awk + add column without adding to header

Question

I am working on this awk tutorial here.

$ head file
Name,Number,Letter
Unix,10,A
Linux,30,B
Solaris,40,C
Fedora,20,D
Ubuntu,50,E

this command inserts a new column after the last column. But i do not want to add to the header, how do I do this? I will also, after this, want to add a header name after.

$ awk -F, '{$(NF+1)=++i;}1' OFS=, file
Name,Number,Letter,1
Unix,10,A ,2
Linux,30,B ,3
Solaris,40,C ,4
Fedora,20,D ,5
Ubuntu,50,E,6

Your output shows that your input is sloppy w.r.t trailing blanks. — Jonathan Leffler
– Jonathan Leffler, Commented Oct 1, 2014 at 23:33
tks,how do i fix that? Ah i see you mean the space after the Letter column. — HattrickNZ
– HattrickNZ, Commented Oct 2, 2014 at 0:32
It depends on your choice in editors. I have various implementations of a program stb (strip trailing blanks) which I use to ensure no trailing blanks. In vim, you could use :g/[ ^I][ ^I]*$/s/// (where the ^I is a tab) to eliminate trailing blanks and tabs. At the office, vim is set up to display ghastly yellow when there's a trailing blank or tab on a line in a source file. Etc. Basically, trailing blanks and tabs are untidy; avoid them when you can. — Jonathan Leffler
– Jonathan Leffler, Commented Oct 2, 2014 at 0:55
That i variable is redundant since NR provides a count of lines. Setting FS and OFS to the same value in 2 different places is a terrible idea and that's not a reason to set vars in the file list so I'd treat anything else in that tutorial with a lot of suspicion. Use awk 'BEGIN{FS=OFS=","} .. ' file instead of awk -F, '...' OFS="," file and get the book Effective Awk Programming, Third Edition, by Arnold Robbins. — Ed Morton
– Ed Morton, Commented Oct 2, 2014 at 2:02
tks ed, found that book here if only I had the time to read all that :) — HattrickNZ
– HattrickNZ, Commented Oct 2, 2014 at 21:07

P.P · Accepted Answer · 2014-10-02 07:40:40Z

4

You can skip the first line with:

awk -F, 'NR>1{$(NF+1)=++i;}1' OFS=, file

If there can be blanks in the file, then you can use regular expression to mask them out:

awk -F, '!/^$/ && NR>1{$(NF+1)=++i;}1' OFS=, file

To add a new header, you can do:

awk 'NR==1{$0=$0",New_header";} NR>1{$(NF+1)=++i;}1' OFS=, file

Like @jaypal said, it could be simply:

awk -F, 'NR==1{$0=$0",New_header";} NF && NR>1{$(NF+1)=++i;}1' OFS=, file

edited Oct 2, 2014 at 7:40

answered Oct 1, 2014 at 23:31

P.P

122k20 gold badges193 silver badges251 bronze badges

Sign up to request clarification or add additional context in comments.

5 Comments

jaypal singh Over a year ago

+1: You can use NF to ignore blank lines. awk -F, 'NF && NR>1{$(NF+1)=++i;}1' OFS=, file.

HattrickNZ Over a year ago

How do I overwrite the existing file, or do i just do a > to the same filename?

P.P Over a year ago

@HattrickNZ No, that'll destroy the file! If you have latest awk (4.1 or later) then you can use the -i option to do in-pace modification. Otherwise, use a temp file: awk -F, '{...}' file > tmpfile && mv tmpfile file

HattrickNZ Over a year ago

with 4.1 how do i do it? is it awk -Fi, ... file? when in command line it works, but then when I cat the file it has not changed

P.P Over a year ago

@HattrickNZ If your awk has -i option then : awk -i -F, '{...}' file Read GNU awk manual here.

Ed Morton · Accepted Answer · 2014-10-02 02:03:50Z

2

$ awk 'BEGIN{FS=OFS=","} {print $0, (NR>1?NR-1:"new header")}' file
Name,Number,Letter,new header
Unix,10,A,1
Linux,30,B,2
Solaris,40,C,3
Fedora,20,D,4
Ubuntu,50,E,5

answered Oct 2, 2014 at 2:03

Ed Morton

209k18 gold badges90 silver badges212 bronze badges

Collectives™ on Stack Overflow

awk + add column without adding to header

2 Answers 2

5 Comments

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

5 Comments

Comments

Your Answer

Sign up or log in

Post as a guest

Related