859 |nd784 -5 -84 -5

NC_689.1 og456 |nd784 -5 -54 +8

File3

NC_54456 og789 |nd784 -5 -56 -6

NC_98123 og859 |nd784 -5 -84 -5

NC_689.1 og456 |nd784 -5 -54 +8

I want to keep the only first two columns (NC_12345 og855) and discard rest of it. How can I do this?

With awk you can just use | as column separator and print the first column:

awk -F '|' 'print $1' file1.txt file2.txt file3.txt

output will be concatenaded. If it's necessary to keep output in separate files, consider using a for loop in shell around awk

# assuming they're all in the same directory, hence `*`
for fname in ./file*.txt ; do
 # add extension to current file in "$fname" variable indicate new file
 # > does the actual redirection
 awk -F '|' 'print $1' "$fname" > "$fname".new
done

Having new output in .new might be desirable for backup. Otherwise, we can use sed -i to perform in-file replacement. Run it without -i first for test trial

# use file*.txt if they're all in the current directory
sed -i 's/|.*$//' file1.txt file2.txt file3.txt
sed -i 's/(^.*)|.*/1/g' file1.txt file2.txt file3.txt

Another option is via Python:

#!/usr/bin/env python3
import sys

for fname in sys.argv:
 with open(fname) as fd_read, open(fname+'.new','w') as fd_write:
 for line in fd_read:
 fd_write.write(line.split('|')[0] + 'n')

This script is intended to be used as ./script.py file1.txt file2.txt file3.txt and will write output to new files with .new extension

For cutting out some text, I always think of cut first, separating by | as in jno's comment:

cut -d| -f1 file

Or it looks like the format is always 8 characters, then 2 spaces, then 5 more characters, so you could just cut the first 15 characters with

cut -c 1-15 file

Or you could separate fields by spaces, but since there's two spaces between the first & second fields that means you'd cut out fields 1, 2, and 3:

cut -d" " -f 1-3 file

Cut doesn't have in-place file editing like sed, but you can output to a new file, then move over the original file, for example

for file in file1,file2,file3
do
 cut -d" " -f 1-3 "$file" > "$file.2"
 mv "$file.2" "$file"
done