Shell Programming and Scripting

Hi Friends,
I want to delete specific columns from a file.
Say my file content is as follows:

"1","a","ww1",1234"
"2","b","wwr3","2222"
"3","c","erre","3333"

Now i want to delete the column 2 and 4 from this file.

That is I want the file content to be:
"1","ww1"
"2","wwr3"
"3","erre"

This file contains lakhs of records.
How is this possible in shell scrpit.

Regards,
Prema

You can awk.

Code:

awk -F, '{ print $1","$3 }' input.txt > output.txt

Hi Vino
Thank you

The code you have mentioned will copy the specified columns of the file to a new file.
Is there any way to delete the specified columns from existing file?

cat file | cut -d',' -f1,3

AFAIK, awk cannot do a read and write on the same file. You have to make use of the temporary file and then rename it to the original file.

You can use either perl or sed if you dont want the redirection of output to a different file.

The sed solution might not very portable. Because, not every sed has a -i option. You would need GNU sed for that.

Code:

sed -i -e 's_\(.*\),.*,\(.*\),.*_\1,\2_g' input.txt

same way you could alternatively use perl
perl -pie ...... rest same as the above sed command