How can I pad a file beyond 1024 characters in a unix environment?

Question

I have a file that I need to pad each line with spaces to around 1100 characters from a bash script. Each line in the file currently has 900 characters.

The natural way to to this is

awk -F, '{printf("%-1100s\n",$0)}'  src.txt > dst.txt

However, I get an error telling me

awk: formatted string too long
 record number 1

After some experimentation and searching on the internet, I've determined that the maximum line length of formatted string awk can handle is 1024.

Is there a better way to get around this limitation?

(note: I'm runnin on SunOS 5.10, and I can't add GNU tools to it, etc.)

Answer 1

Get GNU awk.

$ awk 'BEGIN{printf "%-1100s\n", "foo"}'
     foo

You're probably using old, broken awk - use nawk or /usr/xpg4/bin/awk on Solaris. If you're having this problem with one of them, then use the other one.

Answer 2

There was one other solution I came up with:

awk -F, '{printf("%-900s\n",$0)}'  src.txt > tmp1.txt
awk -F, '{printf("%200s\n","")}'  src.txt > tmp2.txt
paste -d "\0" tmp1.txt tmp2.txt > dst.txt

This results in the same file as

nawk -F, '{printf("%-1100s\n",$0)}' src.txt > dst.txt 

Answer 3

If you have perl installed on that system (probably yes), you can set a script like this, and then run it like pad.pl input.txt 1100 > output.txt.

#! /usr/bin/perl

open (INPUT, "<$ARGV[0]");
$LENGTH=$ARGV[1];

while (<INPUT>) {
    chomp($_);
    while (length($_) < $LENGTH ) { $_ = $_." "; }
    print $_."\n";
}

close INPUT;

Answer 4

You could try:

awk '{ pad=1100-length($0); x = $0; 
    while (pad > 0) { x = x " "; pad--}; print x }' src.txt >dst.txt

... which avoids use of printf() completely but does multiple string concatenations instead.