Hello,
Can anyone plz suggest a way to accomplish the following:
I have a semi-column ( separated text file which has coumn headers in the first line and the values in the lines that follow. I need a script that would skip the first line with the column headers and fetch me only the values that follow.

Thanks,
Ravi.

awk 'NR > 1 {print $0}' <filename>

or

tail -n <filename> // where n is one less than the number of lines in the file

Hi,
The second command works...thanks. But just to understand: when I issue the first command it just goes to next line on the console...doesn' do anything. Am I doing something wrong? I issued the command:
awk 'NR > 1 {print $0} Test
where Test is the name of my file.

Thanks,
Ravi.

Sorry...my bad. I didn' give the second quote. Thanks for the help again.

Ravi.

I wrote a small script in perl on the fly which solves the problem you have described.

filename: original file
newfilename: new file with first line of original file truncated


#! /usr/bin/env perl

open (INFILE, "filename")|| die ("File not found. Program exits!");
open (NEWFILE, ">newfilename") || die ("Could not create file _newfilename_. Program exits!");

my @file = <INFILE>;
print NEWFILE @file[1..$#file];
close(NEWFILE);
close(INFILE);


store the above given code in a file and execute it in the same directory as your original file.

cheers

Hi Luqman,
Can you plz reply to my earlier post on corrupt hard-disk problem. I know you suggested to use bootable cd and then re-mount root partition....but how do I mount the root partition after I boot using the bootable cd.

Thanks,
Ravi.

@jkobrien:

your

"awk 'NR > 1 {print $0}' <filename>"

statement works fine with my system but shouldn't it be "gawk 'if (NR>1) print $0' <filename>" ?

I am a bit confused why your command works without if clause. Isn't general structure of awk commands like this?

/pattern/ STATEMENT

Can you please point to certain line of manpage, doc or explain in your own word.

thanks,

Hi,

I don't really know how to answer your question. I've always used awk as

awk -switches 'pattern {command}' filename

"man awk" shows the same thing (gawk is just gnu awk). "if" is always implicit in the pattern and usually you don't need any switches.

I've never seen your syntax before and it gives me parse errors on my system. Does it work on your system?

rpm -q gawk -> gawk-3.1.1-9

John

Yes my command was working at my work place with gawk 3.1.2 but at home an older version 3.1.1. gives syntax error.

I just had a look at the gawk home pages and I realise now what you mean.

The two slashes signify a regular expression ("man awk" or the link above for more on these - they're for matching patterns). If you just specify a regexp, e.g. awk '/ABC/ {print $0}' file, awk will return any lines with the text "ABC" in them. As we're trying to return everything beyond the first record, regexp's aren't apropriate.

Also, the slashes and single quotes create difficulties with the t and c shells. In fact if you search the link above, you'll see that the author says "If you use csh, you're on your own".

John